VTT CREATES BUSINESS FROM TECHNOLOGY

Technology and market foresight • Strategic research • Product and service development • IPR and licensing
• Assessments, testing, inspection, certification • Technology and innovation management • Technology partnership
VTT SYMPOSIUM 257

• • • VTT SYMPOSIUM 257

IASMiRT

20th International Conference on

SMIRT 20 – Book of abstracts. Vol. 2

STRUCTURAL MECHANICS IN REACTOR TECHNOLOGY
Dipoli Congress Centre, Espoo (Helsinki), Finland
August 9–14, 2009

The international conferences on Structural Mechanics in Reactor Technology (SMiRT)

have traditionally provided innovative and practical mechanics-based solutions to
the planning, design, construction, operation, and regulation of NPPs and related
facilities. SMiRT 20 will continue this tradition, bringing together experts and prac-
titioners from around the world to share their knowledge of technology that is most
relevant at this time in the nuclear energy industry for both current operations and
future development like Generation IV design.

SMiRT 20 – Book of abstracts

Vol. 2
20th International Conference on
ISBN 978-951-38-6337-1 (soft back ed.) ISBN 978-951-38-6338-80 (URL: http://www.vtt.fi/publications/index.jsp)
STRUCTURAL MECHANICS
ISSN 0357–9387 (soft back ed.) ISSN 1455-0873 (URL: http://www.vtt.fi/publications/index.jsp)
IN REACTOR TECHNOLOGY
 SMiRT 20 Secretariat
VTT Technical Research Centre of Finland
P.O. Box 1000, FI-02044 VTT, Finland
Tel +358 20 722 111
Fax +358 20 722 7053
www.vtt.fi/smirt20, www.iasmirt.org

INTERNATIONAL ASSOCIATION FOR STRUCTURAL MECHANICS in

REACTOR TECHNOLOGY
 Keywords: Nuclear power plants, Nuclear facilities, Nuclear
VTT SYMPOSIUM 257 safety, Structural safety, Advanced reactors, Mechanics of
materials, Aging, Plant life management, Inspection, Maintenance,
Design and qualification, Fracture Mechanics, Structural
evaluation, Structural reliability, Probabilistic safety assessment,
Extreme loads, Earthquakes, Fuel and core structures, Severe
accident management, Computational mechanics, Metal materials,
Concrete materials, Containment structures, Seismic loads,
Seismic analysis, Design methods

20th International Conference

on Structural Mechanics in
Reactor Technology
SMiRT 20
Book of abstracts
Vol. 2
Espoo, Finland 2009

Edited by
Seppo Vuori & Rauno Rintamaa

Organised by
VTT
 ISBN 978-951-38-6337-1 (soft back ed.)
ISSN 0357–9387 (soft back ed.)
ISBN 978-951-38-6338-8 (URL: http://www.vtt.fi/publications/index.jsp)
ISSN 1455–0873 (URL: http://www.vtt.fi/publications/index.jsp)
Copyright © VTT 2009

JULKAISIJA – UTGIVARE – PUBLISHER

VTT, Vuorimiehentie 5, PL 1000, 02044 VTT

puh. vaihde 020 722 111, faksi 020 722 7001

VTT, Bergsmansvägen 5, PB 1000, 02044 VTT

tel. växel 020 722 111, fax 020 722 7001

VTT Technical Research Centre of Finland

Vuorimiehentie 5, P.O. Box 1000, FI-02044 VTT, Finland
phone internat. +358 20 722 111, fax + 358 20 722 7001

Edita Prima Oy, Helsinki 2009

Preface
The two biggest global challenges facing the energy world today are
climate change and the huge increase in energy consumption. Global
electricity consumption is expected to double by the year 2030. There will
be increasing competition in the world for energy resources. On the other
hand there is a need to lower energy intensity of the economy and turn to
CO2 -free forms of energy production. Inevitably, the era of cheap energy
is over. These are driving forces favourable for increasing nuclear energy
production capacity.
In Europe, the EU has put forward a very concrete target. The Heads of
State set very ambitious EU mandatory targets for the new Energy and
Climate Change policy for 2020. One of those targets requires at least
20% less CO2 emissions compared to 1990. Targets will not be realistic
without considerable investments on new nuclear power plants and their
development. Consequently, the European Commission with the support
of nuclear industry have launched a Sustainable Nuclear Energy
Technology Platform (SNE-TP) in September 2007. In Finland, a
decision to build a new (fifth) nuclear power plant (EPR) was already
made in 2002, and the licence for the construction was granted in early
2005. Recently new initiatives have been made by the industry for three
more nuclear power plants in Finland. The applications for the Decision
in Principle for these initiatives are being handled by the Government.
Some other countries in Europe have also made initiatives for new builds.
The international conferences on Structural Mechanics in Reactor
Technology (SMiRT) have traditionally provided innovative and practical
mechanics-based solutions to the planning, design, construction,
operation, and regulation of NPPs and related facilities. SMiRT 20 will
continue this tradition, bringing together experts and practitioners from
around the world to share their knowledge of technology that is most
relevant at this time in the nuclear energy industry for both currently
operating facilities and future development. Around 400 papers will give

3
answers to Challenges Facing Nuclear Renaissance. Besides technical
papers the SMiRT20 Programme consists of Leadership Forum, Technical
Plenary Sessions, Panel Workshops and Tutorial Workshops having topics
of major interest.
This book of abstracts gives an outline of all technical papers presented in
each Technical Divisions. Full papers are published in the DVD Proceedings.
We wish all the SMiRT20 participants and authors a successful conference
and pleasant stay in Finland 2009.

Rauno Rintamaa Seppo Vuori

Chairman Chairman
SMiRT 20 Conference Local Organizing Committee

4
Contents

Preface 3

5. Modeling, Testing and Response Analysis of Structures, Systems

and Components 17
Numerical studies on shear reinforced impact loaded concrete walls (5-1844) 19

Qualification against seismic and other external vibration, experiences from the oversight of
Olkiluoto 3 (5-1851) 20

Seismic motion incoherency effects on SSI response of nuclear islands with significant mass
eccentricities and different embedment levels (5-1853) 23

Finite element analysis of the primary shield structure and evaluation for postulated reactor
pressure vessel head drop (5-1857) 26

Study on radiation shielding performance of reinforced concrete wall (2): shielding analysis
(5-1865) 28

Study on radiation shielding performance of reinforced concrete wall. (1) Loading test on
concrete walls and modeling of concrete cracks (5-1866) 31

Seismic response impact of incoherent SSI analysis by new hard-rock coherency model
(5-1874) 34

Experimental study on modal identification and dynamic amplification of a steel frame

structure (5-1875) 36

Performance of X-plate elasto-plastic dampers. A passive seismic supports for nuclear

piping under cyclic loading (5-1877) 38

Inelastic axisymmetric analysis of BARC prestressed concrete containment model (5-1885) 39

Flow induced vibrations for reactor internals of PWR – Less art, more science (5-1883) 41

Structural modeling and analysis of the SMART-2008 shaking table specimen (5-1892) 43

Structural dynamic analysis of a non symmetrical RC building within the scope of a blind
prediction contest – Project SMART 2008 (5-1900) 45

5
Dynamic analysis of a concrete shear-wall (5-1908) 47

Application and evaluation of “design by rule” procedures applicable to nuclear power plant
ASME B & PVC section iii class 2 and 3 piping (5-1911) 49

Numerical studies on pre-stressed impact loaded concrete walls (5-1921) 51

Seismic analysis and upgrading of suspended ceilings and air ducts over main control room,
emergency control room and control and protection systems of units 3 & 4 of WWER-440 MW
NPP Kozloduy (5-1931) 52

Seismic analyses of safety important piping systems situated in reactor building of units 5
and 6, Kozloduy NPP (5-1932) 54

Strain-based acceptance criteria for section III of the ASME boiler and pressure vessel code
(5-1940) 56

A simple dynamic model for estimating the effect of gaps on response of a spent fuel
transportation cask closure lid during a drop impact (5-1941) 58

Numerical model of the thermal and mechanical behavior of a CANDU 37-element bundle
(5-1942) 60

A rational seismic design approach for reinforced concrete walls for nuclear power plants
(5-1943) 63

Response and failure criteria of large cylindrical vessels to rapid pressurization in CANDU
severe accidents (5-1945) 65

Update of ASCE standard “Seismic analysis of safety-related nuclear structures and

commentary” (5-1947) 67

Investigation of building structures response to heavy item drop (5-1951) 69

LS-DYNA impact analyses of nuclear power plant structures for tornado missile risk analysis
(5-1961) 71

Thermo-mechanical analysis of a helium cooled divertor of a fusion reactor (5-1963) 73

A failure mode evaluation of a 480V MCC in nuclear power plants at the seismic events
(5-1970) 75

Seismic FEM analysis of reinforced concrete structure in SMART-2008 project (5-1974) 76

Salient aspects of analysis and design of large integrated safety related structures (5-1973) 77

Indian PHWR pre-stressed concrete containment performance evaluation with BARCOM

and round robin analysis program (5-1977) 80

The FSI seismic analysis for FBR core assemblies (5-1983) 83

The effects of seismic spectrum on seismic analysis (5-1988) 86

Seismic analysis of primary sodium system components for the loop type fast breeder
reactor (5-2000) 87

6
Qualification of creep, fatigue and fracture design of PFBR components based on tests
(5-2002) 89

Inelastic strain at sliding joint between primary ramp and primary tilting mechanism of
prototype fast breeder reactor (5-2004) 91

Structural integrity assessment of DHX under CDA pressure loading (5-2006) 92

Experimental compression behavior of Stiffened Steel-Plate Concrete (SSC) structures under

compressive loading (5-2008) 94

Developing a numerical model to describe the mechanical behaviour of die-formed expanded

graphite for valve stems sealing (5-2009) 97

Tests on reinforced concrete slabs with pre-stressing and with transverse reinforcement
under impact loading (5-2015) 100

Protection of seismic structures using variable FPS typed TMD system (5-2016) 103

Crack opening in steam generator tubes submitted to an internal pressure: experimental and
numerical modelling (5-2019) 105

Numerical simulation of plain concrete fracture experiments with fictitious crack model
(5-2035) 107

Study of incident water hammer in an engineering loop under two-phase flow experiment
(5-2042) 109

Advanced analysis of gasketed pressure vessel closure systems (5-2043) 112

Update and comparative study on seismic wave incoherence in soil-structure interaction

(5-2048) 114

Effect of geometrical defects and cracks on the collapse of heat exchanger U-bent tubes
submitted to external pressure (5-2049) 116

Study of liquid dispersal from a missile impacting a wall (5-2050) 118

Structural evaluation of drop load effects on buried structures (5-2057) 120

Experimental and numerical simulation of radiolysis gas detonations in BWR exhaust pipes
and mechanical response of the piping to the detonation pressure loads (5-2062) 122

A study on seismic behavior of nuclear power building in strong nonlinear area and fragility
evaluation using 3 dimensional FEM. Part 1. Ultimate seismic condition of building (5-2074) 125

A study on seismic behavior of nuclear power building in strong nonlinear area and fragility
evaluation using 3 dimensional FEM. Part-2. Fragility evaluation (5-2075) 127

The effect of foundation embedment on seismic SSI response of EPR nuclear island
structures (5-2076) 128

Component mode synthesis based SSI analysis of complex structural systems using SASSI
(5-2089) 132

7
 Seismic capacity test of overhead crane under horizontal and vertical excitation – element
model test results on non-linear response behavior (5-2148) 135

Experimental determination of the interaction of blast waves proceeding in air and ground
(5-2465) 137

Spectra-compatible time histories for the ACR NPP in Eastern North America (5-2471) 139

Soil-structure analysis for ACR nuclear island (5-2472) 140

Reactor head stand evaluation using simplified non-linear analysis (5-2474) 141

NPP seismic protection against shock and vibration loads (5-2479) 145

The gearbox for the helium cycle of 10 MW high temperature gas-cooled reactor (5-2508) 146

Reactor building 3D-model for evaluating the pressures on concrete regularization and
foundation waterproofing membrane (5-2514) 147

On the generation of inelastic secondary system seismic response spectra (5-2526) 149

Seismic response of a two-degree-of-freedom system with friction based on the mass ratio
(5-2542) 152

Response of graphite dowel-socket structure under various loads (5-2564) 155

Internal pressure capacity evaluation of prestressed concrete containment buildings

considering multiple aging effects (5-2580) 156

Qualification seismic test on control rod driving mechanism of CEFR (5-2588) 159

Seismic assessment of the sellafield B38 mobile caves (5-2615) 161

Response and seismic margin of Kashiwazaki Kariwa Nuclear Power Plant building to
Chuetsu-oki earthquake (5-3192) 164

Simulation analysis of reactor buildings on Niigataken Chuetsu-oki earthquake at

Kashiwazaki-Kariwa Nuclear Power Plant (5-3193) 167

6. Design and Construction Issues 171

On the design of pipe supports and steelwork regarding revised German nuclear safety
standards (6-1587) 173

Active control of vibrations in piping systems (6-1658) 175

Performance-based design of SSC wall in fire (6-1675) 177

Upgrade and modification of fuel handling equipment in Korea (6-1693) 180

Damping values for seismic design of nuclear power plant SC structures (6-1697) 181

Performance-based fire design of half SC slabs in nuclear power plants (6–1698) 184

Serviceability limit state and crack width analysis of concrete structures in nuclear power
plants (6-1706) 187

8
Demands on anchor systems for concrete structures of nuclear facilities (6-1709) 189

Establishment on slip coefficient of slip resistant connection (6-1712) 191

The SWR 1000 containment – civil design aspects in view of high robustness (6-1714) 192

Guidelines and dataware for life cycle management for NPP pipeline supports (6-1774) 194

A case study on a radiation shielding structure for the cold neutron guide at HANARO –
focused on a mixed proportion design and fabrication of heavy weight concrete for a
radiation shielding (6-1791) 196

An efficient structural form for concrete containment structures (6-1806) 198

Timber mat protection design for buried utilities subject to impact loads (6-1809) 201

Design of modular composite walls subjected to thermal and mechanical loading (6-1820) 203

The effects of design parameters on the thermal response of an LBE capsule (6-1821) 204

An investigation on the fuel assembly structural performance for the PLUS7 fuel design
(6-1824) 205

Friction coefficient measurement test on 13MN class tendon of PC strands for prestressed
concrete containment vessel (PCCV) (6-1825) 206

Analytical study for failure probability of PCCV under pressure load after seismic experience
(6-1826) 208

Civil engineering experiences from the oversight of Olkiluoto 3 (6-1850) 211

Seismic motion incoherency effects for AP1000 nuclear island complex (6-1852) 214

Out-of-plane shear strength of steel plate concrete walls dependent on bond behavior
(6-1855) 217

Development of the simplified fuel assembly model for the fuel assembly SSE and LOCA
analysis (6-1858) 219

Concentration of plastic strain in the steel liner near the equipment hatch in a 1:4 scale
prestressed concrete containment model (6-1903) 223

Structural design of replacement emergency core cooling filtration system (6-1907) 225

Assessing the reliability of seismic base isolators for innovative power plant proposals
(6-1918) 227

Low-activation concrete design methodology for reducing radioactive waste. Categorization

of low-activation concrete by low-activation factor (6-1923) 229

Concrete shrinkage taken into account as crack width assessment (6-1924) 232

Design of suspended ceilings in main control room of units 5 and 6 of Kozloduy NPP
(6-1933) 233

Soil remediation for seismic design of independent spent fuel storage installation (ISFSI)
pad (6-1935) 234

9
 Improving constructability of the new generation nuclear construction through improvements
in design efficiency and use of high-strength reinforcement (6-1937) 235

Implementation of high-performance concrete in the ACR-1000 containment structure for

100 year design life (6-1969) 237

Structural analysis, design and detailing of reactor vault in prototype fast breeder reactor
(6-1972) 240

Development and in-reactor verification of three types of advanced nuclear fuels for PWRs
(6-1986) 243

Investigation of possible corrective actions during manufacturing of fast breeder reactor

components towards assessing the structural integrity (6-2003) 244

Structural analysis towards erection of prototype fast breeder reactor components (6-2005) 246

SSI analysis for a reactor building with high frequency seismic ground motion (6-2040) 247

Evaluation of local stresses at the vessel shell to nozzle intersection (6-2117) 249

A study on optimization of seismic strengthening for the plant facilities in terms of plant
management (6-2225) 250

Comparing European and American codification in the field of NPP civil engineering
(6-2493) 253

Aspects of the design and construction of a new feedwater line for Angra 1 Nuclear
Power Plant as a part of the steam generator replacement program (6-2497) 255

Analysis of the stress-strain state of containment depending on temperature fluctuations

in the environment (6-2560) 256

Using the stressed frame for blast resistant fenestration design of full containment structures
(6-2601) 258

A new device for the study of early-age cracking in massive concrete structures (6-3140) 260

7. Safety, Reliability, Risk and Margins 265

Adjusting the fragility analysis method to the seismic hazard input. Part I: The intensity-
based method (7-1567) 267

Adjusting the fragility analysis method to the seismic hazard input. Part II: The energy
absorption method (7-1568) 269

On the treatment of dependency of seismically induced component failures in seismic

PRA (7-1581) 271

Time-dependent reliability of reinforced concrete beams considering variability in

degradation due to reinforcement corrosion (7-1590) 273

Component degradation effect on seismic risk of NPP (7-1643) 275

The reactor coolant circuit strength and the safety and reliability issues (7-1651) 278

10
Improvement of the seismic fragility analysis by use of the methods of structural reliability
and safety analysis (7-1655) 280

The seismic fragility assessment of the feed water tanks plant using robust prediction
concept of structural response (7-1664) 281

Longevity curves for probabilistic lifetime analysis (7-1684) 282

Reliability and safety analysis of raft foundations under dynamic loading (7-1688) 285

Challenges in the application of probabilistic safety goals for nuclear power plants (7-1769) 287

Integrated soil-structure fragility analysis method for nuclear structures (7-1771) 288

Risk-informed implementation of results from modern seismic hazard analyses into the
design of new buildings of the existing NPP’s (7-1773) 290

PSA Level 2 – Experience with the review process from the perspective of the independent
evaluator (7-1778) 292

Probabilistic fracture mechanics: PTS Screening Criteria for RTNDT, application of FAVOR
code to a German KONVOI plant (7-1785) 294

Effects of AAR on seismic assessment of nuclear power plants for life extensions (7-1789) 296

Site-specific ground motion models for soil sites with thick sedimentary layers (7-1795) 297

Seismic performance assessment for safety-related nuclear structures (7-1818) 299

Development of a reliability data handbook for piping components in Nordic nuclear power
plants (7-1837) 301

Benchmark exercise on risk-informed in-service inspection methodologies (7-1841) 302

CANDU pressure tube degradation and probabilistic safety criteria (7-1847) 304

Results and insights from interim seismic margin assessment of the Advanced CANDU
®
Reactor (ACR) 1000 reactor (7-1849) 305

Application of CFD code PHOENICS for simulating CYCLONE SEPARATORS (7-1867) 306

Seismic fragility of a civil engineering structure (7-1871) 307

Thermal-hydraulic analysis for accidents in OPR1000 and evaluation of uncertainty for PSA
(7-1878) 309

Evaluation of the seismic damage index of structures using fuzzy logic (7-1890) 311

A procedure for the computation of seismic fragility of equipment components in NPPs

(7-1904) 313

Detailed plant seismic walkdown of the Armenian NPP – Unit 2 (7-1949) 315

A quantitative method for RI-ISI assessment (7-1975) 316

Evaluation for run-out distance distribution of rocks falling from slopes (7-1979) 318

Reliability analysis of slope stability at nuclear power plant site (7-1982) 320

11
 Seismic damage assessment by probabilistic seismic demand models applied to NPP
structures (7-1993) 321

Safety margins in mechanical integrity assessments for passive NPP components (7-2014) 324

Load factor in case when separating aleatory uncertainty and epistemic uncertainty (7-2023) 326

Research associated with the July 2007 NCO earthquake at the Kashiwazaki-Kariwa
nuclear power plant (7-2064) 328

Seismic risk analysis utilizing the PGA and PGV simultaneously as ground motion
measures (7-2389) 330

Seismic isolation of the IRIS NSSS building (7-2399) 332

Insights gained from the Beznau Seismic PSA (7-2405) 333

Thermal-hydraulic analysis for accidents in OPR1000 and evaluation of uncertainty for PSA
(7-2478) 334

Estimation of leak and break frequencies for probabilistic safety analyses of piping systems
(7-2529) 336

A temperature characteristic diagnosis algorithm of the abnormal signal simulation analysis

module by using probabilistic techniques (7-2546) 337

Fragility functions for seismic performance assessment of safety-related reinforced concrete

nuclear structures (7-2557) 338

Experience from a seismic probabilistic safety assessment of a German PWR (7-2566) 339

8. Issues Related to Operations, Inspection and Maintenance 343

Fretting wear resistance nuclear fuel design & operating experience (8-1619) 345

Aging problems and residual life time evaluation of the WWER-1000 MW containment shell
structure (8-1622) 347

Monitoring relative humidity and temperature for life-time assessment of sandwich-type

concrete structures (8-1647) 349

German nuclear power plants utility ageing management – long term fatigue evaluation of
safety relevant components (8-1652) 352

Considerations related to long-term operation for CANDU 6 NPP (8-1663) 354

Service life management system of concrete structures in nuclear power plants (8-1685) 357

Statistical assessment method for the optimization of the inspection need for nuclear steam
generators based on existing inspection data (8-1731) 359

Development of RI-ISI at STUK (8-1794) 361

Monitoring pipe thinning using two accelerometers (8-1801) 363

12
Activities of OECD/NEA in the fields of integrity and ageing of components and structures
(8-1804) 364

Effects of concrete creep and shrinkage on the stress conditions of a post-tensioned

containment structure for steam generator replacement project (8-1812) 366

Microbially influenced corrosion in cooling water systems – development of a new protection

concept for system components conveying brackish water (8-1815) 369

PAMS – piping and component analysis and monitoring system application and visualisation
(8-1835) 371

Update on Canadian regulatory oversight of ageing management for nuclear power plants
(8-1842) 374

Performance surveillance of Gentilly-1 reactor building GFRP repair using fiber optic sensors
and strain gauges (8-1848) 375

Recent advances in seismic non-destructive testing, and associated finite element based
evaluation, utilized on a pre-stressed concrete reactor containment at a NPP in operation
(8-1882) 377

Study on the boric acid corrosion behavior of disk/seat materials in SI check valves (8-1889) 380

The complex approach to the determination of NPP Steam Generators Heat-Exchange

Tubes Plugging Criterion (NPP SG HET PC) on the basis of analyses of the processes of
the tubes damaging during NPP operation (8-1910) 381

Seismic qualification and upgrade of safety important pipelines support systems in reactor
building of units 5 and 6, Kozloduy NPP (8-1934) 383

Proactive Management of Materials Degradation (PMMD) and enhanced structural reliability

(8-1954) 385

Ageing management of steam generator internals (8-1959) 387

Assessment of gaseous pollution from hot cutting processes in NPP disassembling (8-1960) 389

Repair and strengthening of damaged reinforced concrete slabs with CFRP (8-2047) 390

Development of integrity evaluation program for pipe wall thinning (8-2071) 392
1
Fabrication flaw density and distribution in piping weldments (8-2476) 393
1
Fabrication flaw density and distribution in weld repairs (8-2477) 394

Graphite blocks reloading consideration in HTR-PM (8-2486) 395

Korean experience in aging management for long term operation of NPP (8-2533) 399

Numerical and analytical framework for analysis of crack initiation and propagation under
thermal fatigue loading (8-2581) 401

A model to monitoring real-time fracture of concrete subjected to the load from tendons by
AE technique (8-2603) 402

Methodology research on prediction for operating lifetime of PWR RPV (8-2619) 403

13
 The effectiveness of chemical cleaning in reducing the risk of leakage in steam generator
tubing: a Bayesian approach (8-2621) 404

A proposal for a unfied model on nuclear power plant life management including
maintenance optimisation (8-3173) 405

9. Waste Management, Fuel Cycle Facilities and Decommissioning 407

Development of neutron shielding materials for nuclear fuel storage facilities (9-1707) 409

Modelling the aging of concrete as a technical barrier in nuclear waste disposal facilities
(9-1834) 410

Numerical and experimental structural assessment of a half scale model of a nuclear spent
fuel elements transportation package under 9 m drop tests (9-1927) 412

Technical challenges related to the spent nuclear fuel dry cask storage/transportation
analysis and design (9-1936) 413

Optical strain measurement of plastic strain localization in nuclear waste copper canisters
(9-2012) 414

Planning of one-piece removal of BWR reactor pressure vessels at Barsebäck Nuclear

Power Plant, Unit 1 & 2 (9-2571) 415

Stability analysis of storage of spent fuel in stack of trays in pool (9-2604) 418

10. Challenges of New Reactors 419

The challenge of nuclear reactor structural materials for Generation IV Nuclear Energy
Systems (10-1586) 421

Design considerations for developing a steam generator for integral modular reactor
SMART (10-1589) 422

Dynamic analysis methodology for stacked graphite fuel blocks of a VHTR using a
commercial structural analysis code (10-1678) 423

RCC-MR 07 code: specificities and recent developments (10-1686) 425

FE analysis of ITER 40º vacuum vessel sector and stress assessment according to French
nuclear code RCC-MR (10-1705) 426

A high temperature gas loop to simulate VHTR and nuclear hydrogen production system
(10-1870) 428

Impact of engineered safety features on AHWR containment (10-1881) 431

Preliminary analysis of the structural effects due to dynamic loads of the isolated next
generation lead cooled reactor (10-1887) 433

Radiotoxicity perspectives for different ELSY working hypotheses: towards a sustainable

fuel cycle (10-1905) 434

Material challenges of the new advanced gas cooled systems (10-2038) 436

14
The stress assessment of reflector graphite bricks in high-temperature gas-cooled reactor
(10-2044) 438

Study on steam generator helical tube integrity assessment of HTR (10-2051) 439

Researches for development of regional energy reactor, REX-10 (10-2070) 441

Summary of SMIRT20 preconference topical workshop – Identifying structural issues in

advanced reactors (10-2504) 442

Development of high efficiency and high capacity gas/gas heat exchanger for gas-cooled
reactors (10-2535) 443

Computational assessment of the reactor vessel cooling options in a prismatic core VHTR
(10-3139) 445

Generation IV material issues – case SCWR (10-3143) 447

The Finnish Sustainable Energy (SusEn) project on New Type Nuclear Reactors (NETNUC)
(10-3163) 449

Thermal hydraulic transient analysis of the high performance light water reactor using
APROS and SMABRE (10-3164) 451

15
16
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

5. Modeling, Testing and Response

Analysis of Structures, Systems and
Components

Modeling and response analysis of structures (including foundations),

systems, lifelines and components subjected to extreme loads. Validation
of analytical methods based on experimental results.

(Part 2. 1844–3193)
Part 1 of the abstracts of Division 5 is included in Volume 1 of the SMiRT20
abstracts

17
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

18
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Numerical studies on shear reinforced

impact loaded concrete walls (5-1844)

Arja Saarenheimo1, Kim Calonius2, Markku Tuomala3, Ilkka Hakola4

1
VTT Technical Research Centre of Finland, e-mail: [email protected]
2
VTT Technical Research Centre of Finland, e-mail: [email protected]
3
Technical University of Tampere (TUT), e-mail: [email protected]
4
VTT Technical Research Centre of Finland, e-mail: [email protected]

Protective concrete barrier walls of nuclear power plants are required to

withstand the effects of impacts by projectiles. Numerical methods are
developed and taken in use for predicting the response of reinforced concrete
structures subjected to impacts of deformable projectiles. Structural behaviour,
in terms of collapse mechanism type and the damage grade, are predicted both
by simple analytical methods and by extensive non-linear FE-models.
Experimental data is needed in order to verify the accuracy of numerical models.
Numerical results obtained using different kinds of methods are compared with
experimental data and observations on impact loaded reinforced concrete walls
with shear reinforcement.
An experimental set-up has been constructed at Technical Research Centre of
Finland VTT for medium scale impact tests. The main objective of this effort is
to provide data for the calibration and verification of numerical models.
Reinforced concrete walls have been tested in the IMPACT project using
deformable aluminium missiles. The IMPACT test facility is described in
Lastunen et al. [1].

Reference
1. A. Lastunen et al. Impact Test Facility in Transaction SMiRT-19 August 2007, Toronto,
Canada.

19
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Qualification against seismic and other

external vibration, experiences from the
oversight of Olkiluoto 3 (5-1851)

Pekka Välikangas, Jukka Myllymäki, Jorma Sandberg

Radiation and Nuclear Safety Authority (STUK), Finland
e-mail: [email protected]

Introduction
This paper presents experiences which the Finnish Radiation and Nuclear Safety
Authority (STUK) has gained during the inspection and supervision of
qualification of structures and equipment against seismic and other external
vibration for Olkiluoto 3. Feedback from the Olkiluoto 3 is presented in order to
bring some useful information to be considered in future projects for nuclear
power plants. Experiences presented in this paper are from a nuclear safety
authority’s point of view.

Background
Qualification of nuclear safety related structures and equipment against external
vibrations is a combination of studies in different technical domains. The first
issue is to define the external source from which the vibrations are induced.
Such sources are seismic earthquake, aircraft crash and explosion pressure wave.
The vibration behaviour of a building framework must be analysed and,
depending on selected equipment, corresponding design response spectra’s must
be defined for further qualification of nuclear safety related equipment.
The common procedure is to qualify different components separately against
induced external vibrations. Beside that the equipment are studied as larger
units, formed by, e.g., pump, motor, switchgear, foundation and connection to
piping. Qualification against vibrations can be made by analysis, shake table
tests and/or the combination of these analyses and vibration tests.
STUK is overseeing the construction and component manufacturing of nuclear
power plants by inspections and supervision of design, manufacture and
construction at the site. Also the installation and commissioning of nuclear
safety related equipment will be followed as soon as these phases begin. If
quality control and assurance from these later phases bring information in the
form of non-conformance reports and audit findings, they will also be described
in this paper.

20
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Nuclear safety related vibration resistance questions are dealt with

correspondence between STUK and the licensee of the Olkiluoto 3 in order to
ensure, that all solutions fulfil nuclear safety requirements and from licenser
point of view that these solutions are accepted by STUK. Important part for
efficient handling of different, mostly complicated technical questions has been
the close cooperation between the licensee and supplier of the nuclear power
plant so that also specialists from STUK have participated in technical
discussions.

Essential results
Feedback from STUK’s studies and decisions and from non-conformance
reports of Olkiluoto 3 qualification against external vibrations has been collected
so that it can serve in future projects as well as in the further development of
regulatory guides on nuclear safety [1…3].
Experiences are presented from design process point of view so that is
possible to understand how design criteria have been formed from the initial
design assumptions to final equipment qualification against external vibrations.
Such design phases are sources of external vibration, behaviour of building
framework and equipment qualification.
An important issue is cooperation between different technical domain areas,
like civil engineering and equipment design.

Summary
Experiences from Olkiluoto 3 nuclear safety related equipment qualification
against external vibrations will be described. In order to see the total picture,
brief description of how STUK has learnt its lessons will be presented. Feedback
is collected by design process so that is possible to understand the development
of criteria for nuclear safety related equipment qualification against external
vibrations. The Finnish seismic conditions and regulatory requirements will also
be described.

Selective references from standards, regulatory

guides and research studies, which are directly
related to the reported feedback/experiences:
1. Saarenheimo, A., Tuomala, M., Calonius, K., Lastunen, A., Hyvärinen, J. Myllymäki
Jukka, Numerical Studies on Impact Loaded Reinforced Concrete Walls, 2007.
International Association for Structural Mechanics in Reactor Technology. 8 p.
Transactions of the 19th International Conference on Structural Mechanics in
Reactor Technology, SMiRT 2007, Toronto, Canada, August 2007.

21
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

2. Saarenheimo, A., Tuomala, M., Hakola, I., Hyvärinen, J., Aalto, A., Myllymäki, J.
Impact of deformable missiles of reinforced concrete walls, Proceedings of
CONSEC'07. 5th International Conference on Concrete unser Severe Conditions:
Environment & Loading. Toutlemonde, F. et al (eds.), Tours, France, June 2007,
8 p.

3. Calonius, K., Hakola, I., Hostikka, S., Kankkunen, A., Lastunen, A., Saarenheimo, A., Silde,
A., Tuomala, M. Experimental and Numerical Studies on Impacts, Räty, H., Puska,
E.K. (eds.). 2006. SAFIR. The Finnish Research Programme on Nuclear Power
Plant Safety 2003–2006. Final Report. VTT Research Notes 2363. VTT. Espoo
(2006). Pp. 207–224. http://www.vtt.fi/inf/pdf/tiedotteet/2006/T2363.pdf.

22
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Seismic motion incoherency effects on SSI

response of nuclear islands with significant
mass eccentricities and different
embedment levels (5-1853)

Ghiocel Dan Mircea1, Short Stephen2, Greg Hardy2

1
GP Technologies, Inc.
6 South Main St., 2nd Floor, Pittsford, New York 14534, USA
e-mail: [email protected]
2
Simpson, Gumpertz and Heger
th
4000 MacArthur Blvd., 7 Floor, Suite 710, Newport Beach, California, 92660, USA
e-mails: [email protected], [email protected]

The paper shows the effects of seismic motion incoherency on the soil-structure
interaction (SSI) response of a nuclear structures founded on different site
conditions. The paper presents results obtained from a sequence of parametric
SSI studies using the AP1000-based stick SSI model that was also employed in
recent EPRI studies (Short, Hardy, Merz and Johnson, 2006 and 2007). The
paper focuses on the effects of foundation embedment on incoherent SSI
response. Different stochastic and deterministic incoherent SSI approaches are
employed. These incoherent SSI approaches are a part of those used in the EPRI
studies called the SASSI-based approaches (deterministic AS and SRSS
approaches, and stochastic simulation approach). In addition, an alternate
version of the SRSS approach is included. The 2005 and 2007 Abrahamson
incoherency models for all sites, hard-rock sites and soil sites are applied. No
wave passage effects are considered. The computed SSI results show that
incoherent SSI effects are significant for both non-embedded and embedded
structures. Conclusions and recommendations are stated at the end of the paper.
We considered both stochastic and deterministic incoherent SSI approaches.
In addition to the stochastic simulation apprach, three deterministic approaches
were considered: i) linear superposition, or algebraic sum, of the scaled incoherent
spatial modes (AS in EPRI studies), ii) quadratic superposition of the incoherent
modal SSI complex response amplitudes (transfer function amplitudes) assuming a
zero-phase for the incoherent SSI complex response phase (SRSS in EPRI
studies), and iii) quadratic superpostion of the incoherent modal SSI complex
response amplitudes (transfer function amplitudes) assuming a non-zero phase
for the incoherent SSI complex response that is equal to coherent SSI complex
response phase (not used in EPRI studies). The last implementation is an
alternate version of SRSS approach that does not neglect the complex response
phase.

23
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Two case studies are considered: i) A typical PWR Reactor Building (RB)
with three different embedment levels, and ii) the AP1000-based stick model
used in the EPRI studies with different embedment levels, and two different
foundation mat sizes. The AP1000-based stick model that was surface founded
in the EPRI studies was embedded in the analyses of this paper. The embedded
foundation walls are modeled by shell elements. Seismic input and soil layering
were assumed: i) identical with those used in EPRI studies for AP1000-based
stick model, ii) typical hard-rock site-specific GRS with 2007 Abrahamson hard-
rock coherency model and hard-rock soil profile with Vs of 9000 fps, iii) RG
1.60 GRS with 2007 Abrahamson soil coherency model and soil layering with
Vs of 1000 fps. It should be noted that the 2007 Abrahamson soil coherence
function is currently not accepted by US NRC. Only the 2007 Abrahamson hard-
rock coherence function is permitted by US NRC at this time.
Based on the investigated case studies shown in this paper, the following
conclusions are drawn:
1) The effects of motion incoherency are similar for non-embedded and
embedded nuclear structures. The SSI results shown herein indicate that
motion incoherency effects are significant for both rock and soil sites.
Typically, motion incoherency effects are larger for rock sites in high
frequency range.
2) Combined effects of embedment and motion incoherency are much
more complex for soil sites than for rock sites. For rock sites, it appears
that motion incoherency effects are to reduce the SSI response at all
frequencies, but more drastically in the high frequency range, above 10-
12 Hz. For soil sites, the motion incoherency effects manifest visible at
much lower frequencies, below 10Hz, where global, dominant structural
vibration modes exist.
3) For structures with significant mass eccentricities, motion incoherency
effects could amplify the torsional SSI responses, as shown herein for
the AP1000-based stick model on a soil site, in Y direction.
As practical recommendations, we believe that for soil sites, the combined
effects of motion incoherency and embedment have to be considered. For hard-
rock sites, since motion incoherency effects have similar trends for non-
embedded and embedded SSI models, the use of simple reduction factors might
be acceptable.
We believe that more study is worthwhile to propose and gain acceptance for
the use of soil coherence function by the US NRC.

24
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

References
1. Ghiocel, D.M., Li, D., Coogler, K., Tunon-Sanjur, L. 2009. Seismic Motion Incoherency
th
Effects on AP1000 Nuclear Island Complex, The 20 SMiRT Conference, Paper
1852, Helsinki, Finland, August 14–18.

2. Short, S.A., Hardy, G.S., Merz, K.L., Johnson, J.J. 2006. Program on Technology
Innovation: Effect of Seismic Wave Incoherence on Foundation and Building
Response, Electric Power Research Institute, Palo Alto, CA and US Department
of Energy, Germantown, MD, Report No. TR-1013504, December.

3. Short, S.A., Hardy, G.S., Merz, K.L., Johnson, J.J. 2007. Validation of CLASSI and
SASSI to Treat Seismic Wave Incoherence in SSI Analysis of Nuclear Power
Plant Structures, Electric Power Research Institute, Palo Alto, CA and US
Department of Energy, Germantown, MD, Report No. TR-1015111, Novermber.

4. Short, S.A., G.S. Hardy, K.L. Merz, J.J. Johnson. 2006. Program on Technology Innovation:
Effect of Seismic Wave Incoherence on Foundation and Building Response.
Electric Power Research Institute, Palo Alto, CA and US Department of Energy,
Germantown, MD, Report No. TR-1013504, December.

5. Short, S.A., G.S. Hardy, K.L. Merz, J.J. Johnson. 2007. Validation of CLASSI and
SASSI to Treat Seismic Wave Incoherence in SSI Analysis of Nuclear Power
Plant Structures. Electric Power Research Institute, Palo Alto, CA and US
Department of Energy, Germantown, MD, Report No. TR-1015111, November.

25
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Finite element analysis of the primary

shield structure and evaluation for
postulated reactor pressure vessel
head drop (5-1857)

Necip Onder Akinci, Jaspal Singh Saini, William H. Johnson

Bechtel Power Corporation

Reactor pressure vessel head (RPVH) lifts are required for head replacement operations
and refueling outages at nuclear power plants. To ensure that the reactor core
remains covered with coolant and sufficient cooling is available following a
postulated RPVH drop, it is required to evaluate the consequences of impact
loading from a postulated concentric flat head drop (approximately 400 kips
free-falling through 40 ft. or more) onto the vessel flange (Ref. 1). Available
literature for this non-design basis event is limited and structural response
acceptance criteria for it are currently under development jointly by the US
Nuclear Regulatory Commission (NRC) and the industry group Nuclear Energy
Institute (NEI) (Ref. 2). Determination of system responses for implementation
of these criteria entails transient linear or nonlinear dynamic analysis of the
nuclear steam supply system, internal concrete structure and containment system
by Finite Element (FE) software such as ANSYS (Ref. 3). As it is convenient
and necessary, due to contractual and division of responsibility requirements, to
separate the design/analysis considerations for the nuclear steam supply system
(NSSS) from those for the supporting internal concrete structure (ICS), a
decoupled approximate approach must be taken. Also, there are unique and
special response and design evaluation considerations for this impact loading
condition which must be addressed. This investigation provides general
guidance and recommendations for addressing the NSSS interface with
boundary and supporting structures, i.e., soil, containment building/basemat, and
ICS in the performance of FE analysis and design evaluations for this beyond
design basis RPVH postulated drop event.
The important factors that need to be considered in development of an
analysis methodology are discussed by addressing the various types of RPVH
support interfaces. The structural response to the postulated impact loading is
characterized by inelastic deformation of the NSSS system components and its
structural steel supports, accompanied by linear or nonlinear response of the
concrete in compression and a rebound response in tension. Tension response
due to the rebound may yield or rupture the concrete anchorages (rupture of the
anchorages per se does not constitute a “failure” condition provided that the
deformation is limited and the system pressure boundary integrity is

26
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

maintained). The actual supporting concrete behavior varies in accord with the
supporting concrete configuration which can be a shelf, ledge, a corbel, or the
base slab itself. Other factors considered are the estimation of inelastic response
utilizing a linearly elastic model, the effects of decoupling, soil-structure
interaction effects, model boundaries and extent of model needed to capture
response, the various concrete inelastic response modes, problem size and
solution time/computational resources, anchorage stiffness and strength
uncertainty, dynamic load factors and equivalent static load approaches, shear
friction utilization as an energy absorber, and concrete local bearing response.

References
1. NUREG-0612. Control of Heavy Loads at Nuclear Power Plants. Office of Nuclear Reactor
Regulation, U.S. Nuclear Regulatory Commission, Washington D.C. 1980.

2. NEI 08-05, Rev. 0. Industry Initiative on Control of Heavy Loads. Nuclear Energy Institute,
Washington D.C. 2008.

3. ANSYS Version 11. User Manual. ANSYS, Inc. Software Products. Pennsylvania. 2007.

27
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Study on radiation shielding performance of

reinforced concrete wall (2):
shielding analysis (5-1865)

Takashi Maki1, Yoshinari Munakata2, Yoshiyuki Sato3, Keiji Sekine4,

Yoshinori Sakai5, Koji Oishi6, Kazuyuki Torii7
1
Japan Nuclear Fuel Limited, 4-108, Aza Okitsuke, Oaza Obuchi,
Rokkasho-Mura, Kamikita-gun, Aomori-ken 039-3212, Japan
e-mail: [email protected]
2
Japan Nuclear Fuel Limited, 4-108, Aza Okitsuke, Oaza Obuchi,
Rokkasho-Mura, Kamikita-gun, Aomori-ken 039-3212, Japan
e-mail: [email protected]
3
Japan Nuclear Fuel Limited, 4-108, Aza Okitsuke, Oaza Obuchi,
Rokkasho-Mura, Kamikita-gun, Aomori-ken 039-3212, Japan
e-mail: [email protected]
4
Japan Nuclear Fuel Limited, 4-108, Aza Okitsuke, Oaza Obuchi,
Rokkasho-Mura, Kamikita-gun, Aomori-ken 039-3212, Japan
e-mail: [email protected]
5
Ohsaki Research Institute, Inc., 2-2-2 Uchisaiwai-Cho, Chiyoda-ku,
Tokyo 100-0011, Japan, e-mail: [email protected]
6
Shimizu Corporation, 1-2-3 Shibaura, Minato-ku, Tokyo 105-8007, Japan
e-mail: [email protected]
7
Shimizu Corporation, 1-2-3 Shibaura, Minato-ku, Tokyo 105-8007, Japan
e-mail: [email protected]

Introduction
It is necessary to estimate the decrease in the radiation shielding ability of a
cracked concrete wall in a nuclear facility for safety reasons. Since the crack
width is very small, it is difficult to measure the change in the penetration rate of
radiation through the thick concrete shield. In this study, the shielding abilities
of cracked and uncracked walls were assessed by simulation using the three-
dimensional Monte Carlo code MCNP51) and its nuclear data library
MCPLIB042).

Calculation method
The calculation models were two models, simple and practical building model of
the repository for low level radioactive wastes. Only gamma rays were used as a
source of radiation, and their spectrum was conservatively obtained from the

28
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

calculated spectrum of vitrified radioactive wastes. A cubic room model with

dimensions of 3 m × 3 m × 3 m was used as the simple model, and the thickness
of the wall was 1200 mm. The source was set at the center of the room, and its
volume was 1 m3. It was assumed that a straight crack extending from the floor
to the ceiling would be formed in one of the walls. The crack width was varied
from 0 to 10 mm for the assessment. The cell tally was set outside the center of
the crack, and had two types. One was rectangular type that was 100 × 100 × 10
mm in size for simple model, and the other was cylindrical type that was 39 mm
in diameter and 60 mm in thickness for practical model. The dimensions of the
cylindrical one were the same as those of an ionization chamber. The number of
gamma rays generated during the simulation was up to 2.0 × 1011. The
calculation was performed for about 2 months using 16 CPUs, each of which
had a frequency of 4 GHz or more. Results obtained by calculation were
normalized with the strength data of practical radioactive wastes. The calculated
gamma-ray fluxes were converted into the dose rate by the flux-to-dose rate
conversion factors based on ICRP Pub.74.

Calculated results
The ratio of the penetration rate of radiation through a cracked concrete shield
to the penetration rate of radiation through an uncracked concrete shield was
obtained from the analysis. The thickness of the concrete shield was 1200 mm.
Ratios of 5, 50, and 1000 were obtained at crack widths of 1, 2, and 10 mm,
respectively. Since it was experimentally verified that the maximum crack width
should be less than 1 mm it was expected that the decrease of shielding ability
would become about 1/10 in the minimum. Parametric estimates of the ratio
were obtained at different wall thicknesses and a constant crack width of 1 mm.
Up to a wall thickness of 500 mm, the penetration rate of radiation through the
cracked wall was almost the same as that of radiation through the uncracked
wall. This was because the number of gamma rays that penetrated the cracked
wall was negligibly smaller than the number of gamma rays that penetrated the
uncracked wall. The ratio of the two penetration rates, as defined above,
increased when the wall thickness exceeded 500 mm. However, the ratio was at
the most 10 when the wall thickness was 1200 mm. In the case of the practical
model, the wall thickness was 1000 mm, and cracks were formed at intervals of
200 mm; further, the ratio was less than 3.

Conclusion
The penetration rate of radiation through a cracked concrete shield was
compared with that of radiation through an uncracked concrete shield in order to
estimate the decrease in shielding ability by using the Monte Carlo code
MCNP5. When the crack width was 1 mm, which is a practical value, and the

29
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

wall thickness was less than 500 mm, there was almost no change in the two
penetration rates. It was found that the penetration rate increased with the wall
thickness. The ratio of the two penetration rates was at the most 10 when the
wall thickness was 1200 mm. It was concluded that the increase in the
penetration rate of radiation through the concrete shield due to crack formation
is not a serious problem since the surface of the crack practically generated in
concrete walls is irregular and never smooth as calculated above.

References
1. X-5 Monte Carlo Team, MCNP: A General Monte Carlo N-Particle Transport Code,
Version 5, LA-UR-03-1987, LANL (2003).

2. RSICC, CCC-710/MCNP: Data Libraries for MCNP5 (2003).

30
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Study on radiation shielding performance of

reinforced concrete wall. (1) Loading test on
concrete walls and modeling of concrete
cracks (5-1866)

Yoshinari Munakata1, Takashi Maki1, Yoshiyuki Sato1, Keiji Sekine1,

Takamasa Nishioka2, Nobuyuki Niwa2, Osamu Kontani2
1
Japan Nuclear Fuel Limited, 4-108, Aza Okitsuke, Oaza Obuchi,
Rokkasho-Mura, Kamikita-gun, Aomori-ken 039-3212, Japan
e-mail: [email protected]
2
Kobori Research Complex Inc.
6-5-30 Akasaka, Minato-ku, Tokyo 107-8502, Japan

Introduction
Reinforced concrete structures that store radioactive materials are required to
maintain shielding capability as well as aseismic capability even after suffering
damage, mainly cracks, due to earthquakes. Many researches have been performed
on the aseismic capability of walls cracked during an earthquake, but few have
been conducted on their shielding capability.
In order to evaluate the shielding capability of concrete walls, horizontal
loading tests were conducted on RC and SC (steel plate reinforced concrete)
wall specimens to confirm their aseismic capability, and to obtain a better
understanding of their crack patterns and roughness of fracture surfaces. Cyclic
horizontal loads were applied to the specimens. Images of crack patterns were
taken with a high-resolution digital camera and crack width and length were
obtained by analysing the images. Core samples were taken through cracked
portions of wall specimens to investigate the roughness of the fracture surfaces.

Loading tests
Two RC wall specimens and one SC wall specimen were tested in this research.
The rebar ratio and rebar spacing of the RC wall specimens and the plate
thickness ratio and stud spacing of the SC wall specimen were made the same as
those of real walls so that the crack patterns obtained from loading tests would
be similar to real walls.
The RC wall specimens were 250 mm thick. One, called RC-1, had a rebar
ratio of 0.508%, which corresponds to the minimum rebar ratio of real walls.
The other, called RC-2, had a rebar ratio of 0.796%, which corresponds to the
average rebar ratio of real walls. The rebar spacing was 200 mm, as for real

31
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

walls. The SC wall specimen, called SC, was 300 mm thick. The SC steel plate
was 3 mm thick, based on the proportion for real walls. The stud spacing
200 mm, was the same as for real walls.
For RC specimen, cyclic loads were applied step by step up to a shear strain of
4 × 10-3, and then increased until shear failure occurred. For the SC specimen,
cyclic loads were applied step by step up to an ultimate shear strain of 6 × 10-3,
and it was then unloaded in order to observe the crack patterns of an unbroken
SC specimen.

Investigation of cracks
A digital camera having 4500 × 3000 pixels was employed to obtain the crack
patterns of the wall specimens. The resolution was 0.1 mm in crack width. The
images were digitally analyzed to obtain crack lengths and widths.
The maximum crack width was 2.6 mm for RC-1 and 2.0 mm for RC-2 when
subjected to 4 × 10-3 shear strain. After unloading, the maximum width was
reduced to 1.1 mm and 0.7 mm respectively. The maximum crack width was
0.2 mm for SC in the unloaded condition from 6 × 10-3 shear strain.
In order to investigate the fracture surface roughness of the wall specimens,
core samples were taken through cracked portions. Each core was easily broken
into two pieces and the fracture surfaces were digitally scanned to create 3D
images.
There were 6 core samples obtained from RC specimens. The maximum depth
of the fracture surface was 25.6 mm and the minimum depth is 9.6 mm. There
were 3 core samples obtained from SC specimen. The maximum depth of the
fracture surface was 29.7 mm and the minimum depth was 9.8 mm.

Modeling of cracks
In order to perform analyses of shielding capability of walls, the crack model
was established. A through crack was modeled by a straight slit having a
constant width in order to make the model conservative against all through
cracks. Then, the slit width was determined 1.0 mm considering the maximum
crack width observed in RC specimens. The crack patterns were simplified to
equally spaced parallel slits crossing at 45 degrees to the horizontal.

Conclusions
– It can readily be said that the loading tests of the RC and SC wall
specimens simulated the shear failure behaviour of real RC and SC walls
and that the crack patterns and the roughness of fracture surfaces observed
were appropriate for investigating shielding capability of the real walls.

32
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

– Cracks corresponding to 95% of total crack length had widths of less than
0.25 mm for both RC specimens. Cracks corresponding to almost 100% of
total crack length had widths of less than 0.1 mm.
– The crack spacing was 200 mm, which corresponded to the rebar spacing
of the RC specimens and the stud spacing of the SC specimen.
– Since depths of fracture surfaces are much greater than crack width, those
are too rough for radiation to go through the walls. It can safely be said
that radiation rays would hardly penetrate walls directly via these through
cracks.
– The through cracks in the RC and SC wall specimens were modelled by a
straight slit having a constant width of 1 mm. The crack patterns on the
walls were simplified to parallel slits of equal spacing crossing at 45
degrees to the horizontal.

33
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Seismic response impact of incoherent

SSI analysis by new hard-rock
coherency model (5-1874)

Sang-Hoon Lee, Joo-Hyung Kang

Korea Power Engineering Company, Inc.
360-9 Mabuk-dong, Giheung-gu, Yongin-si, Gyeonggi-do, South Korea
e-mails: [email protected], [email protected]

Many earthquake recordings show the response motions at building foundations

to be less intense than the corresponding free-field motions. To account for these
phenomena, the concept of spatial variation, or wave incoherence was
introduced. Several approaches for its application to practical analysis and
design as part of soil-structure interaction (SSI) effect have been developed,
which require a coherency model compatible with the soil condition of the given
SSI model. However, conventional coherency model didn’t reflect the
characteristics of earthquake data from hard-rock site, and their application to
the practical nuclear structures on the hard-rock sites was not justified
sufficiently. Seismic response impact of hard-rock coherency model proposed in
2007 on the incoherent SSI analysis is discussed in this study.
Case studies are performed to investigate the effects of site condition,
location, foundation type and spatial variation of input motion. The ground
condition represents typical medium-hard rock and hard-rock sites that have
shear wave velocity of 3,500 ft/s and 8,000 ft/s, respectively. A typical reactor
building of pressurized water reactor type is converted into three-dimensional-
beam-stick model. To identify the responses of rocking and torsion behavior,
additional edge points and rigid beam elements are added to the model. The
foundations of two different models are placed on surface and embedded into the
ground, respectively. Site specific response spectrum is defined as input motion,
which is constructed through near field earthquake sources considering the effect
of a fault around the site. Acceleration time histories composed of two horizontal
components and one vertical component are artificially generated, and those
spectral accelerations comply with 5% damped site specific response spectrum.
Each ground motion has total duration of 24 seconds with interval of 0.005
seconds. Seismic response from incoherent SSI analysis is also compared with
that from U.S. NRC Regulatory Guide 1.60 spectrum to identify the spectral
impact due to site specific response spectrum in high frequency range.
All seismic responses from coherent and incoherent SSI analysis are obtained
through SASSI computer code and INCOH module developed by Tseng to
implement the effect of incoherent spatial ground motion. INCOH was also
revised to add the new hard-rock coherency model developed by Abrahamson in

34
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

2007. Seismic response from new hard-rock coherency model is compared to

that from the conventional coherency model developed also by Abrahamson in
2005. Seismic responses for comparison are calculated applying mode
superposition of each component output and multiple input motions to reflect
randomness of input motion.
Several facts could be concluded from the case study results.
(1) Response reduction by wave incoherence is more obvious for
embedded foundation model than surface foundation case.
(2) The response caused by rocking and torsion effect due to incoherent
motion does not increase remarkably compared to the coherent case.
(3) Response having the unique peak at lower than 10 Hz does not show
response reduction at even higher than 10 Hz range.
(4) Response reduction effect at high frequency range due to incoherent
motion can be expected under hard-rock site as well as medium-hard
rock site.

35
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Experimental study on modal identification

and dynamic amplification of a steel
frame structure (5-1875)

Sung Gook Cho1, Yang Hee Joe2, Sung Tak Kim3, Sang-Kook Lee4, Gi Sung Pang5
1
JACE KOREA, Gyeonggi, Korea, e-mail: [email protected]
2
University of Incheon, Incheon, Korea, e-mail: [email protected]
3
KEPCO, Seoul, Korea, e-mail: [email protected]
4
Korea Institute of Nuclear Safety, Teajon, Korea, e-mail: [email protected]
5
Electric Power Technology Evaluation & Planning Center, Seoul, Korea

The seismic qualification of electrical equipment in nuclear power plants can be

normally performed by shaking table tests. However, these tests cannot be used
for the equipment that have already been installed in operating plants since the
equipment cannot be moved from the plants and mounted on shaking tables.
This case requires another method of seismic qualification, i.e., analysis or
combination of analysis and test. In addition, if complex equipment is to be
seismically qualified by analysis, modal identification test is needed to consider
reasonably the dynamic characteristics of analytical model. The dynamic
properties of electrical cabinets, in most cases, are calculated from an analysis
using finite element method. In some cases, experimental data obtained from
either an in situ modal testing or a shaking table testing has also been used to
estimate the cabinet dynamic properties.
Specifically for comparatively complex equipment with small devices which
is not easy to be mathematically modeled, the testing method is preferably
selected for the qualification. In the course of seismic qualification test program,
an identification test for the dynamic characteristics of the equipment, usually
called exploratory test, is performed prior to the main seismic proof tests to get
useful information for the determination of the best method and interpretation of
results of qualification tests. The modal identification test is also frequently used
for the verification of analytical models used in seismic qualification by analysis.
In this paper, the modal parameters of a steel frame model with 1-bay and
3-stories are identified by impact hammer tests. The shaking table tests are also
performed to analyze the dynamic amplification of the model. Forced vibration
tests have been conducted using harmonic input motions with varying frequencies.
Blind predictions and post-correlation analyses have been performed for the
forced vibration test. The results obtained from detailed finite element analyses
are verified with the corresponding results from shaking table testing of the
frame model.
The paper presents a structural identification procedure for the finite element
model updating in steel frame structures. This study consists of three parts: (i)

36
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

impact hammer tests and shaking table tests for harmonic loading, and modal
identification, (ii) baseline FE Modal updating using the stiffness modification
technique, and (iii) estimation of the dynamic amplification.
The effectiveness of the proposed procedure has been verified through an
experimental study on a steel frame structure with 1-bay and 3-stories. This
paper is also purposed to give a guideline for effective analytical modeling of
cabinet-typed electrical equipment by comparing the test results with the
analysis results from several different models.
The test specimen is a simple steel frame fabricated with square pipe with size
of 620 mm × 420 mm × 1500 mm. The total weight of the specimen is 56.4 kg.
The specimen was welded on a base plate with a thickness of 4 mm which was
affixed with bolts to the shaking table. The shaking table has a maximum
loading capacity of 500 kg.
The results of the modal analyses show good agreement with those obtained
by modal tests. And the analysis results of the updated finite element model and
the harmonic excitations are improved. Dynamic amplification factors of the
structure under El Centro earthquakes obtained respectively by experimentally
and numerically also show good agreement.

37
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Performance of X-plate elasto-plastic

dampers. A passive seismic supports for
nuclear piping under cyclic loading (5-1877)

P.N. Dubey*, G.R. Reddy, K.K. Vaze, A.K. Ghosh, H.S. Kushwaha
*e-mail: [email protected]

Keywords: passive supports, Snubbers, locking, EPDs, radiation hazards, damping

In the nuclear power plant (NPP) piping design, major loads considered are
pressure, dead weight, seismic and loads due to restraint to thermal expansion.
The thermal stresses and seismic stresses are contradictory to each other. To
reduce the former, piping should be flexible and for the later it should be rigid.
Hence it becomes very tedious to meet these two contradictory requirements
using conventional supports. In this condition snubbers are used, which allow
the gradual thermal expansion and arrest the sudden motion due to earthquake.
From the past experiences snubbers have proved to be very costly, expensive
and need frequent maintenance, leakage problem in hydraulic snubbers and they
also congest the space because of more space requirement for installation.
Sometimes it is also observed that the mechanical snubbers lock during normal
operation and cause undue thermal stresses in the piping and nozzles. Recently a
trend has been started to use dampers in place of snubbers. Normally X- shaped
plate is chosen as an elastoplastic energy absorbers such that the strain is
constant over the height of the device, thus ensuring that yielding occurs
simultaneously and uniformly over the full height of the damper. X-plate
elastoplastic dampers (EPDs) are preferred because of their high seismic energy
absorbing capacity, simple design, low cost and maintenance free operation.
EPDs are based on plastically deforming steel components or layered laminated
plate in flexure, shear, torsion or a combination thereof. For critical applications
like NPP, where safety of public and environment from undue risk of radiation
hazards is prime concern, it is necessary to evaluate the performance of supports
under seismic loads by testing before implementation. In the present paper
testing of 6 mm thick X-plate EPDs made of SS316L material has been
performed for evaluation their performance under cyclic loads at different
frequencies and tip displacements. By testing if was found that they can sustain
many cycles of stable yielding deformation, resulting in high levels of energy
dissipation (damping).

38
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Inelastic axisymmetric analysis of BARC

prestressed concrete containment
model (5-1885)

Tarvinder Singh*, R.K. Singh, A.K. Ghosh

Reactor Safety Division
Bhabha Atomic Research Centre
Trombay, Mumbai, India
*e-mail: [email protected]

Reactor Safety Division (RSD) of Bhabha Atomic Research Centre (BARC) has
initiated an experimental program at BARC Tarapur Containment Test Facility
to evaluate the ultimate load capacity of Indian Pressurized Heavy Water
Reactor (PHWR) containments. For this study, BARC Containment Model
(BARCOM), which is 1:4 scale representation of Tarapur Atomic Power Station
(TAPS) unit-3&4 540 MWe PHWR Inner Containment of Pre-stressed Concrete
has been constructed. The model includes all the important major design features
of the prototype containment structure and simulates Main Air Lock (MAL),
Steam Generator (SG), Emergency Air Lock (EAL) and Fueling Machine Air
Lock (FMAL) openings. The design pressure (Pd) of BARCOM is 1.44 kg/cm2
(g), which is same as the prototype containment structure. For the experimental
program of the ultimate load capacity evaluation it is desirable to identify the
critical locations of various types of sensors on concrete and steel members of
BARCOM. In addition, it is also desirable to understand the behavior of
containment model under internal pressure and study the various failure modes
and the expected elastic/inelastic response at the identified critical locations,
which are important for instrumentation / monitoring during the experiment. In
the present work the pretest analysis of BARCOM has been performed with
finite element axi-symmetric modeling. The analysis has been performed using
2D axi-symmetric finite element model representing the 157.5 degree azimuth.
This free field region is free from local stresses due to discontinuity of buttresses
or penetrations and has been developed to predict the structural response for
static over-pressurization load. The concrete structure is modeled with 8-node
continuum axi-symmetric elements. The hoop reinforcements and tendons are
modeled as rebar elements, which are represented as steel layers of equivalent
smeared thickness in a particular continuum axi-symmetric element. These rebar
elements have uni-axial behavior resisting only the axial force in the bar
direction, which is the hoop direction in the present model. The longitudinal
reinforcements and tendons were modeled as embedded axi-symmetric
membrane elements with orthotropic material properties so that all the bars carry
stress only along their individual axial directions. The thicknesses of the steel

39
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

layers have been calculated so that it represents the BARCOM reinforcement

and pre-stressing tendons in the axi-symmetric model. A constant pre-stress has
been applied as initial stress in the hoop rebar elements and embedded
longitudinal tendon elements. It has been assumed that there is no slip between
the concrete and steel rebar / embedded members. To consider the effect of the
reinforcement, the tension stiffening is used in concrete material model. The
structural response and the various failure modes of BARCOM were assessed
through non-linear analysis. Elastic/inelastic properties of the concrete and steel
members were taken into consideration to trace the response of the structure
during over-pressurization. It was concluded that the BARCOM has an ultimate
load capacity factor of 3.54 Pd for the identified failure modes which were
studied on the axi-symmetric finite element model. Further analysis of the
BARCOM with 3D shell element model is reported for the ultimate load
capacity evaluation after considering the influence of various openings.

40
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Flow induced vibrations for reactor internals

of PWR – Less art, more science (5-1883)
Nicolas Jobert1, Jean-Luc Chambrin1, Thierry Muller2, Benoît Migot2
1
AREVA-NP: Primary Components Dept, Courbevoie, France
2
AREVA NP: Tehnical Center of Le Creusot, Le Creusot, France

Aim of the study

In the design process of any component, some aspects are always considered as
‘too difficult to be really predictable’ and rather wisely, engineering wisdom
leads to favor robust design over fine-tuned design. As a general rule, such grey
areas occur at the interface between engineering disciplines, and FIV most
surely belongs to that category.
Within the scope of Reactor Pressure Vessel Internals, it is the wish of the
authors to shed some light on the matter firstly by reviewing the important
features of FIV analysis and secondly by recognizing which particular topics can
be adequately described using current tools and which advances may be
desirable and/or achievable.

Short description of the work

This paper will first cover a review of available analysis techniques, from hand
calculations to fully detailed numerical simulations. After discussing the
compared merits of these increasingly complex approaches, some considerations
will be made about the parameters needed to significantly enhance the reliability
of such evaluations.
The different aspects covered will include the whole calculation process, from
forcing functions models to structural response evaluation. The key assumptions
under each method will be reviewed and, where applicable, sensitivity studies
will be performed in order to evaluate the potential bias induced by the
simplifying and idealization procedures routinely followed.
A particular emphasis will be put on the possible synergy between experimental
and analytical approaches, and their relative strengths and weaknesses.

Conclusions
After having reviewed the physical phenomena involved, and how adequately they
can be captured by various approaches, the primary contributors to a robust
evaluation of FIV levels are identified. For each of them, a ‘best practice’ approach
is described and some recommendations for future development are provided.

41
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Figure 1. Example mockup used for hydraulic testing.

Figure 2. Typical FEM model used for a modern Core Barrel FIV analysis (4-loops PWR).

42
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Structural modeling and analysis of

the SMART-2008 shaking table
specimen (5-1892)

Marco Domaneschi, Maria Gabriella Mulas

Department of Structural Engineering, Politecnico di Milano, Milano, Italy
e-mails: [email protected], [email protected]

The problem of analysis and design of 3D buildings prone to relevant torsional

effects when subjected to seismic excitation is still an open one. These effects
can be particularly relevant for the reinforced concrete (RC) structures typical of
the nuclear industry. For this reason the two French companies EDF and
Commissariat à l’Energie Atomique have recently launched the SMART-2008
project (Seismic design and best-estimate Method Assessment for Reinforced
concrete buildings subjected to Torsion and non linear effects). A 3D scaled
model (scale 1/4th) representative of a nuclear RC structure is currently being
tested on the Azalee shaking table (at CEA research center in Saclay, France)
under a series of bi-directional accelerograms of increasing amplitude and able
to induce non linear behavior in the structure. The 3-storey building has an
irregular shape in plan and several openings in the walls: both factors are able to
emphasize the torsional effects under seismic loading. An international
benchmark1 contest has been launched in parallel to the experimental tests. The
participants have to predict the structural behavior under different seismic
loading, to the aim of: (a) modeling and analyzing typical buildings of nuclear
power plants; (b) predicting the excitation experienced by equipments within the
building; (c) determining fragility curves for structural components.
In this paper the initial work done by the authors within the context of the
benchmark will be described. A brief review of the European codes prescriptions
on 3D effects is presented. A few linear analyses have been performed according
to code prescriptions, making use of both simplified plane models and complete
3D models. The capability of plane models in predicting the dynamic behavior
of the specimen is thus assessed. High stress levels are expected in the structure
elements at high amplitude levels of seismic excitation: the numerical
environment of the ANSYS FEM code is adopted to describe the structural non
linear behavior. The building is discretized in finite elements so as to perform
the blind prediction of the structural behavior, which is the main goal of the first
phase of the benchmark problem. The walls and the slab are firstly modeled as
solid linear elastic elements to tune the structural model and avoid the
difficulties one can meet in the analysis of a complex structure. The linear model
is representative of the specimen under low levels of excitation. After the first
linear elastic step, the nuclear specimen is modeled by nonlinear elements which

43
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

consider the real behavior of the material when subjected to strong ground
motions. In particular, the reinforced concrete is modeled through the ANSYS
available material model, which allows to simulate the concrete behavior with
the presence of steel bars. The concrete model describes the failure of brittle
materials; both cracking and crushing failure modes are accounted for.
Geometric nonlinearities are also considered. The model is analyzed in the
dynamic range: both spectral and time-history analysis are adopted in the linear
range, while only time-history analyses are performed in the non linear range.
The numerical results, both in the linear and in the non linear range, will
provide useful indications on the modeling and analysis strategies to be adopted
for RC structures prone to 3D effects when subjected to seismic excitation. The
comparison with experimental results, when available, will suggest possible
improvements in the numerical models and analysis, further increasing the
usefulness of the numerical analyses here presented.

Reference
1. Lermitte, S., Chaudat, T. Presentation of the blind prediction contest, Project SMART
2008 Specification. CEA, DM2S, SEMT/EMSI/PT/07-003/C, May 2007.

44
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Structural dynamic analysis of a non

symmetrical RC building within the scope
of a blind prediction contest – Project
SMART 2008 (5-1900)

Yves E. Mondet1, Adele Klein2, Urs Bumann3

1
Basler & Hofmann Consulting Eng., Forchstrasse 395, CH-8032 Zürich
e-mail: [email protected]
2
Basler & Hofmann Consulting Eng., Forchstrasse 395, CH-8032 Zürich
e-mail: [email protected]
3
HSK – Swiss Federal Nuclear Safety Inspectorate, CH-5232 Villingen
e-mail: [email protected]

The Commissariat a l’Énergie Atomique (CEA) has launched a blind prediction

contest project named SMART 2008 (Seismic design and best-estimate Methods
Assessment for Reinforced Concrete buildings subjected to Torsion and non-
linear effects) in May 2007 which will be concluded with a final workshop in
2010. The objective of the benchmark is to predict the dynamic behaviour of a
non-symmetric 3-dimensional RC building that is designed according to the
French nuclear methods and subjected to seismic excitation. More than 30
international teams consisting of industrials, nuclear/energy corporations and
research facilities are taking part in this blind prediction contest. The project has
got two phases: phase one – seismic analysis of the nuclear reinforced concrete
building and phase two – variability quantification and fragility assessment.
Phase 1 includes a blind predictive benchmark, an experimental test and a
prediction upgrading.
Subject to this paper is the numerical seismic analysis for the structure carried
out during phase 1 which is taking place from May 2007 to December 2008 by
the team HSK (Swiss Federal Nuclear Safety Inspectorate) and Basler und
Hofmann Consulting Engineers. The team belongs to the group of industrials
participating in the contest. Methods and results of the best estimate analysis and
the post-adjustments due to the findings during the experimental tests will be
described.
The best estimate structural dynamic analysis has been carried out by means
of modal time history analysis using different specified synthetic and real
accelerograms. A 3-dimensional model has been developed for the concrete
structure. The properties of the numerical model have been modified in order to
account for the specimen’s real behaviour during dynamic excitation. Non linear
material behaviour due to higher loads is taken into account.

45
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

The experimental tests have been carried out at the 3-dimensional RC

specimen on the AZALEE shaking table at CEA. The results that are presented
to the participants can be compared to the results predicted by means of the
numerical analysis. Insights into the genuine characteristics of the model, findings
about when exactly strength degradation starts and about the actual amount of
energy dissipation can be gathered. The comparison of the test results and the
results of the participants’ modal analysis show that the shaking table must have
an important influence on the mode shapes of the structure. The effect of the
shaking table as well as the other previously mentioned effects have been
evaluated and taken into account in the refined numerical analysis.

46
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Dynamic analysis of a concrete

shear-wall (5-1908)

Jaegyun Park1, Chul-Hun Chung1, Jang Seok Y2, Byung-Moo Jin3

1
Professor, Dankook University
2
Researcher, Hyundai Institute of Construction Technology Development
3
Researcher, Daewoo Engineering & Construction

Introduction
Current trend of earthquake resistant design is performance based one which
limits the maximum displacement under the load. To evaluate the effectiveness
of the displacement control under the near-field ground motion due to
earthquake, IAEA initiated CRP program. As a first step, they performed an
analysis of a shear wall and compared the result with the shaking table test
results. In this paper, we use better concrete and rebar model to regenerate the
test results using ABAQUS, a general purpose nonlinear FE program, and
compare the result with other calculations.

Analysis
We used ‘Concrete Damage Model’ embedded in ABAQUS 6.4.1, which is
originally proposed in Lubliner et al. (1989) and further developed in Lee and
Fenves (1998).
The model of the concrete shear wall came from the previous report
KINS/GR277 (Hyun et al. 2004) by KINS. Figure 1 described the original shear
wall for the real test and Figure 2 presents the two dimensional model of the
shear wall with rotational and translational springs. The concentrated masses
were put as the Figure 3 with the same intervals. A dynamic analysis on this
model resulted in the 3 initial modes of the structure (Figure 4), which are
similar to the modes of beam-stick model in KINS/GR277 report.

Conclusions
The model analysis results are similar to the previous results, which show that
current material model is in good agreement with test specimen within elastic
limit. The strength of the structure is related to behavior beyond the elastic limit
such that we calibrated the material property parameters to regenerate the
dynamic response of the real test structure under serial earthquake loading,
where the nonlinear damage model plays an important role.

47
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Figure 1 Figure 2

Figure 3 Figure 4

References
1. Hyun et al. IAEA Coordinated Research Program (CRP), Safety Significance of Near Field
Earthquake/Assessment of Near Field Earthquake Effect, KINS, GR-277, 2004.

2. Anil. K. Chopra. Dynamics of Structures, Prentice Hall, 1995.

3. Clough, R.W., Penzien, J. Dynamics of Structures, McGRAW-HILL, 1975.

4. Lubliner, J., Oliver, J., Oller, S., Onate, E. A Plastic-Damage Model for Concrete, Int. J.
Solids Structures, Vol. 25, No. 3, pp. 299–326, 1989.

5. Lee, J., Fenves, G.L. Plastic-Damage Model for Cyclic Loading of Concrete Structures,
Journal of Engineering Mechanics, Vol. 124, No. 8, pp. 892–900, 1998.

6. Park et al. Static and Dynamic Analysis of a Concrete Shear-Wall, SMiRT 19, Toronto, 2007.

48
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Application and evaluation of “design by

rule” procedures applicable to nuclear
power plant ASME B & PVC section iii
class 2 and 3 piping (5-1911)

John D. Stevenson
J.D. Stevenson, Consulting Engineer

The text of the abstract is placed here. The abstract should be concise and should
present.
This paper describes a “Design by Rule” procedure that could be used to
seismically design nuclear safety related cold (to < 150°F, 66°C) piping. The
procedure pre-engineers the location of transverse pipe supports, which will
maintain the piping system within applicable code stress limits which include
seismic design loads.
Commercial nuclear power plants and other nuclear material and waste
processing plants typically contain over 150,000 feet (48,000 meters) of cold
safety related piping requiring seismic design. The engineering effort using
conventional stress analysis procedures typically requires over 400,000 man
hours per plant. By use of the “Design by Rule” procedure suggested herein this
engineering man hour effort could be reduced by at least 70 percent with no loss
of design conservatism.
The “Design by Rule” procedure consists of locating transverse pipe supports
as multiplier of dead weight support spacings. Typical piping construction codes
such as ASME B&PVC Section III, Subsection NF and ASME B31.1 recommend
dead weight support spacings which result in a prescribed longitudinal dead
weight stress in the piping typically defined as around 0.1 Sc where Sc is the
allowable stress in the pipe.
By specifying multiples of these dead weight support spacings it is possible to
determine the following:
• The dominate frequency of the piping of the piping in transverse and
vertical directions
• The resultant seismic force to be applied to the piping taken from the
applicable seismic response spectra
• The limiting seismic and total longitudinal stress in the piping system as
a function of the support spacings.
With this information and the procedure developed in this paper it is possible to
demonstrate that a piping system is within code allowable limits without the

49
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

effort to prepare an analytical computerized model of the piping system and to

determine the frequency of the piping system.
The paper will demonstrate the application of the “Design by Rule” analysis
procedure for typical piping system materials and layouts to include detailed
analytical finite element response spectral analysis of piping systems which
verify its applicability.

50
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Numerical studies on pre-stressed impact

loaded concrete walls (5-1921)

Markku Tuomala1, Kim Calonius2, Arja Saarenheimo2, Pekka Välikangas3

1
Department of Civil Engineering, Tampere University of Technology
Tampere, Finland, e-mail: [email protected]
2
VTT Technical Research Centre of Finland, Espoo, Finland
3
Radiation and Nuclear Safety Authority (STUK), Helsinki, Finland

The various protective concrete barrier walls of nuclear power plants are
required to withstand the effects of impacts by accidental or intentional missiles.
Structural systems and solutions are under development both in building
framework and in detail level, which require more sophisticated tools for
different design phases. For example detailing of shear reinforcement is under
development. Therefore numerical methods have been developed and taken in
use for predicting the response of pre-stressed shear reinforced concrete
structures subjected to impacts by hard projectiles. The impact load function on
reinforced concrete wall caused by a hard missile is studied. Predicted impact
loads are further used in structural analyses. Alternatively, the dynamical contact
between the projectile and the target plate is modelled with nonlinear FEM.
Structural behaviour of the impact loaded pre-stressed walls has been predicted both
by analytical methods and by involved non-linear FE-models. Analysis methods to
predict associated damage mechanisms like crater formation, penetration, shear cone
formation and perforation are examined. Experimental data is needed in order to verify
the accuracy of numerical models. In this paper, numerical results obtained using
different kinds of methods are compared with experimental data and observations on
impact loaded pre-stressed reinforced concrete walls with shear reinforcement and
with pre-stress levels of practical applications.
An experimental set-up has been constructed at VTT for medium scale impact tests.
The main objective of this effort is to provide data for the calibration and verification
of numerical models intended to be used in full scale practical applications.

51
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Seismic analysis and upgrading of

suspended ceilings and air ducts over main
control room, emergency control room and
control and protection systems of units 3 &
4 of WWER-440 MW NPP Kozloduy (5-1931)

Stanislav Georgiev1, Marin Jordanov2

1
Structural and Seismic Engineer
EQE Bulgaria AD, H. Smirnenski Blvd. 1, Bulgaria, Sofia
e-mail: [email protected]
2
Structural and Seismic Engineer
EQE Bulgaria AD, H. Smirnenski Blvd. 1, Bulgaria, Sofia
e-mail:[email protected]

The objective of the seismic analysis of the suspended ceilings and air ducts was to
certify that the existing structure can / can not withstand an earthquake defined as
Floor response spectra (FRS) from Review level earthquake (RLE) and Local
earthquake (LE) for the specified location of installation (elevation +9.60) at
Units 3 & 4 of Kozloduy NPP. The system of suspended ceilings and air ducts
was classified as seismic category I because of direct danger for the personnel
and the equipment for control of the nuclear reactor situated under the ceilings in
Main control room (MCR), Emergency control room (ECR) and the Control and
protection systems (CPS). The main bearing structure of the ceilings was built of
double T steel sections in longitudinal direction united with U profiles in
transverse direction which were hanged to reinforced concrete girders via
hangers made mainly of angle and U profiles with length 3–4 m. For ECR on the
bottom of double T sections were welded gratings of bars to which Rabitz net
was attached and then lime-cast to the net from below. For MCR on the bottom
of double T sections were welded gratings of T profiles to which gypsum plates
were mounted. Part of the ceiling over MCR was additionally lowered with 1 m
by using bars for hangers and gypsum plates below. The air ducts had
rectangular shape and were also attached to the girders by hangers made mainly
of bars. The vertical branches of the air ducts had contact with the ceiling
structure via angle profiles closely fit from all four sides to each vertical branch
and welded to the bearing structure of the ceilings.
Two separate models of the steel structure were made on SAP 2000 software –
one model for the suspended ceiling over MCR and one over ECR and CPS. The
air ducts were also modeled and the contact to the ceiling structure was taken
into account. As the contact of the suspended ceiling on its periphery to the
surrounding concrete walls and columns was via the lime-cast or the gypsum

52
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

plates that have very small rigidity, the horizontal supports in the models were
neglected. Because of the lack of vertical X braces and relatively small rigidity
of the hangers in case of horizontal excitation (earthquake), the calculated
displacements were relatively high which lead to breaking of the lime-cast and
the gypsum plates at the periphery which is inadmissible. At the same moment
the internal forces in the hangers were very high and the calculations made
according to ANSI N690 code showed that they did not have enough bearing
capacity in case of seismic excitation defined with FRS for I category structure.
A lot of upgrading measures (connecting of the steel structure to the
reinforced concrete walls via anchors, additional vertical X braces, especially for
the lowered ceiling with 1 m over MCR, additional supports of the air ducts etc.)
were made for Units 3 & 4. The upgraded models were run. The displacements
were reduced to minimum and the forces in the elements lowered as well. The
calculations for the bearing capacity of the existing and the new upgrading steel
elements of the suspended ceilings and air ducts over MCR, ECR and CPS
showed that the system of suspended ceilings and air ducts can bear the
prescribed seismic excitation defined for I category structure of Units 3 & 4 of
NPP Kozloduy.

53
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Seismic analyses of safety important piping

systems situated in reactor building of units
5 and 6, Kozloduy NPP (5-1932)

Pavel Spassov1, Marin Jordanov2, Maya Kancheva3,

Georgy Kostov4, Petar Bakardjiev4
1
Systems Division Manager, EQE Bulgaria AD
e-mail: [email protected]
2
Structural and Seismic Engineer, EQE Bulgaria AD, Sofia
e-mail: [email protected]
3
Senior Engineer, EQE Bulgaria AD
e-mail: [email protected]
4
Mechanical Engineer, EQE Bulgaria AD
e-mails: [email protected], [email protected]

The aim of the task was to qualify seismically the existing safety important
piping installed in the reactor building. The need for seismic qualification came
from reassessment of review level earthquake (RLE) established for Kozloduy
NPP site. The pipelines should be properly qualified according to the new higher
seismic level. The task was included in the Modernization Program developed
for Units 5 and 6. The program was executed in two basic stages: Basic
Engineering Phase (BEP) and Main Contract Phase (MCP). The selection of the
piping to be seismicaly qualified was done in BEP, together with some
preliminary analyses by Framatome ANP. The final analyses, and the detailed
design for improvement of the seismic stability of piping were performed by
EQE Bulgaria experts, during the MCP.
Some of the safety important systems pipelines, which were analyzed, are
located in the containment of the units (pipelines from systems TQ, TG, TK, TX,
YR and YP), and the others are in the reactor building outside the containment
(TQ, VF, TG, TC, UJ). The systems listed are: reactor core cooling, containment
spray, primary circuit purification, emergency feedwater for steamgenerators,
pressurizer – bubbler surge lines, spent fuel pool cooling, fire extinguishing and
essential service water.
For the analysis of the pipelines, a spectral method was used with suitable for
the purpose calculating software, which allows the use of response spectra
analysis and uses the ASME Boiler and Pressure Vessel Code, Section III: Rules
for Construction of Nuclear Power Plant Components, for pipelines Class 1, 2
and 3.
For the analysis of Class 1 pipelines, the software PepS was used, which
incorporates the processing software of DST Pipestress and the graphic interface
for input and output by Tractebel Engineering’s EditPipe.

54
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

For the analysis of Class 2,3 pipelines the Algors’s software Pipeplus was
used.
The loads and the load combinations incorporated into the models are
according to the requirements of the ASME code. The results of the analyzed
pipelines were compared to the AMSE criteria.
For the input dynamic loads from seismic, a reduction coefficient Fμ was
used, which takes into account the ductile behavior of the pipeline systems. The
used floor response spectra were for Review Level Earthquake (RLE) with local
component LLE, which were developed for Kozloduy NPP.
As a result from performed analyses the pipelines are divided into two groups
– the first one include those pipes which need upgrade of support system, and
the second one group comprise the pipes, which models demonstrated adequate
capacity to load combinations, including seismic load corresponding to latest
defined level. In general, 1/3 of the analyzed pipelines did not satisfy the criteria,
totally 46 out of 132 lines. Most of the “weak” pipelines belong to the fire
extinguishing system.
Corresponding measures for upgrade of pipe support systems were proposed.

55
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Strain-based acceptance criteria for section

III of the ASME boiler and pressure vessel
code (5-1940)

Gordon S. Bjorkman, Jr.1, Doug Ammerman2

1
Nuclear Regulatory Commission, Washington DC, USA
e-mail: [email protected]
2
Sandia National Laboratory, Albuquerque, NM, USA
e-mail: [email protected]

Modern finite element codes used in the design of nuclear material transportation
and storage casks can readily calculate the response of the packages beyond the
elastic regime. These packages are designed to protect workers, the public, and
the environment from the harmful effects of the transported radioactive material
following a sequence of hypothetical accident conditions. Hypothetical accidents
considered for transport packages include a 9-meter free drop onto an essentially
unyielding target and a 1-meter free fall onto a 30-cm diameter puncture spike.
For storage casks, accident conditions can include drops, tip-over, and aircraft
impact. All of these accident events are energy-limited rather than load-limited,
as is typically the case for boilers and pressure vessels. Therefore, it makes sense
to have analysis acceptance criteria that are more closely related to absorbed
energy than to applied load. Strain-based acceptance criteria are the best way to
meet this objective.
The U.S. NRC has a long history of assuring the safety of the public from the
potential hazards associated with the transportation of radioactive material. For
most of this history, the design of the packages used to transport this material
has been based upon the ASME Boiler and Pressure Vessel Code and guidance
has been provided by U.S. NRC Regulatory Guide 7.6. For the past decade, the
section of the Code that is most relevant to the design has been Section III,
Division 3. This section of the Code is based upon the concept of stress intensity,
which is twice the maximum shear stress. The allowable stress intensities vary
according to loading case and type of stress. For some of these, the allowable
stress intensity is larger than the yield stress of the material, a tacit approval for a
limited amount of plasticity. This approach was necessary when stresses were
determined with hand calculations and was still beneficial during the early days
of finite element analyses. As finite element calculations became more detailed,
it has become possible to determine the stress state at any point in the package
and the associated strains. Since the Code has allowed limited plasticity, modern
package designers would prefer to use inelastic analysis techniques to calculate
the stresses and strains that result from the required loading conditions. There

56
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

are two ways to implement inelastic analysis: continue using stress-based

acceptance criteria, or; develop strain-based acceptance criteria.
This paper will briefly discuss the efforts within the ASME, detail the
advantages of using strain-based criteria, discuss the problem areas associated
with establishing strain-based criteria, and provide insights into inelastic
analyses as applied to radioactive material transportation and storage casks in
general. The views expressed represent those of the authors and not necessarily
those of their respective organizations or the ASME.

57
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

A simple dynamic model for estimating the

effect of gaps on response of a spent fuel
transportation cask closure lid during
a drop impact (5-1941)

Gordon S. Bjorkman, Jr.

Nuclear Regulatory Commission, Washington DC, USA
e-mail: [email protected]

During an impact event, gaps between the various components of a spent fuel
transportation cask may create secondary impacts that result in higher dynamic
loads than would have occurred if the gaps had not been present. A condition of
particular interest is the gap that may exist between the fuel assemblies and cask
closure lid and the effect this gap may have on amplifying the response of the
closure lid during an impact.
When spent fuel is transported in a transportation cask, gaps exist between the
cask closure lid (the containment boundary) and the cask internal components
(fuel assemblies, fuel basket, etc.). If a transportation accident was to occur these
gaps may lead to a secondary impact on the lid that could significantly increase
the response of the lid above the values that would have occurred if the gap had
not been present. For the 30 foot drop, this is why the regulations in 10 CFR Part
71.73(c) (1) require the cask to be dropped “in a position for which maximum
damage is expected.” Position is made up of both the orientation of the cask as
well as the geometric position of the cask and its internal components relative to
one another. Thus to comply with the regulation, gaps whose size is sufficient to
significantly influence the dynamic response of the closure lid or internal
components must be incorporated in tests and finite element analyses of
transportation casks.
Through the use of a simple dynamic model this paper investigates the effect
of a secondary impact due to a gap between the cask internals and the cask
closure lid on the response of the closure lid during a 30 foot end drop (or c.g.
over corner drop). The dynamic model consists of five components (parameters):
(1) The mass of the internals traveling at the impact velocity for a 30 foot drop
(44.4 ft/sec), (2) the gap between the internals and cask lid, (3) the cask lid,
assumed to be a simply supported circular plate, (4) the modal mass of the lid,
and finally, (5) an impact limiter that applies a constant deceleration to the cask
overpack. In addition, the dynamic model assumes elastic behavior. This is
consistent with the Standard Review Plan (NUREG-1617), which recommends
that the closure lid bolts and closure lid system within the region of the lid bolts
remain elastic in order to demonstrate leak-tightness by finite element analysis.

58
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

The response results are presented in terms of the Dynamic Load Factor
(DLF) for the closure lid. Response is shown to be a nonlinear function of the
impact limiter deceleration, gap size and closure lid diameter and thickness.
These results provide valuable insights into the parameters that effect response
and show the conditions under which gaps may be of sufficient size to
significantly influence response. The NRC Staff plans to compare these results
with drop results from detailed LS-DYNA models.

59
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Numerical model of the thermal and

mechanical behavior of a CANDU
37-element bundle (5-1942)

Lei Jiang1, Ken MacKay1, Robert Gibb2

1
Senior Research Engineer, Martec Limited, Halifax, Nova Scotia, Canada
2
Reactor Thermalhydraulics Branch, CNSC, Ottawa, Canada

Investigation of the bowing deformation of fuel elements is important for

assessing the integrity of fuel and the surrounding structural components under
different operating conditions including accidents. The bowing of a fuel element
is defined as the lateral deflection of the element from the axial centerline and
the magnitude of bow is the maximum deflection between points of restraints.
Bowing can have significant effects on performance and safety of the nuclear
reactor. For instance, during normal operations, bowed fuel elements could
reduce sub-channel flow area resulting in poor heat transfer due to local coolant
starvation, causing these elements to deflect as a result of local overheating.
Under accident conditions, bowing of fuel elements can lead to safety concerns
for fuel coolability and the pressure tube integrity.
The integrity of the pressure tube under accident conditions is a requirement
for effective trips in CANDU reactor design. As a result, an understanding of all
of the behaviors of the pressure tube in various different accident scenarios is
crucial for understanding the trip effectiveness. One of the important pressure
tube behaviors that needs to be investigated is the coupled thermal-mechanical
interaction between the fuel bundle and pressure tube. In the early stages of a
loss of coolant event, it is known that fuel bundles deform or barrel so that the
outer elements bend outward. But the magnitude of the distortion is not readily
quantifiable. The bundle barreling could result in the bundle coming into
increased contact with the pressure tube. If this were to happen, the contact force
between bearing pads and the pressure tube would be increased which increases
the contact area on the bearing pad and pressure tube, and enhances the heat
transfer from the fuel elements to the pressure tube, creating localized hot spots
and high stress areas. The combination of higher temperature and higher local
stress could lead to pressure tube failure. If this were the case, then the
likelihood and numbers of pressure tube failures in loss of coolant and other
accident conditions would be greater than presently predicted. This would
significantly affect the current under-standing of margins to severe core accidents.
The numerical simulation of the fuel bundle to pressure tube interactions
requires a reliable numerical bundle model which is capable of predicting
macroscopic thermal and mechanical behavior of the fuel bundle assembly under
different accident conditions. In order to obtain realistic solutions, all the key

60
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

thermal and mechanical features of the fuel bundle, such as temperature-

dependent nonlinear material properties, fuel-to-sheath interactions, endplate
constraints and contact between fuel elements, need to be included. In this paper,
we present a finite element based numerical model for nuclear fuel bundles,
verify its accuracy and demonstrate its suitability for being utilized to investigate
fuel bundle to pressure tube interaction in future nuclear safety analyses. The
present study was carried out in three stages.
The first stage involved development and verification of a finite element-
based numerical model for predicting temperature distributions and thermal-
induced barreling deformations of a complete 37-element fuel bundle. For
structural analysis, a 3D beam model was developed which was able to compute
thermal-induced macroscopic deformations caused by axial and radial
temperature variations in fuel elements and endplates. On the other hand, a 2D
heat transfer model was used to predict asymmetric temperature distributions
over the mid-section of fuel elements due to the effect of neutron flux gradient
and uneven coolant conditions. The results from this 2D heat transfer model
were then combined with a previous axi-symmetric temperature solution to form
a complete 3D approximation of the temperature field in fuel bundle. Both the
structural and heat transfer analysis models were verified against available
experimental data and numerical solutions, but no comparison of model
predictions against an integrated bundle deformation data was attempted.
The second stage of the present study involved application of the bundle
model to predict static deformations of a fuel bundle to various steady-state
power and coolant conditions. Setting up the heat transfer problem required
extraction of asymmetric coolant conditions from the results of thermo-hydraulic
calculations and adjustment of power generation rate to account for neutron flux
gradients. The endplates were assumed to be at the coolant temperature, which
was normally variable over the bundle cross-section. The approximate 3D
temperature distribution was then applied to the structural model to predict
bundle deformation and solutions were obtained for two different steady-state
thermal conditions. These solutions indicated that all the fuel elements bowed in
the anticipated direction and the overall pattern of the deformation was
consistent with the temperature distribution. Element-to-element interactions
through the spacer pads played a very important role in maintaining the shape of
bundle cross-section. If these interactions were not considered, the maximum
barreling deformation of each fuel element occurred at the mid-span and the
magnitude of deformation was nearly proportional to the temperature gradient
over the fuel element, suggesting that the temperature gradients in fuel rods are
the primary driving force for bundle barreling.
In the third stage, the steady-state solution procedure was extended to handle
transient thermal conditions, such as the loss of coolant accident (LOCA)
conditions. This required implementation of the nonlinear visco-plastic material
model for Zircaloy-4 into the beam finite element and generalization of the axial
scaling technique for combining the planar and axisymmetric heat transfer

61
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

results to form a full 3D transient temperature field of the entire bundle. An

investigation was conducted to identify the most important mechanism for
barreling deformation of the fuel bundle and it indicated that although the
differential on rod elongations played a role on bundle barreling, the most
important driving force was the temperature gradients in the fuel rods.
Numerical studies presented in this paper demonstrate that the present finite
element based fuel bundle model is reliable and predicts credible bundle
behavior in response to accident type conditions in a fuel channel. Note,
however, the ability of the present model to predict accident behavior is limited
by lack of knowledge of many of the sub-phenomena; two examples are fuel
column rigidity and effect of endcap welds on end plate rigidity. The intended
application of this model is to assess pressure tube integrity in large break loss of
coolant accidents. An evaluation of the completeness of the models and how the
application should be performed is the next step towards this goal.

62
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

A rational seismic design approach for

reinforced concrete walls for nuclear
power plants (5-1943)

Carlos Coronado. Ph.D.1, Sanjeev R. Malushte, Ph.D., S.E.2,

Javeed Munshi, Ph.D., S.E.3
1
Civil Engineer, Bechtel Power Corporation, USA
e-mail: [email protected]
2
Sr. Principal Engineer, Bechtel Power Corporation, USA
e-mail: [email protected]
3
Principal Engineer, Bechtel Power Corporation, USA
e-mail: [email protected]

Keywords: shear wall, finite element, soil structure interaction, seismic, shell
element

Most nuclear power plant buildings are box-type structures consisting of several
interconnected shear walls. In recent practice, it is common to use refined finite
element models involving shell and plate elements for structural analysis of this
type of structures to accurately capture the response of walls and slabs to seismic
and other loads. Results from these analyses, in the form of element forces and
moments are used by the structural engineer during the design process. In the
United States such design is typically conducted according to the requirements
of ACI-349 provisions, which are mostly derived from ACI-318 building code.
ACI-318 provisions for shear walls are based on experimental results obtained
from individual wall specimens tested up to global failure for in-plane forces and
moments acting on the wall panel. The ACI design equations (and the tests
backing these equations) do not directly address the wall adequacy for
simultaneous in-plane and out-of-plane forces/moments; the assumption being
that the two sets of demands are likely out of phase (which is often true for
earthquake excitation along a single axis). Results from 3D seismic analyses
involving simultaneous multi-directional excitation (using 100-40-40 or SRSS
combination rule) however often show that a wall may be subject to significant
in-plane and out-of-plane demands simultaneously. In the absence of clear rules,
the designer is however left to perform independent checks for the in-plane and
out-of-plane demands, which disregards the effect on wall adequacy under such
simultaneous demand conditions.
Full sectional resultants (global) are required when conducting the design of
shear walls according to ACI-349 code. Nevertheless such resultants are not
readily available from the finite element models used for typical nuclear
structures; they must be calculated using section cuts by numerical integration of

63
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

individual elements results. Some mainstream structural analysis programs

provide postprocessing capabilities for these purposes. Nevertheless, such
options are not available in soil structure interaction programs widely used in
nuclear applications such as SASSI2000. Therefore custom postprocessing tools
must be developed by the structural engineer for such purpose; or alternatively
the shear wall design can be directly conducted using the results from individual
finite elements. In this paper shear wall designs conducted using global and
element based approaches are discussed and compared against experimental and
numerical results. In particular, experimentally tested shear walls available in the
literature are redesigned according to the aforementioned methodologies to resist
the experimentally observed ultimate load. It is shown that both global and
element based methodologies result in wall designs meeting ACI-349 code
requirements. Nevertheless, their outcome is different in terms of reinforcement
ratios and distributions. Therefore different performance levels can be expected
for each design approach. Nonlinear finite element simulations considering
concrete cracking and reinforcement yielding are conducted in order to
investigate such performance. Wall designs are evaluated in terms of drift and
cracks distribution at different loading levels. Conclusions and recommendations
are drawn from this evaluation and used to guide the design of walls in box-type
nuclear structures. The practical implementation of these recommendations is
presented for a typical wall in the Ultimate Heat Sink (UHS) building of a
generic nuclear power plant. The design seismic loads are obtained from soil
structure interaction (SSI) analyses conducted for the design-basis event or Safe-
Shutdown Earthquake (SSE). In addition, sensitivity analyses are carried out in
order to account for uncertainties in the soil properties. The results of the seismic
SSI analysis are summarized for design purposes in three different forms: first as
static equivalent seismic loads based on average peak structural accelerations;
second as absolute peak element force and moment resultants; and third in the
form of time histories of element resultants. The use of each set of results for
design purposes is addressed and discussed. In particular the applicability of
ACI-349 provisions is discussed pointing the advantages and limitations of using
a particular set of results at the design level.

64
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Response and failure criteria of large

cylindrical vessels to rapid pressurization
in CANDU severe accidents (5-1945)

David L. Luxat1, John C. Luxat2

1
AMEC NSS, Toronto, Ontario
2
McMaster University, Hamilton, Ontario

Severe accidents in CANDU reactors are typically categorized into two distinct
classes of accident progression. One class of accidents involves a slow
progression to core damage and the second, lower likelihood class involves rapid
progression to core damage. Due to the positive void reactivity in operating
CANDU reactors, postulated accidents do exist in which engineered safeguards
(i.e. reactor shutdown) are required to mitigate a reactivity transient that is either due
to or a consequence of some initiating events. Depending upon the effectiveness of
reactor shutdown, the subsequent progression to severe accident core damage is
characterized by either very rapid core degradation (loss of shutdown) or a slow
degradation of the core resulting from a power-cooling mismatch following loss
of a significant number of mitigating emergency heat sinks.
Since CANDU reactors have two independent and dedicated shutdown
systems, Level 1 Probabilistic Risk Assessments (PRA) indicate that severe
accidents involving a loss of shutdown are very low frequency and are not
dominant contributors to the total core damage frequency. However, due to the
rapid core degradation that occurs following an unmitigated reactivity transient,
this accident class has the potential for early releases and must be considered
outside of the scope of PRA by undertaking consequence analysis. Understanding
the response of the reactor system to large and rapid energy releases, and the
resulting dynamic loads imposed on the large calandria vessel, is critical to
assessing the nature of the initial consequences and the potential for subsequent
mitigating actions following core degradation.
Loss of shutdown accidents are characterized by the near-simultaneous rupture of
a large number of fuel channels. The discharge of coolant from ruptured
channels with ensuing rapid pressurization of the calandria vessel is sufficiently
large to be beyond the capacity of the pressure relief system for the vessel.
This paper presents a detailed finite-element model for the dynamic structural
response of the calandria vessel following fuel channel failure. This detailed
model is used to assess:
• The dynamic response of the calandria vessel to impulse loading
transients typical of two-phase coolant discharge arising from multiple,
near-simultaneous fuel channel failures,

65
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

• The impulse loading threshold to induce calandria vessel failure,

• The location of calandria vessel failure, and
• Potential pathways for subsequent fluid discharge following calandria
vessel failure.
In order to evaluate the dynamic response of the vessel, a separate hydrodynamic
model is developed that calculates the pressure propagation in the vessel
originating from failed fuel channels, including the effect of collapse of
calandria tubes onto their pressure tubes due to overpressure. This model
provides the transient hydrodynamic pressure loading on the vessel wall. Short
term interaction between the hydrodynamic transient and displacement of the
vessel walls is taken into account using an elastic representation for the vessel.
The finite element model is subjected to the calculated loading and dynamic
estimates of strain are generated. In order to account for uncertainties associated
with mechanical feedback from the vessel on the hydrodynamic loading
transient, a series of parametric runs are performed to evaluate the sensitivity of
this coupling. Results show that there is relatively low sensitivity arising from
this coupling due to the fact that:
• The initial pressure wave acts as the dominant impulse loading of the
vessel wall,
• Subsequent reflected pressure waves are significantly attenuated by the
steam bubble formed by the ruptured channels, and
• The transient displacement of the vessel wall is governed by the
dominant structural modes of the vessel.
High stresses are generated at the weld discontinuity at either end of the main
vessel shell where it connects with a smaller diameter sub-shell extension. The
stresses at the welds exceed ultimate tensile strength and the vessel fails by a tearing
rupture around the circumference. A critical number of near-simultaneously failed
channels are required for the vessel failure to occur.
Additional sensitivity analysis is performed to quantify the uncertainty in both
the likelihood and extent of vessel failure. The extent of vessel failure is an
important factor which governs the rate of moderator fluid displacement out of
the vessel and the subsequent pressurization of containment.

66
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Update of ASCE standard “Seismic analysis

of safety-related nuclear structures and
commentary” (5-1947)

Short, Steve1, Orhan Gurbuz2, Mike Salmon3

1
Simpson, Gumpertz and Heger, 4000 MacArthur Blvd., 7th Floor, Suite 710,
Newport Beach, California, 92660, USA, e-mail: [email protected]
2
Bechtel Corp., 16162 Tortola Circle, Huntington Beach, CA 92649
e-mail: [email protected]
3
LANS, LLC., Los Alamos, New Mexico 87545, e-mail: [email protected]

The American Society of Civil Engineers Standard 4 (ASCE 4) has been the main
guidance document for the seismic analysis of the nuclear safety-related facilities and
other critical or important facilities in USA for more than two decades. The Standard
was last revised in 1998. This standard was developed mainly for U.S. Department of
Energy non-reactor nuclear facilities. A working group of the ASCE Dynamic
Analysis of Nuclear Structures Standards Committee undertook a task in 2005 to
update the standard to implement recent developments in seismic analysis of these
facilities. Recently, a resurgence of the nuclear power industry has made it even more
important to have up-to-date seismic provisions.
ASCE 4 is a companion document to ASCE Standard- 43, “Seismic Design
Criteria for Structures, Systems and Components in Nuclear Facilities.” ASCE
Standard-43 has been used as a reference document by several utilities in
submittal of combined license applications (COLS) for New Reactors.
The working group responsible for this revision is part of the ASCE Dynamic
Analysis of Nuclear Facilities Committee. Its members are drawn from the
industry, academia and governmental organizations, thus encompassing the
potential users of the standard in different groups. It is expected that the revised
document will be available in 2009.
The paper summarizes the changes being made to the document. The entire
standard has undergone a major editorial change in order to make it more user-
friendly. The standard now is composed of 7 Sections, appendices, and related
commentary. Section 1 was expanded to highlight the “target performance goal”
approach adopted both in ASCE 4 and its sister standard, ASCE 43.
Performance based seismic design criteria have been implemented for
Department of Energy Facilities for many years. Only recently, has the nuclear
power industry employed such criteria. The ASCE 4 provisions for determining
seismic demand include sufficient conservatism that when combined with the
seismic design provisions in ASCE 43, the probabilistic target performance
goals are achieved. The target performance goals are expressed as annual

67
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

frequency of exceeding unacceptable behavior for structures, systems, and

components being designed. . Note that these criteria provide many levels of
criteria in a graded approach that can be implemented based on the hazards and
importance of the facilities. As a result, these criteria are especially valuable for
nuclear facilities such as fuel processing facilities where the hazards are
significant but much less than those associated with nuclear power plants. Such
criteria have not existed in the past and these facilities have not been designed in
a consistent manner.
Section 2 brings in the latest developments in seismic ground motions and the
various input ground motion definitions which have come to use in recent years.
Guidelines for utilizing the output from a probabilistic seismic hazard analysis to
establish the input to a seismic soil-structure interaction analysis are presented. Section
3 discusses modeling for seismic analyses, providing guidance on issues such as mesh
sizes, material properties, stiffness, damping and requirements for special structures.
Section 4 describes acceptable methods of analyses that range from simple
equivalent static approach to complex nonlinear dynamic analysis. This section
has been expanded to include pushover analysis and frequency domain method.
Section 5 provides a significant update on soil-structure interaction analysis with
emphasis on sub-structuring method and three-dimensional response
calculations. Section 6 continues with the sub-system analysis, offering guidance
on input motions, in-structure response spectra, in-structure time history
motions, and coupled system analysis.
Section 7 is devoted to analysis of special structures, including buried structures,
earth-retaining walls, above-ground vertical tanks, distribution systems, sliding
and rocking calculations and base isolation. Seismic design provisions for
distribution systems and for unanchored components provided in this Standard
are strongly needed by the nuclear industry. In addition, the subject of dynamic
soil pressures on underground walls has been significantly expanded to consider
responses from soil-structure interaction analyses.
An appendix addresses the evaluations beyond design basis and presents
available methods for such evaluations. Both the probabilistic seismic risk and
seismic margin assessments are discussed and compared.
In summary, the revised standard will reflect the state-of-the-art approaches for
determining seismic demands on nuclear safety-related structures, systems and
components. This document is expected to become the definitive source for seismic
analysis in the nuclear industry and resolve many outstanding controversial issues.

References
1. DOE-STD-1020. Natural Phenomena Hazards Design and Evaluation Criteria for
Department of Energy Facilities.

2. ASCE 43-05. Seismic Design Criteria for Structures, Systems, and Components in
Nuclear Facilities.

68
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Investigation of building structures

response to heavy item drop (5-1951)

Renatas Karalevicius, Gintautas Dundulis, Sigitas Rimkevicius

Laboratory of Nuclear Installation Safety, Lithuanian Energy Institute
3 Breskaujos str. LT-44403 Kaunas, Lithuania, e-mail: [email protected]

Introduction
The Ignalina Nuclear Power Plant (NPP) has two RBMK-1500 graphite
moderated boiling water multi-channel reactors. The Ignalina NPP Unit 1 was
shutdown at the end of 2004 while Unit 2 is foreseen to be shutdown at the end
of 2009. The projects of dismantling and decontamination of equipment of Unit 1
were initiated. One of important parts of such projects is analysis and assessment
of all hazards that are possible during dismantling and decontamination activities.
The equipments installed in NPP are heavy. The dropping of heavy dismantled
item can destroy dismantling building and adjacent buildings and equipments.
Therefore the drop of heave parts of items should be evaluated during
decontamination. The methodology of the evaluation of consequences to
building structures of heavy item drop is presented in this paper. The structural
integrity analysis of the building structures in case dropping heavy part was
carried out. The finite element method was used in this analysis.

Analysis
This paper present the structural integrity analysis of building structures in case
drop of heavy dismantled item during dismantling and decontamination
activities in the of Ignalina NPP Unit 1. The drop of heavy item potentially can
damage building structure and as result can cause collapse of building. The
maximal hazard, which could be caused by the load drop during dismantling
activities in the building of Ignalina NPP, is drop of the cut ring of Emergency
Core Cooling System (ECCS) pressure vessels. Potential damage of the building
slab and adjacent structures due to drop of a cut ring of ECCS pressure vessel on
to this slab has been assessed in this paper. The maximum possible drop height
(16 m) was chosen in order to evaluate the loading to the slab at level 0.0 m of the
building. Finite element model for the cut ring of ECCS pressure vessel drop force
calculation was developed. The state-of-the art computer code ABAQUS/Explicit
was used for load analysis. The maximum possible drop forces were calculated
from this analysis. This maximum force was used for the static analysis of
structural integrity of this compartment using finite element software
ABAQUS/Standart.

69
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Summary and conclusions

According to analysis results it is possible to conclude that the impacted
reinforced concrete slab at level 0.00 m and supporting columns of this
compartment will experience cracking of concrete, but the structural integrity of
these slab and columns will be maintained during impact of a cut ECCS ring.
The analysis results show that the structural integrity of the building will be
maintained, it will not collapse and it will be capable of performing its intended
function.

70
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

LS-DYNA impact analyses of nuclear power

plant structures for tornado missile
risk analysis (5-1961)

Robert T. Bocchieri1, Lawrence A. Twisdale2, Claudia Navarro-Northrup3,

Gary P. Gauthier4, Steven W. Kirkpatrick5
1
Applied Research Associates, Inc.
2672 Bayshore Parkway, Suite 1035, Mountain View, CA 94043, USA
e-mail: [email protected]
2
Applied Research Associates, Inc.
8537 Six Forks Road, Suite 6000, Raleigh, N.C. 27615-2963, USA
e-mail: [email protected]
3
Applied Research Associates, Inc.
2672 Bayshore Parkway, Suite 1035, Mountain View, CA 94043, USA
e-mail: [email protected]
4
Gary P. Gauthier (while at Applied Research Associates, Inc.)
California Department of Water Resources, Division of Safety of Dams
2200 X Street, Sacramento, CA 94236, USA
e-mail: [email protected]
5
Applied Research Associates, Inc.
2672 Bayshore Parkway, Suite 1035, Mountain View, CA 94043, USA
e-mail: [email protected]

The TORMIS methodology was developed to estimate the probability of damage

to nuclear power plant structures and components from debris missile impacts in
extreme winds. A critical component of performing this risk analysis is the
assessment of the damage characteristics to the power plant structures from
various missiles and the critical velocity at which this damage results. For some
target/missile combinations, analytical methods have been historically used to
predict damage. However, for more complex structures and missiles, either
testing or more detailed analyses are required.
This paper details LS-DYNA finite element analyses that were conducted to
determine the critical speeds for various missile/target combinations. LS-DYNA
is a nonlinear explicit finite element code for the dynamic analysis of structures,
and is particularly well-suited for impact and penetration analysis. Targets in this
study included various parts of Emergency Diesel Generator (EDG) Exhaust
Vents, EDG Fuel Oil Day Tank Vents and Steel Floor Plates. Critical speeds for
vent structures were defined by the degree of closure of the vent cross section,
which restricts the flow of exhaust gases. To account for uncertainties in the as-
fabricated target strength, two welded joints were considered; a ‘strong weld’
that has the same strength as the baseline material and a ‘weak weld’ that fails in

71
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

shear at the design allowable strength. These two cases bound the actual strength
of the joint and were necessary due to the lack of sufficient data on the welded
joint characteristics.
In these analyses, both the missiles and targets were modeled explicitly to
correctly model their interaction and the resulting impact damage. A variety of
missile types, including metal pipes, concrete pavers, wood beams, steel grates,
metal siding, plywood panels, and storage bins were considered. Impact analyses
were conducted to determine the impact response for a large variety of missile
impact orientations. Results from these analyses were used to determine the
critical orientations and speeds for each missile against each type of target. Over
three hundred separate analyses were conducted in various missile orientations
at several speeds.
Finally, the feasibility of scaling the critical speeds for the various missiles by
their initial kinetic energy was examined. This approach had mixed success
because of the large differences in stiffness and strength of the missile types as
well as changes in critical orientation. In general, however, missiles of similar
size and stiffness to the targets took the least energy. Missiles of similar size but
softer and weaker required higher energies and the softest and weakest missiles
required the greatest energy. A more general scaling rule would require
consideration of the missile strength and geometry characteristics.

72
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Thermo-mechanical analysis of a helium

cooled divertor of a fusion reactor (5-1963)

Igor Simonovski1, Boštjan Končar2, Leon Cizelj3

1
Jožef Stefan Institute, Reactor engineering division
Jamova cesta 39, SI-1000 Ljubljana, Slovenia
e-mail: [email protected]
2
Jožef Stefan Institute, Reactor engineering division
Jamova cesta 39, SI-1000 Ljubljana, Slovenia
e-mail: [email protected]
3
Jožef Stefan Institute, Reactor engineering division
Jamova cesta 39, SI-1000 Ljubljana, Slovenia
e-mail: [email protected]

The helium cooled divertor based on modular design concept is envisaged within
the framework of a fusion power plant conceptual study. Several modular design
variants have been developed and analyzed in the past few years at the
Forschungszentrum Karlsruhe (FZK) [1, 2]. An advanced modular divertor
concept with multiple helium cooling jets (HEMJ) was found to be the most
appropriate for several reasons. Helium cooled divertor should master high heat
flux of more than 10 MW/m2 to ensure that the material constraints are not
exceeded. The main design requirements of the helium cooled divertor are to
increase the heat removal capability of the divertor and to minimize the pumping
power for the coolant. In the presented work thermal and stress loadings on the
divertor components cooled by multiple helium jets are investigated numerically
taking into account different boundary conditions. Two different thermal
loadings on the plasma-facing side of the divertor are analysed: a) stationary
loading with a constant heat flux and b) cyclic heat-flux loading. Thermal and
structural analyses are carried out with the code ABAQUS. Thermal boundary
conditions at the fluid-facing side of the divertor are obtained from the CFD
solutions of the turbulent helium flow [3]. The experiments at high heat flux and
high mass flow rate, performed at EFREMOV Institute in Sankt Petersburg are
simulated in our study. Results of thermal and stress loadings in the divertor
structures are provided.

References
1. T. Chehtov, A. Gervash, R. Giniyatulin, T. Ihli, R. Kruessmann, V. Kuznetsov, A.
Makhankov, I. Mazul, P. Norajitra, I. Ovchinnikov, J. Weggen, B. Zeep. Status of
He-cooled divertor development (PPCS Subtask TW4-TRP-001-D2), FZKA
7100. Technical report, Forschungszentrum Karlsruhe – Wissenschaftliche Berichte
(accessible through http://bibliothek.fzk.de/zb/berichte/FZKA7100.pdf), 2005.

73
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

2. L.V. Boccaccini, T. Chehtov, R. Giniyatulin, S. Gordeev, T. Ihli, G. Janeschitz, A.O.

Komarov, W. Krauss, R. Kruessmann, V. Kuznetsov, R. Lindau, P. Norajitra, I.
Ovchinnikov, V. Piotter, M. Rieth, R. Ruprecht, V. Slobodtchouk. Conceptual
design of a He-cooled divertor with integrated flow and heat transfer promoters.
(PPCS subtask TW3-TRP-001-D2). Part I: Summary. FZK 6974. Technical
report, Forschungszentrum Karlsruhe -Wissenschaftliche Berichte (accessible
through http://bibliothek.fzk.de/zb/berichte/FZKA6974.pdf), 2004.

3. B. Končar, M. Draksler, P. Norajitra, V.Widak. Numerical Investigation of Multiple-Jet

Cooling Concept for Helium Cooled Divertor. In Proceedings of the International
Conference Nuclear Energy for New Europe 2008, Portorož, Slovenia,
September 8–11, 2008, September 2008.

74
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

A failure mode evaluation of a 480V MCC

in nuclear power plants at the seismic
events (5-1970)
Min Kyu Kim1, In Kil Choi2
1
Senior Researcher, Korea Atomic Energy Research Institute
1045 Daedeokdaero, YuseongGu, Daejeon, Korea
e-mail: [email protected]
2
Principal Researcher, Korea Atomic Energy Research Institute
1045 Daedeokdaero, YuseongGu, Daejeon, Korea
e-mail: [email protected]

A 480V MCC Cabinet is one of major equipment system in Nuclear Power Plant. For
the shaking table test, a real MCC cabinet was rented from the manufacturing
company. For the evaluation of a failure mode for Motor Control Centers (MCCs), a
shaking table test was performed. For the shaking table test, two kinds of seismic input
motions were used. One is an artificial seismic input motion based on the NRC Reg.
guide 1.60 design spectrum and the other is also an artificial seismic motion based on
the Korean Nuclear Power Plant site specific Uniform Hazard Spectrum (UHS). The
UHS motion was selected for an evaluation of a High frequency effect of the electric
equipment in a NPP. PGA levels for shaking table test were scheduled by 0.2 g to
5.0 g but the test was stopped at about the 2.5 g level because of the chattering of the
relay systems. The shaking table tests were performed with a one dimensional shaking
which was a front to back direction (horizontal) and a vertical direction.
Functional and structural failure modes were also evaluated by this shaking table
test. For the evaluation of a relay chattering, the electric signal was measured at several
points. There are two kinds of ground fault relays and thermal relays are installed in
the MCC. It is impossible to measure the electric signal of all relays, only some of the
relays and electric equipments were considered. In the case of the US NRC spectrum,
480 V AC power was supplied so it can measure the signals of the equipments related
to a power system like a power transformer. But in case of UHS spectrum, 480 V AC
power wasn’t supplied because of the safety of experiment. Therefore, only the signals
from the relays were measured. For the measurement of a relay system, an arbitrary
input power was supplied to the MCC. Also, for the evaluation of structural failure
modes, in-cabinet responses and response amplifications of MCC, acceleration
responses were measured at major points of cabinet.
Through this test, several kinds of functional failure modes can be found and
the chattering effect of several relays in the MCCs can be certified. As a result, it
can be recognized that the 480 V MCC has a sufficient seismicity as a SSE level
earthquake. But in the case of a higher level earthquake motion, a chattering
happened for both seismic motions, moreover both a horizontal and vertical
shaking cause a relay chattering.

75
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Seismic FEM analysis of reinforced

concrete structure in SMART-2008
project (5-1974)

Jukka Kähkönen, Pentti Varpasuo

Fortum Nuclear Services Ltd, P.O. Box 100, 00048 FORTUM, Finland
e-mails: [email protected], [email protected]

In order to asses the seismic tri-dimensional effects and non-linear response of

reinforced concrete buildings, a reduced scaled model (scale of 1/4th) of a
nuclear reinforced concrete building was tested during year 2008 on AZALEE
shaking table at Comissariat à l’Energie Atomique (CEA Saclay, France). This
test, supported by CEA and Electricité de France (EDF) is part of the
international “SMART-2008” project.
The first part of the project was a blind prediction of structure behavior
subjected to the ground motion levels from 0.1 g up to 1 g. As continuation to
the first part of the project participants were given some test results so that the
participants could adjust their analysis models. This conference proceeding
presents the approach and the methodology adopted by the Fortum Nuclear
Services Ltd research team to adjust a finite element method (FEM) model to
match the test results.
The main difficulty in the analysis was the accounting of concrete surface
cracking in relatively low value of the ground excitation. This difficulty was
overcome by the calibration of the brittle cracking concrete model parameters so
that obtained displacement results corresponded better to the expected displacement
result values. Also the damping played major role in the adjustments procedure.

References
ABAQUS version 6–7.1. 2007. Users documentation, Simulia Inc., Dassault Systèmes.

Comissariat à l’Energie Atomique (CEA). 2008. SMART-2008 project documents, Saclay,

France.

Eurocode 8. 2003. Design of structures for earthquake resistance. Part 1. General rules,
seismic actions and rules for buildings, prEN 1998-1, European Committee for
Standardization.

RCCG. 1988. Regales de conception et de construction du genie civil des pilot’s nucleotides.
REP, Editions AFCEN, France.

Yung-Yen Ko, Shan-Yi Hsu, Cheng-Hsing Chen. 2008. Analysis for seismic response of dry
storage facility for spent fuel, Nuclear Engineering and Design 239 158–168, Elsevier.

76
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Salient aspects of analysis and design of

large integrated safety related
structures (5-1973)

K.V. Subramanian, S.M. Palekar, H.A. Mapari, R. Balaji

TCE Consulting Engineers Ltd
243 Matulya Center A, Senapathi Bapat Marg, Mumbai, India
e-mail: [email protected]

Introduction
Nuclear Power Plants often involve buildings which are integrated into a single large
structure due to the functional requirements. Rigorous structural analysis and design
that has to be performed for such large safety related structures, under various
environmental conditions including static, seismic and transient conditions, poses
certain typical issues to be addressed / resolved as enlisted below,
1. Reducing time and cost of the analysis and design performed.
2. Handling certain typical design conditions / loads, and
3. Re-evaluating/re-looking at the different analysis and design parameters
and analysis methods adopted so as to reduce some of the conservatisms
involved and move towards better solutions.
This paper takes cognizance of the experiences gained from engineering
performed for a Proto-type Fast Breeder Reactor (PFBR) under construction at
Kalpakkam in INDIA. It comprises of a single large integrated building, Nuclear
Island Connected Building (NICB) (93 m × 83 m × 74 m) integrating eight
buildings, viz, Reactor Containment Building, Steam Generator Buildings, Fuel
Building, Rad waste building, Electrical and Control Buildings.
Critical aspects of analysis and design have been reviewed in this paper to
identify aspects for possible improvements with suggestions.

Critical areas addressed

1. Soil-structure interaction (SSI) plays a significant role in affecting the
performance of such structures, in which uncertainties are involved in
• Assessing the Stiffness of soil
• Assessing damping ratio of soil
• Idealizing soil in the FE model
• Distribution of soil stiffness.

77
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Results of analysis performed with soil stiffness determined using Vesic’s

equation and distributed uniformly, and, those obtained based on analysis
with stiffness variably distributed based on pressure acting (using
Boussineq’s equations) have been compared. While Boussineq’s formulation
entails extensive computation time, it will be prudent to assess the advantage
of Boussineq’s formulation with respect to Vesic’s formulation from
practical design considerations, under the condition of rocky strata.
Uncertainty due to possible variations in measured value of soil stiffness
has been taken care of by performing a range analysis adopting Soil Stiffness
as ½ K, K and 2 K, findings of which shows variation analysis need not be
performed in case of a rocky strata as the effects are in-significant, details of
which have been elaborated in the paper.
Soil damping evaluated based on provisions of ASCE code, with values
limited to 7% for rocking / torsional motion, 20% for horizontal motion and
30% for vertical motion have been considered. Effect of usage of a uniform
damping of 7% in all directions has been studied and found significantly
conservative. Possible ways of adopting a composite soil damping in future
have been brought out in the paper.
Effect of embedment of structure has been studied and found in-significant
for the type of structure and strata considered.
2. Slabs in such large structures play a significant role in distributing the
horizontal shear forces between different vertical elements and integrating
them together, there-by altering the structural behavior. They are to be
designed adequately to carry the in-plane forces, which will be especially
significant under seismic and thermal loads.
With the available computational capacities and to take care of certain
other aspects, it was sought to go for modeling the slabs purely as diaphragm
elements, in NICB. But, with the current improvements in computational
capacities, it could be possible to go for modeling slabs using a relatively
coarser mesh of Thick Shell elements, instead of Thin Shell diaphragm
elements. The effect of harnessing the out-of-plane bending stiffness (of the
portion of slab acting along with beams) and corresponding economy in
beam designs can be ascertained.
3. Efforts to reduce the size of the model need attention in eliminating the un-
warranted computational costs in such large structures. In NICB, a) all slabs
have been modeled using relatively coarser diaphragm elements, b) items /
equipments that could be de-coupled have been identified and decoupled and
c) stick models for vessels and equipments have been consciously used and
coupled with the global model where possible, and d) refined individual sub-
structure analysis has been performed for certain components which require a
refined analysis, as measures towards reducing the un-warranted size of the
model. Further efforts could be a) towards reducing the size of dynamic
analysis problem by using super-elements and dynamic condensation techniques,

78
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

b) Performing convergence analyses and determining appropriate elements

sizes that could be adopted for different regions based on load variation,
thickness etc, and c) study on effects of usage of certain lower-order elements
for shells and beams in possible areas.
4. Load due to Initial drying Shrinkage of concrete plays a significant role in
such large structures, due to build-up of stresses over large dimensions of
slabs and walls, if applied in the entire FE model of the structure as a single
load step. Efforts to reduce their effects could be towards application of
drying shrinkage loads in the form of construction stage loads and evolving
methods to estimate material properties of concrete under shrinkage and
thermal loads etc.

Conclusion
This paper identifies the critical areas of concern that are to be addressed while
performing structural analysis and design of large structures like NICB.
Principles that had been adopted in the present project to address the issues, and
possible scope and alternatives available for improvement on the process of
analysis and design have been brought out in this paper.

79
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Indian PHWR pre-stressed concrete

containment performance evaluation
with BARCOM and round robin
analysis program (5-1977)

R.K. Singh1, A.K. Ghosh, H.S. Kushwaha

Reactor Safety Division, Health Safety and Environment Group
Bhabha Atomic Research Centre, Trombay, Mumbai 400 085, India
Fax: +91-22-25505151, +91-22-25519613
1
Corresponding author e-mails: [email protected],
[email protected]

The BARC Containment (BARCOM) test model is a 1:4 scale representation of

Tarapur station 540 MWe Pressurized Heavy Water Reactor (PHWR) pre-
stressed concrete inner containment structure. This test model has been
constructed at BARC Containment Test Facility-Tarapur for the functional and
structural performance evaluation of Indian containments. Bhabha Atomic
Research Centre (BARC), Trombay has coordinated an international round robin
analysis program to carry out the ultimate load capacity assessment of
BARCOM with participants from research, academic and industrial organizations.
This is the largest containment model which has been built so far in the world
and would be shortly tested up to its ultimate load capacity and the experimental
results would be analyzed with the inelastic numerical code predictions obtained
from around 15 registered participants in this round robin exercise during the
pre-test and post-test phases.
Indian nuclear power plants have double containment with outer and inner
containment structures of reinforced and pre-stressed concrete construction
respectively. The nuclear containment system of Indian PHWRs is defined as the
primary Inner Containment Wall (ICW) and the secondary Outer Containment
Wall (OCW) enclosures as well as the systems and components provided to
establish an essentially leak tight barrier against the uncontrolled release of
radioactivity to the environment. It is desirable to assure that the integral double
containment system meets the functional requirements, which are important for
safety under various postulated design basis accident (DBA) conditions like
Loss of Coolant Accident (LOCA) or the Main Steam Line Break (MSLB) due
to the largest break in the primary system. Performance of nuclear containment
with improvement in engineered safety features and demonstration of safe
technology for public acceptance of nuclear power program have been backed
up with containment safety research at BARC, Trombay. Indian PHWRs of
220/540 MWe designs have inherent safety features with large containment

80
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

volume, low volumetric power density, low temperature heat sink, low excess
reactivity and long prompt neutron life time. Hence the global threat to the
containment is of lesser severity for the extremely low probable postulated
beyond the design basis severe accidents compared to the PWR and BWR plants
where such scenarios have been postulated and widely studied. However, with a
view to demonstrate the adequacy of this ultimate barrier of the double
containment system and to improve the existing design of the PHWR
containments, structural safety assessment for both the design basis and beyond
the design basis load conditions has been a thrust research area for Indian
PHWR power program.
The objective of the present test program is to obtain the pressure,
displacement and strain data related to the various failure modes of BARCOM.
Following observations would be recorded that would aid in understanding the
functional and structural behavior of the containment model up to the ultimate
pressure.
• First appearance of concrete cracking
• First through thickness cracking
• First yielding of the reinforcement
• First significant loss of leak tightness
• Maximum pressure sustained by the test model before significant leakage
• Ultimate pressure sustained by the containment test model.
In order to predict the ultimate load, elaborate tests have been planned on soil,
concrete, reinforcement and pre-stressing tendons as detailed below
• Soil investigation of the site to determine the geo-technical parameters
(bearing capacity, Poisson’s ratio, bulk modulus and shear wave velocity etc.)
• Compressive and tensile strength of the proposed concrete trial and pour
mixes for the construction of the test model and the associated standard
deviations
• Fracture energy characteristics of plain and reinforced concrete on beam
specimens
• Tensile strength of reinforcement bars to be used for the construction of the
model
• Tensile strength and relaxation characteristics of the pre-stressing strands
• Wobble and friction coefficient of sheath and pre-stressing system assembly.
More than 1500 surface and embedded sensors have been installed on BARCOM
which include vibratory wire embedded and spot-welded strain gauges, electrical
resistance strain gauges, dial gauges, earth pressure cells, tilt meters and camera
systems for structural response, crack monitoring and fracture parameter

81
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

measurement to evaluate the local and global behavior of the containment test
model. The leakage rate measurement shall help to establish the degree of
imperviousness at different milestone pressures reached by the test model and
validate the available analytical and semi-empirical models for the prediction of
the leakage rate with respect to the pressure and crack width data.
All the functional and structural failure modes of BARCOM shall be studied.
Benchmarking of the in-house/commercial non-linear finite element codes and
inelastic analysis procedures for the postulated design basis and extremely low
probable beyond the design basis severe accidents shall be carried out. This
study would give an opportunity to address some of the structural code design
and safety issues relevant for pre-stressed unlined containment structures.
The present paper describes the features of BARCOM and the activities of
design, analysis, construction, engineering and instrumentation shall be
presented for the mega size experiment that has been planned under this Indian
containment research project.

82
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

The FSI seismic analysis for FBR core

assemblies (5-1983)

Yafei MO1, Daniel PROC2, Jing WEN3

1
Master, Research Center of CEFR, China Institute of Atomic Energy, Beijing
e-mail: [email protected]
2
Doctor, Seismic Mechanic Study Laboratory of CEA, Center of Saclay, France
3
Doctor, Research Center of CEFR, China Institute of Atomic Energy, Beijing

Introduction
In this paper, A FEM code CASTEM, is used in the analysis. The dynamic
responses of one row assemblies under the Horizontal Seismic load are
researched. Comparing the results with other similar analyses and the experiments,
the model and the method is reasonable and effective. On this basis, considering
the Fluid Structure Interaction, the dynamic responses of one row and three row
2D assemblies are analyzed. One single horizontal seismic analysis of 3 row
assemblies shows that the interaction between neighboring rows has very small
effect on each other. The estimate response characteristics of 3 row models or
even the whole core under the single horizontal excitation can be replaced by
one center row model.

Analysis model and method

Some pre-analyses on one single assembly are finished before the analyses on
multi-assembly. Based on that one center row of assemblies simulated by
variable cross section beam elements are selected to be performed one single
horizontal seismic analysis. The model is shown as Fig. 1. The shock between
the neighboring assemblies is simulated by combination of one gap element. The
acceleration-time histories are taken as seismic input data. The modal basis
method is adopted. The structural Rayleigh damping ratios and shock stiffnesses
between the neighboring assemblies are assumed. Comparing the results with
other similar analyses and the experiments, the model and the method is
reasonable and effective. On this basis, considering the Fluid Structure
Interaction, the dynamic responses of 2D assemblies are analyzed. The modal is
showed in Fig. 2. For realizing the effect of interaction between the neighboring
rows, 3 neighboring center rows of assemblies are selected to carry out
horizontal seismic analysis. The seismic responses have been compared with
those for one row model.

83
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Figure 1. One single horizontal seismic analysis model of one row of assemblies.

Figure 2. The analysis modal of 2D one row assemblies.

Summary and conclusions

The analysis modal and method of CEFR core assemblies are researched,
considering the fluid structure interaction and some assumed parameters, the
dynamic analysis of one row and three rows assemblies under the horizontal
seismic excitation are finished. We can draw a conclusion as following:

84
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

a, Comparing the results of analyses with other similar analyses and the
experiments, the model and the method is reasonable and effective.
b, The different type elements are adopted to analyze, owing to the
reference coefficient are not same, and the beam element result is
bigger than the 2D shell element.
c, Comparing the results of model consisting of three rows of assemblies
with that of one row, effect between neighbor rows is very small.
Roughly speaking, we can estimate conservatively the response of
three-row model or even the whole core under one single horizontal
excitation by using one center row model.
d, The analysis of 2D structures fluid structure interaction is finished, and
the displacement and impact force are calculated, which will offer
important indemnification for the farther analysis of all core 3D
assemblies.

Reference
1. Jing Wen, Daogang Lu. Single assembly preliminary analysis for horizontal seismic
analysis on FBR core. Atomic Energy Science And Technology, Vol. 41, No. 2,
March 2007, pp. 148–152.

85
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

The effects of seismic spectrum on

seismic analysis (5-1988)

Tian Jinmei, Liu Shubin

China Nuclear Power Engineering Co., Ltd. Beijing, China
e-mail: [email protected]

In this paper, the basic principle of seismic analysis is explained and the factors
which affect the seismic response are analyzed. In the seismic computation of
regenerative heat exchanger of steam generator blowdown system of Lingao II
nuclear power plant, one problem occurs, which is there’s only a little difference
between the seismic spectrum of the same component, but there is great
difference between the seismic response of the same component in different
plants, so the effects of the peak change of seismic spectrum and the magnitude
of the seismic spectrum on results of seismic analysis are analyzed.
The results show that if the first few frequencies of the component are close to
the inherent frequency of the plant, then the modal response of this mode is
magnified remarkably, and these modal responses will contribute most to the
whole structure response. Also, as to the same component, the principle of the
seismic spectrum value change is identical to that of the results of the seismic
responses. So this paper can be served as a reference of analyzing and solving
those problems concerned with seismic analysis.

References
1. Sun Jian-gang, Hao Jin-feng, Wang Zhen. Research on the calculation of the
modeanalysis respons spectrum of the seismic base isolation steel storage tank.
2005 .Journal of Harbin Institute of Technology, Vol. 37, No. 5.

2. Zhou Wenjian, Chen Hong, Wen Bangchun. 2005. Seismic response analysis of
reactorcoolant pump in nuclear power plant. Journal of Vibration and Shock,
Vol. 25, No. 1.

86
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Seismic analysis of primary sodium system

components for the loop type fast
breeder reactor (5-2000)

P. Chellapandi, S.C. Chetal, Baldev Raj

Reactor Engineering Group, Indira Gandhi Centre for Atomic Research
Kalpakkam-603 102, India
e-mail: [email protected]

For 40 MWt (13.5 MWe) loop type fast breeder test reactor (FBTR) operating
with an unique plutonium rich carbide fuel, seismic analysis the primary sodium
system components is reported towards seismic reevaluation purpose. The main
components of the primary sodium system are the reactor vessel, two
intermediate heat exchangers (IHX) and two sodium pumps. As far as sodium
pipings are concerned, the hot pipe lines coming from the reactor joining to
intermediate pipe line between IHX & pump, cold line running from pump to
‘Y’ junction, called ‘cullotte’ and finally reactor inlet pipe are the main sodium
pipings. The main pipe is provided with double envelope, throughout its length.
There are hangers attached in the double envelope of the hot pipings and bellows
in the double envelopes of the pipelines. At a few locations the double envelopes
are welded to the main pipes. At few more selected locations, there are only
guides which allow the axial sliding while constraining all the radial directions,
between them. Thus, the system is complex because of strong coupling of
components, pipelines and double envelopes. Hence, there is a need to analyse
them together with appropriate boundary conditions, which need special
kinematic relations to be implemented in the computer code. To comply the
design code requirements, the analysis is carried out for the dead load, internal
pressure and seismic excitations. For the purpose of seismic reevaluation, review
base ground motion (RBGM) spectra were generated at the ground level.
Subsequently, floor response spectra (FRS) at the primary system support
elevations are generated from the seismic analysis of civil structures. FRS
generated at the elevation of reactor supporting elevation in two horizontal and
one vertical directions corresponding to 5% damping are applied in such a
manner to yield conservative results.
The analysis is aimed at to determine displacements and stresses to check the
functional and design code limits. For preventing mechanical interactions
between main component / piping and their respective double envelopes, the
relative radial displacements are limited to gap between main and double
envelopes at respective locations. For ensuring the structural integrity of
bellows, the effective axial deflections of bellow are limited to the respective

87
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

limits prescribed by the bellow manufacturer. Stresses are limited by the primary
stress limits recommended by RCC-MR (2002 edition).
While pipelines have 1 D feature, components, especially at the junctions and
branch pipes call for 3 D treatment. Addressing these issues, seismic analysis is
carried out by following an integrated approach. Finite element method is used
for the entire analysis, with the computer code called ‘CAST3M’ issued by CEA
France. The analysis is completed in three steps. In step-1, global analysis is
carried out to determine the deflections, forces and moments due to dead load
and seismic loadings using straight pipe elements and bends. The deflections are
used for verifying the deflection limits. The forces and moments are used for the
computation of Pm & (Pm+Pb), following either step-2 or step-3. In step-2, Pm &
(Pm+Pb) are computed using by using the correlations recommended in RCC-MR
for the pipes, bends and branch pipes. The correlations for the tees
recommended in RCC-MR are used for the branch pipes by assuming that the
dimensional restrictions for the fillet radius, etc. are respected, based on which
critical branch pipes are identified for the detailed FEM analysis in step-3.
As the first phase, natural vibration analysis is carried out to determine natural
mode shapes and associated frequencies, which have been extracted up to 50 Hz.
Based on the analysis, it is concluded that the seismic behaviour of components
in east and west loops including double envelope are similar. The deflection
limits to prevent the mechanical interaction between the main and respective
double envelopes are met with comfortable margin. The maximum net axial
deflections are found to be less than minimum acceptable values. The stresses
induced in components namely, reactor vessel, IHX and pumps including their
double envelopes are small. As far as pipings are concerned, the hot lines are
critical, particularly the shell nozzle junctions. However, stress limits are met
with detailed FEM analysis. The pipe bends including cullotte are meeting the
design code limits.
In summary, all the main components in the primary sodium systems in the as
built conditions meet the seismic design requirements.

88
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Qualification of creep, fatigue and fracture

design of PFBR components based
on tests (5-2002)

P. Chellapandi, R. Srinivasan, S.C. Chetal, Baldev Raj

Indira Gandhi Centre for Atomic Research, Kalpakkam-603102 India
e-mail: [email protected]

In 500 MWe Prototype Fast Breeder Reactor (PFBR), the critical out-of-core
components are main vessel (MV), control plug (CP), inner vessel (IV),
intermediate heat exchangers (IHX), steam generators (SG) and hot pipelines
Except for SG which is made of modified 9Cr-1Mo (G91), austenitic stainless
steel (ASS) is used for other components. The salient structural mechanics
features are large size thin walled shell structures, relatively low operating
pressure (< 1 MPa, except SG which operates at 17 MPa), high operating
temperatures (820 K for hot pool) and large thermal gradients (ΔT of 150 K
between hot and cold pool). These components are designed as per French
Design Code RCC-MR (1993)1, for the design life of 40 y. As per the code, the
design is done by analysis for which mainly numerical techniques by finite
element is followed.
The components will be manufactured indigenously. In order to ensure that
the design, analysis, indigenous material and indigenous manufacturing
technology comply with the design and construction code rules, tests are carried
out on a few important full scale components and mockups having component
features such as welds, multiaxiality and stress concentration effects under
simulated loading conditions. Particularly in the domain of creep, fatigue and
fracture design, a series of tests were conducted in Structural Mechanics
Laboratory (SML) with the objectives of qualifying the performance of
components in the reactor and the fracture assessment procedure for the FBR
application and for demonstrating leak before break (LBB) argument for MV,
sodium piping and SG.
This paper highlights the summary of theoretical analyses that have been
carried out on creep, fatigue and fracture design of critical components.
Subsequently, the paper deals with a few of the experimental investigations that
have been carried out essentially to qualify the creep-relaxation behaviour of
IHX tube to tubesheet joint, creep rupture strength of SG tubes, fatigue and
fracture assessment of SG tube bends and LBB justification of a typical full
scale Tee of secondary sodium circuit.
Creep-fatigue and fracture analyses completed for PFBR components are
validated by systematically planned experiments. The following are the
important outcome of these experiments.

89
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

ƒ Rolled and welded tube to tubesheet joints of IHX are stronger than the
basic tubes. Hence, in IHX, tubes decide the pullout strength rather than the
joint even after relaxation.
ƒ Creep rupture life of SG tubes, extrapolated from the test data using
appropriate Larson-Miller parameter, is more than ~ 2000 times the design
plant life without accounting for wall thinning due to corrosion.
ƒ A few thermal duty cycles impose stress range in SG bend tubes, more than
3Sm limit of design codes. However, tests conducted on 12 tubes at room
temperature as per the experimental route recommended by ASME-Section
III, reveal that the tubes can withstand more than 80,000 load cycles
(minimum). After applying a factor of safety of 8.6, computed as per the
code, the allowable number duty cycles is ~ 9300, which is much more than
the design load cycles (~860).
ƒ French guide A16 on LBB assessment, yields reasonably accurate prediction
of crack propagation and global instability of SG tubes. Hence, A16 rules
which are validated mainly for austenitic steel applications, can be applied
confidently for the modified 9Cr-1Mo steel. LEFM approach is found to be
sufficient for this material.
ƒ For the large size pipe Tees and bends, LBB argument can be applied
comfortably. However, buckling should also be considered in the collapse
load assessment procedure for the pipe bends and Tees in general, and Tees
in particular.
ƒ The creep damage estimation procedure specified in RCC-MR: Appendix
A16 which is based on the σ d approach is critically investigated and
improvements needed for A16 were recommended, mainly by application of
(i) appropriate multiaxial creep damage criteria, (ii) improved Neuber’s rule
for predicting elastoplastic stresses and (iii) relaxation of equivalent stresses.
The improved procedure predicts the experimental creep initiation life
satisfactorily.

90
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Inelastic strain at sliding joint between

primary ramp and primary tilting
mechanism of prototype fast
breeder reactor (5-2004)
Bhuwan Chandra Sati, S. Jalaldeen, Sanjeev Kumar, S. Raghupathy,
P. Chellapandi, S.C. Chetal
Nuclear Engineering Group
Indira Gandhi Centre for Atomic Research, Kalpakkam, India
e-mail: [email protected]

Inclined Fuel Transfer Machine (IFTM) is one of the fuel handling machines of
PFBR fuel handling system, which transfers the core subassemblies from in-
vessel transfer position (IVTP) to ex- vessel transfer position (EVTP) and vice
versa. Primary Ramp (PR) and Primary Tilting Mechanism (PTM) are two
important components of IFTM. Both are subjected to high temperature
environment, as they are located inside the hot pool of sodium. PR is fixed at the
top of the roof slab whereas PTM is fixed on the grid plate. Both are connected
with a sliding joint to facilitate the smooth movement of transfer pot. PR and
PTM have differential thermal movements w.r.t. each other as they are fixed at
two different locations. Movement parallel to their axis is allowed at the sliding
joint and hence no restriction of axial differential thermal movement is considered.
Due to differential thermal movements of their support locations there will be
significant movement perpendicular to their axis which is restricted in the sliding
joint. Due to this restriction, bending of PR or PTM may occur and localized
deformation of the component at the edge inside the sliding joint may occur.
To study the deformation of edges of the PR and PTM inside the sliding joint,
analytical and experimental investigation of a simplified sliding joint is carried
out at room temperature. The model of sliding joint is subjected to pure bending
moment. The elasto-plastic analysis of the model is carried out to understand the
local deformation, ovality, plastic strain etc.
To validate the analysis methodology analytical prediction and experimental
observations are compared. The results are matching well between analysis and
experiment. The same methodology of applied boundary conditions has been
applied to PFBR IFTM where there is a sliding joint between PR and PTM.
Elasto-plastic analysis has been carried out to find out the maximum inelastic
strain at the location of the sliding joint for 50 mm relative displacement
between primary ramp and primary tilting mechanism. The maximum local
strain obtained is 0.12% which is well within allowable limit of 5% (local). The
ovality on the primary ramp after applying relative displacement is found to be
negligible. More details of analytical and experimental investigations will be
presented in the paper.

91
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Structural integrity assessment of DHX

under CDA pressure loading (5-2006)

Sajish, S.D, R. Srinivasan, P. Chellapandi, S.C. Chetal

Nuclear Engineering Group
Indira Gandhi Center for Atomic Research-Kalpakkam, India
e-mail: [email protected]

The reactor assembly (RA) of Prototype Fast Breeder Reactor (PFBR) consists
of various components like Main vessel (MV), Inner Vessel (IV), Intermediate
Heat Exchanger (IHX), Decay Heat Exchanger (DHX) and Primary Sodium
Pump (PSP) and Reactor Core etc. The entire primary sodium circuit is housed
within the MV, which contains about 1100 t of radioactive primary sodium.
Hence the integrity of the entire reactor assembly is to be demonstrated even for
most extreme and unlikely events like earthquake and Core Disruptive Accident
(CDA). Among these loadings, CDA resulting from core meltdown is a very low
probability event in an FBR and hence it is considered as beyond design basis
event. Nevertheless as a defense in depth approach, the structural integrity of the
RA is demonstrated through complex numerical analysis, which involves
calculations of fluid transients, structural response and fluid structure interaction
effects [1, 2] and also through experiments on scaled down models. Experiments
simulating the CDA conducted on 1/13th scale model of RA demonstrated the
structural integrity of major components like IHX, DHX and PSP under the
extreme transient pressure loading developed as a result of CDA. This paper
give the details of the numerical analysis carried out on a 1/13th scale model of
the DHX which is used in the CDA mockup studies under transient pressure
loading which simulates the pressure distribution obtained form the mockup
study and response have been compared with the prototype which undergoes a
pressure loading expected during a CDA event predicted from the CDA analysis.
The modeling and analysis of both the prototype and scaled down model of
the DHX has been done using finite element code CAST3M. Fast transient
analysis of reactor assembly of PFBR for an energy release of 100 MJ has been
carried out using in-house code ‘FUSTIN’ and the transient loading in the
vicinity of DHX has been extracted which is used for the analysis of DHX
prototype. Similarly pressure pulse in the vicinity of DHX model has been
obtained from experiments simulating the energy release of 100 MJ for the
1/13th scaled down model of reactor assembly using explosives.
In order to verify the scaling laws maintained for the model and prototype and
to verify the dynamic amplification during the transient loading condition,
natural frequency analysis has been carried out for both DHX model and
prototype and frequency of bending mode has been compared (5.1 Hz and 67 Hz
for prototype and model respectively).

92
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Response has been calculated for the transient pressure loading for the model
and prototype of the DHX based on Newmark time integration. Maximum
response viz displacement and stress have been extracted at different location of
interest. From the analysis it was found the stresses developed in the prototype
under actual CDA loading condition is less than that of the model under the test
condition there by proving that the results obtained from the mockup studies of
1/13th scale model of RA under simulated CDA loading conditions is
conservative. It can be noted that these stresses are the most pessimistic
estimation without considering the communication effect of fluid inside and
outside of the shell. In real case the net stresses will be considerably less than
these values due to the differential pressure acting on the shell. The realistic
stresses acting on the DHX model and prototype therefore can be obtained by
multiplying the stresses with the reduction factor ‘α‘.
Dynamic response analysis has been carried out for the 1/13th scale model as
well as prototype of the DHX of PFBR to determine the dynamic response and
stresses under transient pressure during a CDA event. Transient loading acting
on the DHX is idealised to be a triangular pressure pulse. Along the
circumference of the shell pressure is assumed to have a cosine distribution.
Analysis results show that the maximum stress in 1/13th scale model is higher
than that of the prototype. So it can be concluded that the results of the mock up
study conducted on 1/13th scale model of the RA gives a conservative estimate
of the actual CDA loading. It is worth mentioning that DHX was not damaged in
the experiment and hence the structural integrity of DHX is ensured for the CDA
loading for PFBR.

References
1. P. Chellapandi. FUSTIN- A Code for Structural Analysis of Primary Containment Under
CDA. PFBR/31050/DN/1016.

2. P. Chellapandi. Structural Integrity Assessment of Primary Containment Under CDA.

PFBR/31050/DN/1015/R-A.

3. Alok gupta. Seismic Analysis of DHX (Type-A). PFBR/34130/DN/1001/R-A.

4. R.W. Clough, J. Penzien. Dynamics of Structures. McGraw-HILL Inc.

93
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Experimental compression behavior of

Stiffened Steel-Plate Concrete (SSC)
structures under compressive
loading (5-2008)

Byong Jeong Choi1, Keunkyeong Kim2, Chong-Hak Kim3, Tae Young Kim4
1
Associate Professor, Dept. of Architectural Engineering, Kyonggi University
Seodaemoon-Gu, Chungjeong Ro 2Ga 71, Seoul, Korea
e-mail: [email protected]
2
Korea Hydeo and Nuclear Power Company
Youngdongdaero 411, Kangnam-Gu, Seoul Korea, e-mail: [email protected]
3
Korea Hydro and Nuclear Power Co.
Daejeon, Korea, e-mail: [email protected]
4
Principal Engineer, Dept. of Civil Engineering, Korea Power Engineering Co.
Yongin, Korea, e-mail: [email protected]

Introduction and background

The importance of the research on the steel-plated concrete (SC) structure is now
ever increasing in nuclear structures. The SC structures can be applied to area such
as the first primary shield wall in nuclear structures to reduce the construction period
dramatically. It is necessary to understand the compression characteristic of the new
stiffened steel-plated concrete (SSC) structure under the compression loading.
Compression characteristic is one of important key factors among design variables.
This paper introduced and evaluated the compression characteristics of the stiffened
SC structures under vertical monotonic loading.

Aim of work
The SSC structures were consisted of headed stud bolts, concrete and steel plates
stiffened by W-shaped section. The primary aims of the work are to find and
suggest the empirical compression equation considering the creep effect of
concrete in the SSC structures. The maximum compressive forces resulted from
the experimental test results were also compared with the one of theoretical
finite element method (FEM). To understand of buckling behavior of the
stiffened steel plates in SSC structures are the secondly important aim of this
paper. The buckling patterns of the steel plates are compared with the results
from the egien-value analysis using FEM. The tension stresses of the steel plates
were directly measured from the experimental works during the compressing

94
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

loading period. At the same time, the initial stiffness and the effective length
factor were evaluated in this paper.
To achieve those phenomena of the SSC structures, three types of B/t ratio
were selected. The B/t ratio stands for the ratio between the pitch of headed stud
bolts and thickness of steel plate. The three types of B/t ratio were 25, 33, and 50
that represent inelastic and elastic stress behaviors. The compressive strength of
concrete was 42 MPa and the two types of steel plates are SM 490 and SS 400 in
the test. The thickness of steel plates was 6 mm. The diameter and length of the
headed stud bolt are 9 mm and 71 mm, respectively. The compression loading
was loaded monotonically to the top of the specimen using the Universal Testing
Machine, 10000 kN. The LVDT and strain gauges are installed to measure
stresses at many locations.

Essential results
The experimental works were carried out to suggest the empirical estimation of
compression force including creep and shrinkage effects. First of all, the
buckling shapes of the steel plates after the completion of the test were very
similar the one of FEM analysis, Fig. 1.

Figure 1. Buckling of Steel Plates.

The compression equation can be estimated by Eq. (1) through (2) and well
agreed with the experimental results.
Pn = Fcr As + F ys Ass + 0.85 f ck Ac (1)

Fcr = (1.55 − 0.125λ − 90ε n ) Fyp (2)

In equation (2), the buckling stress of the steel plate included the creep and
shrinkage effects. The quantitative value of compression force for the SSC
structures well agreed to the suggested Eq. (1) in average. The initial stiffness of
the SSC structure in elastic range was very similar to the one from FEM
analysis. The buckling stress measured at the steel plate directly to acquire the
initial stiffness.

95
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Conclusions and discussions

− The buckling shapes of the steel-plated concrete structure stiffened by W-
Shape section placed through the lateral direction. The patterns are well
agreed to the one of FEM analysis.
− The empirical compression equation was suggested and the compression
forces were well agreed to the resulted force from the suggested equation.
− The increase of the yield strength of the steel plate does not affect on the
increase of the maximum strength in most cases.
− The calculated effective length factor between the headed stud bolts was
around 0.5 for the SSC structures.

References
B.J. Choi et al. (2008). Compression Tests for Stiffened Steel Plate-Concrete Structures
with Variation of B/t Ratio. Journal of Korea Association of Steel Structure,
No. 20 Vol. 4, pp. 549–559.

M. Kanchi et al. (1996). Experimental Study on A concrete Filled Steel Structures, part 2.
Compression Test (1). Architectural Institute of Japan Conference, pp. 1071–1072.

M. Takeuchi, M. Narikawa, I. Matsuo, K. Hara, S. Uaami. (1998). Study on a Concrete

Filled Structure for Nuclear Power Plants. Nuclear Engineering and Design,
pp. 209–223.

96
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Developing a numerical model to describe

the mechanical behaviour of die-formed
expanded graphite for valve stems
sealing (5-2009)

Christophe Vasse
Electricité de France Recherche & Développement, Materials and Mechanics of
Components Department, avenue des Renardières – Ecuelles
77818 MORET-SUR-LOING Cedex, France, e-mail: [email protected]

Introduction
Stuffing box packings are very widely used for sealing valve stems, in the
chemical and petrochemical industries as well as in nuclear industry processes.
Following the banishment of asbestos in the 80’s, compressed exfoliated
graphite has become the most commonly used material for the designing of these
seals. However, little knowledge has been gathered so far on the precise
mechanical behaviour of this material.
As part of a project that aims to simulate the operation of nuclear power plants
valves in nominal conditions, the development of a numerical model that could
depict the main characteristics of the mechanical behaviour of CEG is studied. In
consideration of the main objectives of the project, we do not intend to build a
complex representation that would give a detailed account of the “microscopic
behaviour” of CEG ; we are more interested in a simplified model that could
predict the response of the packings, regarding the loadings they endure in
service.

Experimental approach
Compressions and discharges tests are performed on CEG samples on a 10000
daN hydraulic press. These samples are cylinders of diameter 32 mm, prepared
by axial compaction at a given pressure of a graphite “ribbon” of thickness 25
mm wrapped in a die. The tests are performed on the samples whether placed in
the die or free to expand in the radial direction. In the first case, we measure the
axial load and displacement imposed on the sample, as well as the lateral strains
of the outer wall of the die (from which we deduce the pressure applied by the
sample on the inner wall of the die). In the second case, we measure the axial
load and displacement, as well as the radial expansion of the sample.

97
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Finite element analysis

Based on the geometry of the test rig, a 2D axisymmetric finite elements model
was developed using Code_Aster. The Cam Clay model was chosen to represent
the mechanical behaviour of CEG. This model, more commonly used for soil
mechanics, requires the knowledge of eight parameters, four of which being
accessible through the experiments. Two parameters have to be fit numerically,
from which we deduce the two last parameters.
The friction between the stem of the valve and the packing placed around it is
of first importance regarding the operability of the valve. The pressure applied
on the packing must be sufficient to prevent leakage, but one has to ensure that
this pressure makes it possible for the actuator that drives the stem to open or
close the valve whenever required. Thus, the Cam Clay model was further
investigated modelling a whole packing placed between a stem and a stuffing
box. The model includes manoeuvres of the stem, contact and friction between
the packing and the stuffing box on one side, between the packing and the stem
on the other side. The results are compared with previous experiments carried
out in our labs.

Main results
Compressed exfoliated graphite (CEG) shows a very non-linear elastic-plastic
behaviour, resembling strongly that of certain kind of rocks, like clays, or
cohesionless soils, like sands. This similarity can partly be explained by the
porous nature of the material. The key characteristics of this complex behaviour
are: non-linear elasticity, apparition of permanent deformations at low levels of
pressure, linear hardening at higher pressures, and accommodation. As seen for
clays, this behaviour is usually well represented by such models as the Cam Clay
model.
The numerical results obtained with the Cam Clay model show a good
agreement with the “in die” tests, although the model does not account for the
accommodation. Furthermore, the computed axial forces required to drive the
stem are close to those measured experimentally, which reinforces our
confidence in the model.

Conclusions
The Cam Clay model is a first step towards a reliable representation of the
mechanical behaviour of CEG. The FE analysis based on this model shows quite
a good agreement with the fist “static” tests performed on CEG samples. The
model has now to be validated through complementary “dynamic” tests, which
will provide more information on the exact values of the friction ratios between

98
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

the packing and the stem. Future work will also investigate the influence of
temperature and the presence of fluid.

Main references
1. Investigating the load, friction and sealibility characteristics of graphite valve packings,
Maire Ni Ruaidhe, thesis, Trinity College of Dublin, 2003.

2. Poromécanique des joints en graphite expansé matricé, Elisabeth Patron, mémoire de

thèse, Ecole Nationale des Ponts et Chaussées, décembre 1999.

3. Stress-strain aspects of cohesionless soils under cyclic and transient loading, M.P.
Luong, International symposium on soils under cyclic and transient loading, 7–
11 janvier 1980.

4. Soil behaviour and critical state soils mechanics, David Muir Wood, Cambridge
University Press, 1990.

99
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Tests on reinforced concrete slabs with

pre-stressing and with transverse
reinforcement under impact
loading (5-2015)

Nebojsa Orbovic1, Medhat Elgohary2, Namho Lee2, Andrei Blahoianu1

1
Canadian Nuclear Safety Commission – Commission canadienne de sûreté
nucléaire, 280 Slater St. Ottawa, Ontario K1P 5S9, Canada
e-mail: [email protected]
2
Atomic Energy of Canada Limited (AECL)
2251 Speakman Drive, Mississauga, Ontario l5K 1B2, Canada

Introduction
Current empirical formulae for design of concrete elements under impact
loading are based on tests carried out on reinforced concrete elements with
longitudinal reinforcement and they are applicable only to such elements
(References 1, 2). The influence of pre-stressing or transverse reinforcement is
not taken into account. However, the code provisions (References 3, 4) for
punching resistance of concrete elements under conventional loadings take into
account beneficial effect of these parameters. In addition, the design acceptance
criteria according to these formulae are based on the damage of a concrete
element. More precisely, they are based on the visual damage of the rear side of
the concrete element and formulated as scabbing of the concrete surface or
missile perforation through the element. There is no quantification of the
damage in terms of width and depth of the scabbed area or in terms of deflection
of the element.

Aim of the work

The aim of this test campaign (Reference 5) carried out at VTT testing facility in
Espoo, Finland, was to assess the influence of longitudinal, bidirectional, pre-
stressing and transverse reinforcement, separately and combined, on punching
resistance of concrete elements under impact loading. Six tests presented in this
paper were carried out on low aspect ratio concrete slabs (l/h< 10; l is the slab
span, h is the slab thickness) under medium velocity (around 100 m/s) hard
missile impact. The variables were: the level of pre-stressing introduced using
threaded bars and the transverse reinforcement in form of T-headed bars. The
reference test was carried out on a reinforced concrete slab with longitudinal
reinforcement only. In two following tests, in addition to the longitudinal

100
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

reinforcement, pre-stressing was introduced using threaded bars. The slabs were
pre-stressed introduced in both longitudinal directions. The level of pre-stressing
was 5 MPa in one slab and 10 MPa in the other. In three following tests the same
slab design was used with additional transverse reinforcement in form of T-
headed: one slab was reinforced concrete slab with longitudinal and transverse
reinforcement and two remaining slabs with mentioned pre-stressing levels and
transverse reinforcement. Two last tests were performed to asses the combined
effect of pre-stressing and transverse reinforcement.

Essential results
The results are presented in terms of scabbed concrete area and permanent
deflection for each tested slab. It was observed that the punching resistance of
pre-stressed concrete specimens (without transverse reinforcement) was lower
than the reference reinforced concrete specimen, which is not consistent with
code provisions for conventional loadings. However, the difference in damage
between two tested levels of pre-stressing (5 MPa and 10 MPa) was not
significant. On the other hand, the transverse reinforcement, in form of T-headed
bars, increases punching resistance of concrete elements under impact loading.
Therefore, the results are consistent with code provisions for punching resistance
under conventional loadings. The transverse reinforcement combined with pre-
stressing significantly improves punching resistance of concrete elements for
tested missile parameters. However, it was observed that combination of these
two parameters modifies the failure mode. A punching cone, which is a current
failure mode for target (concrete slab) and missile characteristics used in this
campaign, is reduced to a punching cylinder with a diameter comparable to the
missile diameter. Further tests are needed to evaluate ultimate resistance of slabs
with pre-stressing and transverse reinforcement.

Conclusions
The tests performed on concrete slabs under impact loading showed that
introducing pre-stressing in slabs with longitudinal reinforcement does not
increase (even decrease) their punching capacity. Introducing transverse
reinforcement in form of T-headed bars have beneficial effect related to the
punching resistance and the transverse reinforcement combined with pre-
stressing increase significantly the punching resistance for given slab and missile
characteristics. Further tests are needed to asses the benefits of these two
parameters related to ultimate resistance of concrete slabs under impact loading.

101
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

References
1. G.E. Sliter. Assessment of Empirical Concrete Impact Formulas, ASCE, Journal of the
Structural Division, May 1980.

2. E. Buzaud, C. Cazaubon, D. Chauvel. Assessment of empirical formulae for local

response of concrete structures to hard projectile impact, CONSEC’07, Tours,
France.

3. CSA-A23.3-04. Design of Concrete Structures. Canadian Standards Association, 2004.

4. ACI 318-05. Building Code Requirements for Structural Concrete and Commentary.
American Concrete Institute, 2005.

5. A. Lastunen, I. Hakola, E. Järvinen, J. Hyvärinen, K. Calonius. Impact Test Facility,

Paper # J08/2-1, Transactions, SMiRT 19, Toronto, August 2007.

102
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Protection of seismic structures using

variable FPS typed TMD system (5-2016)

Ging-Long Lin, Chi-Chang Lin, Jer-Fu Wang

Department of Civil Engineering, National Chung Hsing University
250 Kuo-Kuang Road, Taichung, Taiwan 40227, ROC
Tel: 886-4- 2287-2221 ex. 225, Fax: 886-4-22851992
e-mail: [email protected]

Introduction
Passive tuned mass damper (TMD) is a wildly used control device in reducing
structural dynamic responses due to environmental excitations such as winds,
earthquakes, and man-made loadings [1–3]. A TMD system consists of an added
mass together with properly designed spring and damping elements that increase
the controlled modal damping for the primary structure. By attaching a TMD to
the structure, the vibration energy of the structure can be transferred to the TMD
and dissipated by the damping mechanism of the TMD. The design and
application of the traditional TMD (linear typed) systems are well developed,
but nonlinear typed TMD systems are still developing.
It is known that energy dissipation technology using friction mechanisms is
an effective means for vibration mitigation of seismic structures. Some
researchers suggested the application of the friction pendulum system (FPS)
typed TMD system. For a FPS typed TMD, the shape of the sliding interface is
made spherical, so that the TMD gravitational load applied on the slider will
provide a restoring stiffness for the TMD to return to its original position after
an earthquake. This restoring stiffness can be determined by the radius of
curvature of the spherical sliding surface, which is a fixed value. Compared with
the traditional TMD, the FPS typed TMD has following advantages: (1) Mass
independent self re-centering ability. (2) Energy dissipating by friction
mechanism. (3) No need to install an extra spring and damping device. (4) No
need to install a suspended mechanism, the space demand is thus much smaller.
On the other hand, since the slip force is a pre-determined fixed value, the
FPS typed TMD will start to slip (be activated) and dissipate seismic energy,
only when the friction force exerted by the seismic TMD motion exceeds the
constant slip force. A FPS typed TMD is not different from an added mass of the
primary structure, if it is not in the slip state. In other words, the TMD may lose
its tuning and energy dissipating ability when the TMD is in its stick state. The
determination of the slip force level is always a key issue in the design of
structures with FPS typed TMD. In the design practice, an earthquake with a
predicted intensity is usually assumed, and the passive FPS typed TMD is then

103
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

designed according to this intensity. As a result, the TMDs may perform well
under earthquakes of the predicted intensity, but may not perform well under
others.

Aim of the work

In order to overcome the problems mentioned above, a variable FPS typed TMD
system is evaluated in this paper. The friction force of the variable FPS typed
TMD system is controllable. A non-sticking friction (NSF) control law [4, 5],
which is able to keep the TMD activated and in its slip state throughout an
earthquake with arbitrary intensity, was conducted. The disadvantages of the
passive FPS typed TMD system can be improved.

Conclusions
The possibility of using a variable FPS typed TMD system for the protection of
seismic structures is considered in this study. The performance of the variable
FPS typed TMD with NSF controller for protection of seismic structures was
investigated numerically. The numerical results show that the proposed variable
FPS typed TMD system can improve the disadvantages of the traditional passive
friction typed TMD system.

References
1. G.W. Housner, L.A. Bergman, T.K. Caughey, A.G. Chassiakos, R.O. Claus, S.F. Masri,
R.E. Skelton, T.T. Soong, B.F. Spencer, J.T.P. Yao. Structural control: past,
present and future. Journal of Engineering Mechanics, ASCE, Vol. 123, No. 9,
pp. 897–971, 1997.

2. C.C. Lin, J.F. Wang, J.M. Ueng. Vibration Control Identification of Seismically-Excited
MDOF Structure-PTMD Systems. Journal of Sound and Vibration, Vol. 240, No.
1, pp. 87–115, 2001.

3. J.M. Ueng, C.C. Lin, J.F. Wang. Practical Design Issues of Tuned Mass Dampers for
Torsionally-Coupled Buildings under Earthquake Loadings. Structural Design of
Tall and Special Buildings, Vol. 17, No. 3, pp. 133–165, 2008.

4. C.L. Ng, Y.L. Xu. Semi-active control of a building complex with variable friction
dampers. Engineering Structures, Vol. 29, pp. 1209–1225, 2007.

5. W.L. He, A.K. Agrawal, J.N. Yang. Novel semi active friction controller for linear
structures against earthquakes. Journal of Structural Engineering, ASCE, Vol.
129, No. 7, pp. 941–950, 2003.

104
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Crack opening in steam generator tubes

submitted to an internal pressure:
experimental and numerical
modelling (5-2019)

J.-C. Le Roux, A. Zeghadi

EDF R&D
MMC Department
Site des Renardières, avenue des renardières – Ecuelles
F-77818 Môret-Sur-Loing cedex
e-mails: [email protected], [email protected]

In the framework of a systems reliability project in steam generator (SG) which

aim is to estimate leak rate in SG tubes, we study the longitudinal crack opening
modelling in the hard rolling transition zone of a SG tube submitted to hydrotest
conditions.
To reach this goal, an experimental system (see figure 1) has been specifically
developed in order to measure crack opening area in a SG tube [1]. Hard-rolled
and virgin tubes with one longitudinal crack are tested. The tubes which are
partly rolled are embedded in a metallic thick-walled cylinder whose interest is
to limit the expansion of the tube due to the rolling and to the internal pressure
applied. It represents the tubesheet’s effect in SG. The first experimental results
show that for low loadings and small crack sizes, the crack opening behaviour is
reversible. Beyond some crack size and a certain internal pressure value, the
crack keeps one residual opening. Besides, it has been shown that for cracks
with a sufficient length, the crack opening does not depend on the zone where it
is (straight section or roll transition zone).
In a second step, finite element calculations using finite element code
Code_aster [2] have been performed in order to estimate the crack opening in
hard-rolled SG tube. The simulations take into account residual stresses due to
the hard-rolling and the effect of the tubesheet on the tube behaviour. Numerical
results are compared with the experimental ones and a simplified model is
proposed in order to estimate the crack opening of hard-rolled tubes in SG.

References
1. C. Sonnefraud, J.-C. Le Roux. 2008. Bilan des essais mécaniques hors débit –
Livrable L2c du projet. Fuite GV en épreuve. EDF R&D. Internal Report.

2. Code_Aster, www.code-aster.com.

105
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Tube with
thick-walled cylinder

Pressurization
system

Figure 1. Experimental system specifically developed to measure crack opening area in

SG tubes. An original pressurization system allows to applied a internal pressure to the
hard-rolled tube.

106
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Numerical simulation of plain concrete

fracture experiments with fictitious
crack model (5-2035)

S.M. Basha1, R.K. Singh2, T. Kant3

1
Architecture & Civil Engineering Division, e-mail: [email protected]
2
Reactor Safety Division, e-mail: [email protected]
Bhabha Atomic Research Centre, Trombay, Mumbai-400085, India
3
Indian Institute of Technology Bombay, Powai, Mumbai-400076, India
e-mail: [email protected]
Corresponding author: S.M.Basha e-mail: [email protected]

Fracture simulation of quasi-brittle material like concrete cannot be carried out

in a generalized manner, unless the material constitutive parameters are identified
and methods for their experimental determination are specified. A variety of
numerical methods to predict crack formation and crack propagation for
concrete has been proposed in recent few years. This paper presents case studies
on numerical simulation of fracture behaviour of plain concrete beam specimens
of grades used in Indian nuclear containments. Three point bend beam specimens
with different sizes and notch parameters are simulated using the fictitious crack
model (FCM). This has been accomplished with computer program for nonlinear
finite element analysis of reinforced concrete structure named ATENA version
3.30. The experimental test beams are of M45 grade concrete (Characteristic
cube strength). The ATENA formulation of constitutive relations is considered
in the plane stress state. A smeared approach is used to model the material
properties, such as cracks or distributed reinforcement. This means that material
properties namely the tensile strength and fracture energy are appropriately
defined within continuum mechanics framework for crack initiation and its
subsequent propagation upto the final collapse for the general finite element
analysis defined for a material point are valid within a certain material volume,
which is in this case is associated with the entire finite element. The constitutive
model is based on the stiffness and is described by the equation of equilibrium at
a material point. The specimen details are as follows.

Beam Specimen sizes (mm) Notch Thickness of

Nomenclature depth (mm) notch (mm)
Length Depth Width
D1P20UB01 376 94 47 23.5 2.0
D2P20UC02 752 188 94 62 .0 2.0

107
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

The load vs CMOD (crack mouth opening displacement) obtained from

experiment and numerical simulations are compared for a typical D2 specimen
identified above in Fig. 1. These studies shows good agreement between the
experimental and numerical simulation, which confirm the constitutive material
models used in the code ATENA is powerful to capture the fracture process of
concrete which is also size dependent.

3500

3000 Experiment D1P20UB01

ATENA simulation
2500
Load(N)

2000

1500

1000

500

0
0.0 0.2 0.4 0.6 0.8 1.0
CMOD(mm)

Figure 1. Comparison of Experimental vs Numerical Simulation of Fracture Process of D2

specimen.

108
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Study of incident water hammer in an

engineering loop under two-phase
flow experiment (5-2042)

R.K. Singh, S.K. Sinha A. Rama Rao1

Vibration Laboratory Section, Reactor Engineering Division
Bhabha Atomic Research Centre
Mumbai, India
1
e-mail: [email protected]

Introduction
Water hammer is a common phenomenon when sub-cooled water flow in pipes
or other containers filled with steam or steam-water mixture. They also appear as
the consequence of fast closing or opening actions of valve or break in the
pipelines conveying single or two-phase flow. In the latter case, shock waves are
generated. In all the cases however, strong dynamic stresses are induced in the
wall of the closed system or equipment. These stresses are very different from
the stresses caused by static loading. The lack of experimental data obtained for
well-defined geometric boundary conditions is a significant obstacle for
validation of codes on fluid-structure interaction problems.
Two incident of loud sound, with in a span of four months was reported by
operators of a full-scale engineering loop erected especially for carrying out
experiments on loss of coolant accident and natural circulation. As the loop was
required to cater for large number of engineering experiments it was
instrumented with vibration and shock transducers and capture sound wherever
it happens in the loop. Condition was recreated for steam condensation induced
water hammer in the loop and the loop parameters were closely studied to
estimate the arrival time and time delay between multiple shock.

Engineering loop
The loop mainly consists of a vertical fuel channel simulator (FCS), steam
drum (SD), down comer, header, feeder and riser. The FCS is made of 54
electrically heated fuel rods cluster [1]. The loop simulates the elevation,
pressure, temperature, velocity and time scales of prototype to carry out study on
thermal hydraulic issues of natural circulation of main heat transport system
(MHTS). The heat sink for the test facility is provided in the form of a secondary
coolant circuit comprising of jet condenser, pool boiling coolers and feed pump.

109
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

The design pressure and temperature of the loop are 100 bars and 315 deg
centigrade.
When the steam is let out from the steam drum, it flows down to jet condenser
where it encounters jet of sub-cooled water and condenses. The condensed water
is pumped back into the steam drum. From lower bottom of the steam drum, the
down comer (26 meters) conducts hot water to a header and from header the
water is fed to FCS through a feeder (16 meters). A separate loop feeds sub
cooled water to jet condenser. A pool boiler connected in this circuit helps in
maintaining spray water temperature.

Incident of water hammer

When the loop was in operation at 2% power (52 KWth) for shut down
cooling experiments, the operators heard a loud sound. The heaters of FCS were
switched of and the loop was brought to low temperature and pressure. After
about a week of operation, the loud sound was again heard. After confirming the
incident with all the crewmembers of the loop, it was decided to investigate the
source and location of loud noise.

Possible mechanisms
Three possible mechanisms were considerer.
• Water hammer due to fall of big water slug in the tail pipe.
• Condensation induced water hammer in the steam line. [2], [4]
• Water hammer due to sudden condensation of steam in the jet condenser.

Recreation of condition for water hammer

The loop was fully instrumented to capture pressure, temperature, flow rate and
fluid levels in various sections of the loop in addition to vibration sensors to
locate the origin of water hammer. The loop was operated to simulate the
condition of all the three possible mechanisms.
When steam line was opened for the steam to flow from SD to JC and spray
water flow was started. The spray water temperature at the start of spray was 43
deg centigrade and the flow rate adjusted 100 LPM. With in a delay of 2
minutes, a series of five shocks were recorded one after the other. Every
subsequent shock was felt with increasing intensity. After the 5th shock, the JC
spray was stopped and the loop was brought to zero power.
Signal Analysis and role of super heat in the steam for time delay between the
shocks.

110
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

An attempt was made to explain the reason for time delay different between
the events through energy balance equation and the time required to quench the
quantity of super heat in the steam was estimated.

References
1. M.P. Paidoussis. Fluid Structure Interactions. Slender Structures and Axial Flow. Vols.
1 & 2. ISBN 0-12-544360-9 (1998).

2. E. Altstadt et al. Fluid Structure Interaction Investigations for piping. Wissenschaftlich-

Technische Berichte FZR-393. Dec-2003. ISSN 1437-322X.

3. Ki Yong Chei et al. Direct Contact Condensation Heat Transfer Model in RELAP5/
MOD3.2 with/without Non Nondensable Gas for Horizontally Stratified Flow.
NED 211(2002). Pp. 139–151.

111
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Advanced analysis of gasketed pressure

vessel closure systems (5-2043)

Billon François1, Batmale Guillaume2, Goussard Pascale3

1
Comex Nucléaire
36 boulevard des océans - BP 137 - 13273 Marseille Cedex 09- France
e-mail: [email protected].
2
Comex Nucléaire
36 boulevard des océans - BP 137 - 13273 Marseille Cedex 09- France
e-mail: [email protected].
3
ESI Group
70 rue Robert - 69006 Lyon - France
e-mail: [email protected].

Introduction/background
The pressure vessels have closure systems allowing visiting the inner of the
component or used for maintenance activities during its operation. For these
closure systems, in particular for nuclear components, their mechanical
resistance and their leak tightness must be demonstrated.

Aim of the work

The aim of the work was to develop a method to model completely a closure
system, including the studs, the gasket, the nuts, the flange or cover and the part
of the pressure vessel, for a 3-dimensional non-linear analysis taking into
account the contacts between the gasket and their seating surfaces as well as the
non linearity of the gaskets and the tightening preload of the studs.

Main results
Therefore, with such a complete representative non-linear 3D model we can:
(a) perform the stress analysis in all parts of the assembly for all the
operating conditions (temperature and pressure) and demonstrate their
mechanical resistance (stress limit, fatigue, fracture, etc.)
(b) analyze the behaviour of the contact area between the gasket and its
seating surfaces, for all the operating conditions. By doing so, we can
assess the history of the compressive forces on the gaskets in order to
demonstrate that the necessary conditions for the leak tightness are met,

112
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

particularly the metal-to-metal contact for multi-ring (metallic-graphite-

metallic) gaskets and as an important consequence, the re-calculation of
the necessary tightening preload applied to the studs by the tensioner.
This method was used, for example, for the behaviour analysis of man
holes, hand holes and eye holes of nuclear components.

Summary/conclusion
The results are meeting the requirements of the nuclear regulations and standards.
This method is applied to openings of nuclear components for the justification
of leak tightness of the in-service operating conditions.
In particular, the method was applied for the calculation of the initial
tightening load performed in hydraulic test condition and for which no leakage
was observed. This test can be considered as a successful scale test.

113
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Update and comparative study on seismic

wave incoherence in soil-structure
interaction (5-2048)

Wen S. Tseng1, Kiat Lilhanand2, Donald Hamasaki3, Julio Garcia4

International Civil Engineering Consultants,
A Division of Paul C. Rizzo Associates, Inc., Oakland, CA, USA
1
e-mail: [email protected]
2
e-mail: [email protected]
3
e-mail: [email protected]
4
e-mail: [email protected]

Introduction
Spatial incoherency of strong ground motions has the effect of lowering
translational earthquake input motion at building foundations but introducing
rocking and torsional behavior.
A methodology to incorporate spatial incoherence of seismic ground motions
in seismic soil-structure interaction (SSI) analysis and practical engineering
application guidelines were developed in a previous study (Tseng and Lilhanand,
1997). Along with the methodology, a computer program module INCOH was
also developed which works together with the SSI analysis computer program
SASSI. The combined SASSI-INCOH program is capable of: (1) calculating the
transfer function vector relating the incoherent motion vector to the motion at a
reference station conforming to a prescribed coherency function for motions
within the structural foundation, (2) modifying the coherent free-field motion
vector into the incoherent free-field motion vector, and (3) calculating the SSI
response of structures with arbitrary foundation configurations due to the
incoherent free-field input motion in the same way as the SSI analysis for the
conventional coherent ground motion input (ICEC, 1998).

Aim of the work

This study describes an update on development of the computer program
INCOH to incorporate: (1) the empirical spatial coherency models based on the
Pinyon Flat array data at a hard rock site, and (2) a coherency model based on an
extensive set of array data at soil sites (Abrahamson, 2007).

114
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Results and conclusions

The new development is illustrated with an SSI study of a coupled Auxiliary and
Shield Building (ASB), Steel Containment Vessel (SCV), and Containment
Internal Structure (CIS) of a nuclear power plant. The coupled structure complex
studied corresponds to a combined nuclear island structure of the AP1000
nuclear power plant located on a rock site. The results of this study are compared
with a recent publication (Short, Hardy, Merz and Johnson 2007). The
comparison shows very good agreement. Results of the study also serve to
demonstrate the effect, especially in the high frequency range, of incoherence of
seismic input motion conforming to the newly updated spatial coherency models
for rock sites on the SSI responses of the typical new generation of nuclear
power plant structures.

References
Abrahamson, N. (2007). Program on Technology Innovation: Effects of Spatial Incoherence
on Seismic Ground Motions. Electric Power Research Institute, Palo Alto, CA,
December. Report 1015110.

International Civil Engineering Consultants – ICEC (1998). Computer Program INCOH,

Internal Report.

Short, S., Hardy, G., Merz, K., Johnson, J. (2007). Program on Technology Innovation:
Validation of CLASSI and SASSI Codes to Treat Seismic Wave Incoherence in
Soil-Structure Interaction (SSI) Analysis of Nuclear Power Plant Structures,
Electric Power Research Institute, Palo Alto, CA, November. Report 1015111.

Tseng, W.S., Lilhanand, K.I. (1997). Soil-Structure Interaction Analysis Incorporating

Spatial Incoherence of Ground Motions, Electric Power Research Institute, Palo
Alto, CA: November. Report TR-102631 2225.

115
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Effect of geometrical defects and cracks on

the collapse of heat exchanger U-bent tubes
submitted to external pressure (5-2049)

A. Limam1, C. Mathon2
1
Université de Lyon, INSA-Lyon, LGCIE, F-69621, Villeurbanne, France

The tube bundle of heat exchangers or steam generators used in nuclear power
plants is very often composed of U-bent tubes. These tubes are generally
characterized by low diameter to thickness ratios (D/t) and high strength
materials. D/t ratios as low as 10 to 20 are currently considered. In some cases, it
is possible for the tube bundle to be exposed to a high external pressure, so that
an adequate margin against buckling is an important design criterion.
For such a range of D/t, the collapse under external pressure is determined by
the inelastic behaviour of the tube material. Moreover, the real geometry of the
tubes has to be considered: during the manufacturing process, initially straight
tubes have to be bent, and their cross-section becomes more or less elliptical in
the curved part. The assessment of the true collapse pressure is quite
problematical, due to the presence of both geometrical and material nonlinearities.
The buckling pressure of a tube can be estimated conservatively using pressure
vessel codes, such as ASME Boiler and Pressure Vessel code [1], RCC-M [2]
(French nuclear code) or the German code [3]. In general these design codes
yield to conservative results, because their most common feature is to provide a
simple calculation formula, which of course cannot account for the complex
behaviour of an elastoplastic, imperfect, U-bent cross-section.
The most complete and reliable procedure for calculating the collapse pressure
of the oval tube is to perform geometrical and material non linear finite element
analysis, knowing that both nonlinearities interact in the problem. Extensive
numerical studies have been conducted to clarify the buckling of thick
cylindrical tubes submitted to lateral pressure or hydrostatic pressure. Effects of
geometrical initial imperfection (initial ovalization) are considered for different
amplitudes and in both configurations, the curved pipe configuration or the
straight one. Different strain hardenings are considered to gauge the effect of
material law on the collapse behaviour. In addition, the effect of other
imperfections, such as local wall thickness variations associated to corrosion or
wear against tube supports, cracks due to stress corrosion, are examined in the
light of the strong dependence on the inelastic behaviour associated to material
properties and the presence of initial ovalization.
Considering this large parametric numerical study conducted with different FE
codes (CAST3M, ABAQUS, Code_Aster), the methodology of the design and
recommendations are proposed. The obtained design curve is then compared to

116
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

current design codes. Finally, this study establishes clear and accurate design
criteria for the collapse of tubular structures under external pressure for heat
exchanger applications.

References
1. ASME Boiler and pressure vessel code, ASME, 2007 (2008a addenda).

2. RCC-M: Design and construction rules for mechanical components of PWR nuclear
rd
islands, AFCEN, June 2000 Edition & 3 addendum, June 2007.

3. AD-Merkblatt B6: Cylindrical shells under external pressure (German pressure vessel
code), 1995.

117
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Study of liquid dispersal from a missile

impacting a wall (5-2050)

1
Ari Silde1, Ari Kankkunen2, Juha Juntunen
1
VTT Technical Research Centre of Finland, Espoo, Finland
e-mail: [email protected]
2
HUT Helsinki University of Technology, Espoo, Finland

Fuel release and spread from impacted projectile are of interest for the
determination of fuel spread and fire risk following an airplane crash on a
structure, and for applying and validating simulation techniques. So far, little
representative experimental information can be found from the literature. Thus,
liquid dispersal processes have been studied at VTT in the selected medium-
scale IMPACT tests where deformable steel or aluminium projectiles are filled
with water and impacted on a solid concrete wall or a steel force plate [1], [2],
[3], [4]. Most of the “wet” missile tests have been conducted using a cylindrical
missile, but in few tests a more prototypical 3-D projectile which consists of
representative fuselage and wings has been used.
The earlier paper [5] concentrated on the measuring methods and procedures
used in the liquid dispersal study, and the main results of the preliminary
simulations of liquid spread using the Fire Dynamic Simulator (FDS) computer
program. Now the focus is on the major experimental findings of liquid release
and spread from ruptured missile impacting a wall. Related measuring
procedures are also dealt with. The main parameters of the liquid phenomena
measured are the velocity of the liquid front coming out from the ruptured
projectile, spreading angle and direction of liquid release, extent of liquid spray
from the target, water pooling area on the floor, and droplet size of the liquid
spray.
The test results show that the initial velocity of liquid spurting out from a
missile may be much higher than the missile impact velocity, but the speed of
droplets of the liquid spray slows down rapidly due to atomization processes and
relating drag.

References
1. Kärnä, T., Saarenheimo, A., Tuomala, M. 2004. Impact loaded structures. In: SAFIR,
The Finnish Research Programme on Nuclear Power Plants Safety, Interim
Report, VTT 2272, Espoo 2004, pp. 113–122.

2. Lastunen, A., Hakola, I., Järvinen, E., Calonius, K., Hyvärinen, J. 2007. Impact Test
Facility. SMiRT-19 Conference. August 2007, Toronto, Canada.

118
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

3. Saarenheimo, A., Hakola, I., Hyvärinen, J., Tuomala, M. 2006. Numerical and Experimental
Studies on Impact Loaded Concrete Structures. Proc., of ICONE 14,
International Conference on Nuclear Engineering, ICONE14-89477, Miami, USA,
July 17–20, 2006.

4. Saarenheimo, A., Tuomala, M., Calonius, K., Lastunen, A., Myllymäki, J., Hyvärinen, J.
2007. Numerical Studies of Impact Loaded Reinforced Concrete Walls. SMiRT-
19 Conference. August 2007, Toronto, Canada.

5. Silde, A., Hostikka, S., Kankkunen, A., Hakola, I., Hyvärinen, J. 2007. Experimental
and Numerical Studies on Liquid Dispersal from Projectile Impacting on Wall.
SMiRT-19 Conference. August 2007, Toronto, Canada.

119
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Structural evaluation of drop load effects on

buried structures (5-2057)

Mehdi S. Zarghamee1, Keng-Wit Lim1, Keith Henshaw2

1
Simpson Gumpertz & Heger Inc., Waltham, MA 02453, USA
2
Progress Energy, Crystal River Energy Complex, Crystal River, FL 34438-6708, USA
e-mails: [email protected], [email protected], [email protected]

Replacement of steam generators is likely to happen at least once during the

service lifetime of pressurized-water reactors. The replacement of old steam
generators requires the use of large rigging systems for lifting and moving of the
old steam generator out the containment structure and of the new steam
generator into the containment structure. These operations are generally
performed over an area adjacent to the containment structure where many buried
structures and utilities are located. Safe load paths are planar regions determined
for each lifted component during the various stages of lifting operation so as to
avoid damage to the buried structures to the extent that would compromise
functionality of the critical buried structure, in case of an accidental drop of a
lifted component. The critical buried structures are those structures that are
needed to remain functional during the generator replacement operation, such as
discharge and intake pipelines and electrical instruments and control panels
related to cooling system of the fuel elements.
This paper describes an approach that includes a combination of dynamic
finite element analyses and structural evaluation of the buried structures, to
perform structural evaluation for drop loads occurring during steam generator
replacement and arrive at final load paths for safe lifting operations.
The procedure is based on development of a three-dimensional dynamic soil-
structure finite-element model, using explicit transient finite element analysis.
The source of elastic wave resulting from impact of the drop load and
propagation through the soil media is simulated through low velocity impact of
objects on the free surface of foundation or soil. Energy dissipation is accounted
for through Rayleigh soil damping and nonlinear response of the dropping
objects, the targets and the buried structures. The transfer of dynamic analysis
results to the standard structural analysis software for structural evaluation is
discussed. A case study is presented in which the structural safety and the extent
of damage and its effect on the ability to function of the underground prestressed
concrete cylinder pipelines, ductile iron pipelines, and of a reinforced concrete
wall is assessed for drop components from a nuclear power plant steam
generator replacement lifting operation. Structural evaluation of the prestressed
concrete cylinder pipe was performed in both circumferential and longitudinal
directions based on limit states criteria used for the design of the pipe using
serviceability, damage and different modes of partial or total failure using

120
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

AWWA C304 (2007) and AWWA M9 (2008). Structural evaluation of ductile

iron pipe was performed in both circumferential and longitudinal directions
based on yielding and buckling limit states using AWWA C150 (2002). Structural
evaluation of the reinforced concrete wall was performed using ACI 318 (2005)
and ACI 349 (2006) and considering damage and was based on evaluation of
strains and the extent of cracking and deflection of the cracked wall.
In conclusion, a procedure based on finite element analysis of soil-structure
interaction is developed for structural evaluation of buried structures subjected
to drop load impact effect in both Defueled and Power Operation Modes The
procedure is used to classify the damage levels to buried structures as: “Code
Compliant”; “Not Fully Code Complaint, but Undamaged”; “Damaged, but
Functional”; or “Failed,” and is used to verify, modify, and/or provide new safe
load paths for lifting operations to avoid failure of the structural systems that
need operate during accidental drop loads. The procedures developed herein are
applicable for the evaluation of the effects of the impact of the accidental drop
loads during steam generator replacement on all types and forms of buried
structures.

References
American Concrete Institute. 2005. Building Code Requirements for Structural Concrete,
ACI 318.

American Concrete Institute. 2006. Code Requirements for Nuclear Safety-Related

Concrete Structures and Commentary, ACI 349.

American Water Works Association. 2002. Standard for Thickness Design of Ductile-Iron
Pipe, AWWA C150.

American Water Works Association. 2007. Standard for Design of Prestressed Concrete
Cylinder Pipe, AWWA C304.

American Water Works Association. 2008. Concrete Pressure Pipe, AWWA Manual M9.

121
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Experimental and numerical simulation of

radiolysis gas detonations in BWR
exhaust pipes and mechanical response
of the piping to the detonation
pressure loads (5-2062)

R. Redlinger1, M. Kuznetsov*1, W. Breitung1

J. Grune2, K. Sempert2, T. Franke3
1
Institute for Nuclear and Energy Technologies, Forschungszentrum Karlsruhe GmbH
P.O. Box 3640, D-76021 Karlsruhe, Germany
2
Pro-Science GmbH, Parkstr. 9, 76275 Ettlingen, Germany
3
EnBW Kernkraft GmbH, Kernkraftwerk Philippsburg
Rheinschanzinsel, 76661 Philippsburg, Germany
*
Corresponding author e-mail: [email protected]

Introduction
Radiolysis gas (2H2+O2) can accumulate in BWR steam piping in case of steam
condensation, with an ensuing detonation of the radiolysis gas being the likeliest
cause of a pipe and/or valve rupture. In the German Reactor Safety Commission
(RSK) recommendations for radiolysis gas control in BWR installations it is
demanded to determine the reaction pressure for the highest possible radiolysis
gas concentration that could arise. In the current work we consider a typical
BWR exhaust pipe, which connects the high pressure steam piping with the
ambient atmosphere, under the following “worst case” scenario: (a) accumulation
of radiolysis gas in an exhaust pipe, (b) fast valve opening to the high pressure
system with steam at 70 bar, and (c) adiabatic pressurization of the radiolysis gas
by the steam. Taking into account a water mirror level of 6 m from the open end,
this leads to an equilibrium state of 20 bar pressure and 602 K temperature for
the pressurized radiolysis gas.
The main purpose of the current work is an experimental and numerical
evaluation of the maximum pressure load plus the integrity of the BWR exhaust
pipe in case of a detonation of the pressurized radiolysis gas.

Experiments
Detonation experiments of radiolysis gas were performed in a real scale exhaust
pipe of 12.25 m length made from stainless steel Nr. 1.4541. The tube consisted
of two parts: (i) with 510 mm outer diameter and 14 mm wall thickness, and (ii)

122
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

with 419 mm outer diameter and 20 mm of wall thickness. To reproduce

the ”worst case” conditions for radiolysis gas after the steam pressurization we
used a radiolysis gas mixture at 10 bar and 293 K which energetically (with
respect to the pressure load) equals the pressurized BWR mixture at 20 bar and
602 K. The inert part of the tube was filled with nitrogen instead of steam to
enable experiments at ambient temperature. The radiolysis gas was included in a
thin shell vessel to separate it from the inert gas, with ignition at a metal plate
side used as an imitation of the gas-water interface in real conditions. 16 strain
gauges were installed along the tube to measure longitudinal and circumferential
deformations under the detonation process, and dynamic pressure loads were
measured using 4 pressure transducers installed on flanges.
Four experiments were performed: at 1.6, 5.0 and two tests at 10.0 bar. The
experiments showed that maximum deformations occurred at the end of the
radiolysis gas vessel. The maximum dynamic strain was measured to be 0.75%
for the radiolysis gas detonations at 10 bar, with the maximum remaining
deformation, using strain gauges and direct measurements, being about 0.15%.
This means that the exhaust tube remains intact even under this “worst case”
detonation scenario of the pressurized radiolysis gas mixture.

Numerical simulations
To simulate detonation of the radiolysis gas mixture at 20 bar and 602 K with
steam as an inert gas, the 3-dimensional CFD code DET3D [1] was used. This
(explicit) code solves the 3D Euler gas dynamic equations for a reacting gas
system taking into account effects of precursor shock waves, reflections and heat
losses on the resulting pressure. After validating the code against available
experimental pressure records, showing very good consistency of the calculated
pressure records, pressure profiles were calculated at all strain gauge positions.
Then, using a simplified 1D model for the mechanical response of a cylindrical
pipe under an internal dynamic pressure load, the dynamic strain corresponding
to these calculated pressure signals was determined. A comparison of these
calculated strain values with the experimental signals showed very good
agreement. This made it possible to use the DET3D code to predict the
detonation pressure loads for the original BWR gas mixture under the worst case
scenario, and to evaluate the resulting dynamic deformations and the integrity of
the exhaust pipe. It turned out that the differences in the calculated pressure
profiles for the 10 bar (293 K) and the 20 bar (602 K) radiolysis gas detonations
were sufficiently small so as to result in basically the same mechanical pipe
response, confirming the validity of the chosen experimental approach.

123
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Summary
Worst case scenario of radiolysis gas detonations in a BWR exhaust pipe at
pressures up to 20 bar was analyzed both experimentally and numerically.
It was demonstrated that the tube can withstand the detonation pressure at
such conditions. The maximum remaining deformation was measured to be not
more than 0.15%.

Reference
1. Redlinger, R. (2008) DET3D – a CFD tool for simulating hydrogen combustion in
nuclear reactor safety. Nucl. Eng. and Design, Vol. 238, pp. 610–617.

124
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

A study on seismic behavior of nuclear

power building in strong nonlinear area
and fragility evaluation using 3 dimensional
FEM. Part 1. Ultimate seismic condition
of building (5-2074)

Shodo Akita1, Takuya Suzuki2, Naohiro Nakamura3,

Masao Koba 4, Tomio Nakano5
1
Nuclear Power Division, The Kansai Electric Power Co., Inc.
Yokota 8, Goshi 13, Mihama, Mikata-gun, Fukui, Japan
e-mail: [email protected]
2
R&D Institute, Takenaka Corporation
1-5-1 Otsuka, Inzai, Chiba, Japan
e-mail: [email protected]
3
R&D Institute, Takenaka Corporation
1-5-1 Otsuka, Inzai, Chiba, Japan
e-mail: [email protected]
4
Nuclear Power Division, The Kansai Electric Power Co., Inc.
Yokota 8, Goshi 13, Mihama, Mikata-gun, Fukui, Japan
e-mail: [email protected]
5
NEWJEC Inc.
2-3-2 Honjo-higashi, Kita-ku, Osaka, Japan
e-mail: [email protected]

The evaluation based on probabilistic safety assessment (PSA) is expected for

nuclear power buildings because the risk of the occurrence of seismic ground
motions beyond the design assumption cannot be denied. For the assessment,
seismic ultimate analyses of the building are necessary.
In this paper, the seismic ultimate behavior was evaluated using an accurate
three-dimensional nonlinear FEM model. In the model, the basemat and the soil
were modeled by solid elements, and shear walls of the building were modeled
by nonlinear layered-shell elements. The uplift behavior was estimated using
joint elements between the basemat and the soil. The response analyses
considering the maximum horizontal acceleration up to 3500 Gal was done.
Then, the influence on the response given by the vertical ground motion and the
basemat uplift was evaluated. Moreover, the response was compared with that of
the lumped-mass model, which is generally used for current seismic design.
From the study, the following results were obtained.

125
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

1) The building reached the ultimate condition at 7 times of the design basis
ground motion input. Shear failure was occurred 3500 Gal input.
2) The horizontal response of the structure for simultaneous horizontal and
vertical input was almost the same as for horizontal only input, thus the
effect of vertical input was relatively small. The vertical response of the
structure for simultaneous input agreed well for vertical only input.
3) The effect of basemat uplift on the horizontal response was relatively
small. However, the effect on the vertical acceleration was not small. The
difference was considered as the vertical induced motion.
4) The shear strain of the lumped mass SR model exhibited almost the same
level as the FEM model in O/S. However in E/B, the SR model
overestimated damage compared to the FEM model.

126
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

A study on seismic behavior of nuclear

power building in strong nonlinear area and
fragility evaluation using 3 dimensional
FEM. Part-2. Fragility evaluation (5-2075)
Takuya Suzuki1, Shodo Akita2, Naohiro Nakamura3,
Masao Koba4, Tomio Nakano5
1
R&D Institute, Takenaka Corporation, 1-5-1 Otsuka, Inzai, Chiba, Japan
e-mail: [email protected]
2
Nuclear Power Division, The Kansai Electric Power Co., Inc., Yokota 8, Goshi 13,
Mihama, Mikata-gun, Fukui, Japan, e-mail: [email protected]
3
R&D Institute, Takenaka Corporation, 1-5-1 Otsuka, Inzai, Chiba, Japan
e-mail: [email protected]
4
Nuclear Power Division, The Kansai Electric Power Co., Inc., Yokota 8, Goshi 13,
Mihama, Mikata-gun, Fukui, Japan, e-mail: [email protected]
5
NEWJEC Inc., 2-3-2 Honjo-higashi, Kita-ku, Osaka, Japan
e-mail: [email protected]

The evaluation based on the probabilistic safety assessment (PSA) is expected

for nuclear power buildings because the risk of the occurrence of the seismic
ground motions beyond the design assumption cannot be denied. In this paper,
the building fragility evaluation of the seismic PSA was carried out using the 3
dimensional nonlinear FEM model based on the result of part-1.
As the fracture modes, the shear failure of the web wall and the flexural
failure and the compressive failure of the flange wall were assumed. The
fragility curves of the FEM model and lumped mass model in each analysis case
were calculated as follows. First, the failure probability was plotted on a diagram
for each input acceleration level where analysis was conducted. The failure
probability is calculated by considering the aleatory uncertainty of the response
and strength value. The plotted points are approximated by a lognormal
cumulative distribution function using the least squares method, which is taken
to be the fragility curve. From the study, the following results were obtained
1) In terms of the failure mode of the envisioned reinforced concrete seismic
walls, shear failure led to flexural failure and compressive failure.
2) In the evaluation of shear strain, the difference between the fragility
evaluations for horizontal input only and simultaneous horizontal and
vertical input was small, and the effect of vertical input was relatively small.
3) The effect of basemat uplift on fragility evaluation was relatively small.
In the fragility evaluation, the lumped mass SR model exhibited a tendency to
estimate damage largely compared to the FEM model.

127
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

The effect of foundation embedment on

seismic SSI response of EPR nuclear
island structures (5-2076)

Mansour Tabatabaie1, Basilio Sumodobila2, Calvin Wong3, Todd Oswald4

1
Chief Engineer, SC Solutions, 1330 Broadway, Ste. 907, Oakland, CA 94612
2
Principal Engineer, SC Solutions, 1330 Broadway, Ste. 907, Oakland, CA 94612
3
Supervisor Engr., New Plants Eng.
AREVA NP, 6399 San Ignacio Ave., San Jose, CA 95119
4
Engr. Mgr., New Reactor Projects
AREVA NP, 7207 IBM Drive, CLT-2D, Charlotte, NC, 28262

Introduction
The Evolutionary Pressurized Reactor (EPR) is an advanced nuclear power plant
design developed by AREVA NP, Inc. The EPR consists primarily of a nuclear
island (NI) and several other significant buildings outside the NI. A plan view of
EPR structures is shown in Figure 1. The structures within the NI consist of the
reactor building (RB), fuel building (FB), safeguard building 1 (SB1), safeguard
building 2/3 (SB2/3) and safeguard building 4 (SB4). The nuclear island is
embedded approximately 11.85 m below ground surface. The EPR standard
design certification considers ten generic soil profiles that range from hard rock
to stiff soil to soft soil conditions that are likely to be encountered at plant sites
in the United States. Three sets of free-field seismic ground motions representing
the rock/soil outcrop motions in the eastern United States are considered for the
standard design. Each soil profile is associated with one or, and in a few cases,
two of the seismic control motions.
This paper presents the seismic soil-structure interaction (SSI) analyses and
results of the embedded EPR NI common basemat structures for the ten generic
soil cases and three postulated seismic ground motions.

Analytical methodology
The soil structure interaction analyses were performed using the computer
program SASSI. SASSI uses finite element and complex frequency response
method to calculate dynamic SSI response of structures supported in
horizontally layered soils system over uniform half-space. The primary soil
material nonlinearity is the strain-compatible soil shear modulus and damping
ratios. The structure model consists of interconnected stick models of RB, FB,
SB1, SB2/3 and SB4 (NI sticks) connected rigidly to a common NI rigid

128
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

basemat (see Figure 2). To model the effect of embedment on the sticks, a series
of horizontal rigid links connect the side soil walls to the sticks below ground
surface. In developing the embedment model, it is assumed that basement walls
that are in contact with soil are rigid in the out-of-plane direction. The
foundation is modeled with horizontal soil layers over uniform halfspace with
control motion applied as outcrop motion at the basemat level.

Discussion of results
The results of the SSI analyses in terms of the envelop of maximum
accelerations and 5%-damped acceleration response spectra for different soil
cases are presented and compared for the embedded and surface-supported EPR
model at the foundation and several selected locations in the structure. In
general, the effect of embedment is to reduce the in-structure maximum
acceleration and acceleration response spectra across the frequency spectrum.
However, the degree of this reduction depends on the soil profile, NI structures
properties and input motion. In general, more reductions in the response due to
embedment effects seem to be associated with the softer soil profiles except
where the spectral response is affected by the structural frequency shift. As the
soil profiles become stiffer, the responses of the internal structures appear to be
less influenced by the embedment effects since they are not directly connected
by the side soils. In general, the structures that have direct contact with the side
soils (SB1, SB2/3, SB4 and FB) appear to be more sensitive to the embedment
effects regardless of the soil stiffness. Typical results are shown in Figure 3.

129
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Figure 1. Plan View of EPR Buildings.

Figure 2. SASSI Finite Element Structural Model.

130
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

2.0 3.0
Reference Outcrop Motion
Surface-Supported EPR Surface-Supported EPR
2.5 Embedded EPR
Embedded EPR

X-Spectral Acceleration (g's)

1.5
X-Spectral Acceleration (g's)

2.0

1.0 1.5

1.0
Damping = 0.05
Damping = 0.05
0.5
0.5

0.0
0.0
0.1 1 10 100
0.1 1 10 100
Frequency (Hz)
Frequency (Hz)

2.0 3.0
Reference Outcrop Motion
Surface-Supported EPR
Surface-Supported EPR
2.5 Embedded EPR
Embedded EPR

Y-Spectral Acceleration (g's)

1.5
Y-Spectral Acceleration (g's)

2.0

1.0 1.5

1.0
Damping = 0.05
Damping = 0.05
0.5
0.5

0.0
0.0
0.1 1 10 100
0.1 1 10 100
Frequency (Hz)
Frequency (Hz)

2.0 3.0
Reference Outcrop Motion
Surface-Supported EPR Surface-Supported EPR
Embedded EPR 2.5
Embedded EPR
1.5
Z-Spectral Acceleration (g's)

Z-Spectral Acceleration (g's)

2.0

1.0 1.5

1.0
Damping = 0.05 Damping = 0.05
0.5

0.5

0.0 0.0
0.1 1 10 100 0.1 1 10 100
Frequency (Hz) Frequency (Hz)

(a) Basemat (b) Reactor Building Internal

Figure 3. Comparison of Acceleration Response Spectra, Soil Case 2sn4u/EUR Medium

Motion.

131
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Component mode synthesis based SSI

analysis of complex structural systems
using SASSI (5-2089)

Mansour Tabatabaie1, Basilio Sumodobila2

1
Chief Engineer, SC Solutions, 1330 Broadway, Ste. 907, Oakland, CA 94612
2
Principal Engineer, SC Solutions, 1330 Broadway, Ste. 907, Oakland, CA 94612

Introduction
Three-dimensional seismic soil-structure interaction (SSI) analysis of nuclear
power plants (NPP) is often performed in frequency domain using programs
such as SASSI. This enables the analyst to properly a) address the effects of
wave radiation in an unbounded soil media, b) incorporate strain-compatible soil
shear modulus and damping properties and c) specify input motion in the free
field using de-convolution method and/or spatially variable ground motions. For
large, complex structural systems with several hundred thousand degrees of
freedom and large foundation impedance matrix associated with deeply
embedded foundations, the conventional sub-structuring analysis approach
employed in SASSI often results in a coefficient matrix that is too large to solve
with currently available computer resources. To address this problem, the
method of component mode synthesis (CMS) is employed in the SSI analysis.
This involves partitioning the structure into several interconnected components,
calculating the reduced-order model of each component, and then assembling the
reduced-order component models into a global model of the total SSI system.
This paper presents the formulation of component mode models, and their
implementation into the global SSI model.

Analytical methodology
The soil structure interaction analysis is performed using the computer program
SASSI. SASSI uses finite element and complex frequency response method to
calculate dynamic SSI response of structures supported in horizontally layered
soils system over uniform half-space. The CMS method has been implemented in
SASSI utilizing the super element capability. According to this implementation,
the structure is first partitioned from the foundation and analyzed as one or
several interconnected components using the ANSYS program to compute the
component mode properties that are used to form super elements. These super
elements are input into the foundation/soil model and analyzed by SASSI to
calculate the foundation response. The foundation response that includes the SSI

132
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

effects is then imposed onto the structural model to calculate the response of the
detailed structure.

Discussion of results
The effectiveness of this procedure is demonstrated by comparing the results of
seismic SSI analysis of a detailed typical NPP model (see Fig. 1) subject to
horizontal excitations. The total SSI system is analyzed with SASSI using the
conventional approach to compute the target solution. Following this, the
structure is partitioned to several components and re-analyzed using CMS
method implemented in SASSI. The results in terms of 5%-damped response
spectra and maximum accelerations at selected key locations (see Fig. 2) in the
structure are computed and compared. Comparison of the responses show close
agreement between the target solution and those obtained using component
mode synthesis.

52.8 m Primary
Containment

13.5 m Reactor Building

(Embedment) Foundation

Diameter = 40 m

Figure 1. Reactor Containment Building.

133
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Node 18241 Z
Y
X

Node 17091

Node 16262 Node 17284

Node 225

Node 25

Foundation Base slab Node 203

Figure 2. Interior View of Reactor Containment Building.

134
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Seismic capacity test of overhead crane

under horizontal and vertical excitation –
element model test results on non-linear
response behavior (5-2148)

Kenichi Suzuki1, Masakatsu Inagaki1, Shirou Fukunishi1,

Tadashi Iijima2, Takashi Matsumoto3
1
Seismic Safety Division, Japan Nuclear Energy Safety Organization
Kamiya-cyo MT, Bldg., 4-3-20, Toranomon, Minato-ku, Tokyo, Japan
e-mail: [email protected]
2
Mechanical Engineering Laboratory
Hitachi, Ltd., 832-2, Horinouchi, Hitachinaka, Ibaraki, Japan
e-mail: [email protected]
3
Plant Design Section, Hitachi-GE Nuclear Energy, Ltd.
1-1, Saiwai-cho, 3-chome, Hitachi, Ibaraki, Japan
e-mail: [email protected]
Design and Construction Issues

The validation of seismic capacity regarding the structural strength or operative

function of equipment has been a key issue for seismic safety assessments of
nuclear power plants. Since the new safety review guidelines for seismic design
of nuclear power plants in Japan, which were established in 2006, required
dynamic response analysis in the vertical direction, it has become an urgent issue
to obtain the seismic capacity data of equipment under vertical excitation.
An overhead crane used in a BWR reactor building is typical equipment that
has lower rigidity and higher response behavior in the vertical direction than in
the horizontal direction. High level horizontal and vertical excitation may cause
the nonlinear response behavior (slipping, leaping and landing) of the overhead
crane, resulting in its derailing from the track.
Accordingly, JNES initiated seismic test programs in fiscal 2006 for the
overhead crane. Seismic capacity tests of a 1/2.5 scale overhead crane model
under horizontal (traveling and traversing direction) and vertical excitation were
planned using the world’s largest high-performance shaking table (E-Defense).
The primary purposes of those tests were to evaluate the structural strength of an
overhead crane and its nonlinear response behavior, and to validate the structural
retrofit of components for prevention of its derailing from the track under high
level excitation, whose retrofit was applied in some of existing plants in Japan.
In order to minimize the seismic capacity test conditions of the scaled overhead
crane model, three types of element model tests were conducted in advance, by
focusing on the effects of key factors on the nonlinear response behavior. The

135
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

key factors were: i) the location of a trolley mounted on a girder (center and end
of a girder); ii) the level of a carrying weight with different rope length (top,
middle and bottom in height from the floor); and iii) rebounding characteristics
of a wheel system after its landing on the rail. Numerical analyses were also
performed to simulate the nonlinear response behavior observed in the element
model tests.

136
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Experimental determination of the

interaction of blast waves proceeding in
air and ground (5-2465)

Javed Iqbal
Directorate of Nuclear Power Engineering Structures
P.O. Box 3297 Islamabad, Pakistan, e-mail: [email protected]

Introduction
Gneiss is considered to have good foundation characteristics. The characteristic
feature of gneiss is its structure: the mineral grains are elongated, or platy, and
banding prevails. The paper deals with the experimental and numerical
evaluation of the influence of simultaneous ground shock and air blast forces on
structural response of containment shell structure founded on gneiss.

Aim of the work

A surface explosion generates both ground shock and airblast pressure on nearby
structures Wu et al. (2007). The ground shock usually arrives at a structure
foundation earlier than airblast pressure because of the different wave
propagation velocities in geomaterials and in the air. However, ground shock
and airblast might act on the structure simultaneously, depending on the distance
between the explosion center and the structure. Therefore, the precise analysis of
structure response and damage to a nearby surface explosion should take both
ground shock and airblast pressure into consideration. The current practice
usually considers only airblast pressure. Many empirical relations are available
to predict airblast pressure. Most of them, however, only predict peak pressure
values.

Essential results
In this experimental study, the experimental relationships of simultaneous
ground shock and airblast forces have been obtained which can be easily applied
in structural response analysis.
The numerical model including both free air and gneiss properties were
programmed and linked to sap2000 (2008) as its user provided subroutines. The
arrival time, peak particle acceleration (PPA), duration and the principal
frequency of the ground motion time history have been determined The
numerical and experimental scaled model results demonstrate good agreement

137
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

with each other. The full scale simulation of a typical reactor containment has
been subjected to surface explosions for structural response analysis. The
variation in stress values owing to time lag between air blast pressure and
ground shock to structure founded on gneiss, structural height and curvature has
been determined.

Conclusions
It is found that the structural damage will be critically underestimated owing to
neglect of simultaneous ground shock and air blast force. The methodology can
be employed to evaluate the blast response of concrete shell type containment
structure and estimating the extent of cracking.

References
Chengqing Wu, Hong Hao. 2007. Numerical simulation of structural response and
damage to simultaneous ground shock and airblast loads. International Journal
of impact Engineering. Vol. 34, pp. 556–572.

Structural Analysis Program, SAP2000 Advanced 12.0.1, 2008. Computers and Structures,
Inc. University Ave. Berkeley, CA 94704.

138
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Spectra-compatible time histories for the

ACR NPP in Eastern North America (5-2471)
G. Atkinson1, N. Allotey2, A. Saudy2, M. Elgohary2
1
Professor, Department of Earth Sciences, University of Western Ontario
Ontario, Canada
2
Atomic Energy of Canada Limited (AECL), Mississauga, Ontario, Canada

The standard design of the Advanced CANDU Reactor®, (ACR®), is developed

by Atomic Energy of Canada Limited to be the next step in the evolution of the
CANDU product line. It is based on the proven CANDU technology and
incorporates advanced design technologies.
In recent years, it has been established that ground motions occurring in
Eastern North America (ENA) are richer in high frequency content than those
occurring in Western North America. This is due to the impact of smaller
earthquakes at shorter distances that govern the computed seismic hazard of sites
in the Eastern North America region.
This phenomenon is particularly relevant for the seismic qualification of
safety related structures, systems and components of the nuclear power plant,
which are characterized by high frequencies. The ACR standard design takes
this phenomenon into account by using a uniform hazard spectra characterizing
the near-field motions in ENA, in addition to the ground response spectra
defined for soil and rock sites per the Canadian Standard Association.
Three sets of artificially generated time histories have been developed to
represent the ENA earthquake design spectrum for ACR Nuclear Power Plant.
Different techniques are applied to generate the three sets: spectral matching,
stochastic finite-fault simulation method and frequency-dependent scaling.
There are 38 time histories generated in total.
This paper presents the approach followed in generating the time histories
compatible with the ENA ground response spectra. In addition, the paper
presents the results of the analyses performed to ensure that the generated time
histories meet both Canadian and international acceptance criteria.

139
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Soil-structure analysis for ACR nuclear

island (5-2472)

N. Alloety, A. Saudy, M. Elgohary

Atomic Energy of Canada Limited (AECL), Mississauga, Ontario, Canada

The standard design of the Advanced CANDU Reactor®, (ACR®), is developed

by Atomic Energy of Canada Limited to be the next step in the evolution of the
CANDU product line. It is based on the proven CANDU technology and
incorporates advanced design technologies. The ACR nuclear island consists of
the reactor building and its adjacent safety-related auxiliary and service
buildings supported on a common base slab.
The ACR standard design is based on a standard uniform hazard spectra
characterizing the near-field motions in Eastern North America, and on two ground
response spectra defined for soil and rock sites per the Canadian Standard Association.
In addition, the standard design is performed for seven soil profiles, enveloping a wide
range of foundation conditions representing potential site conditions.
The nuclear island seismic response due to the three design ground response
spectra is determined, taking the soil-structure-interaction (SSI) effects into
account. The SSI effects are taken into account using the flexible volume
method in the frequency domain. A mathematical model representing the
dynamic characteristics of the nuclear island is developed. The foundation
medium is represented by horizontally layered soil profiles resting on a rigid half
space. The displacement and acceleration responses of the nuclear island
structures as well as the base shear and overturning moments are determined. In
addition, floor response spectra are generated at representative locations inside
the nuclear island structures.
This paper presents the approach adopted in determining the seismic response
of the ACR nuclear island structures due to the specified ground response
spectra while considering the SSI effects.

140
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Reactor head stand evaluation using

simplified non-linear analysis (5-2474)

Taha D. Al-Shawaf1, Lingyah Yen2, Kristin Murray Zaitz3

1
Material and Structural Analysis, AREVA NP Inc., Naperville, IL, USA
e-mail: [email protected]
2
Material and Structural Analysis, AREVA NP Inc., Naperville, IL, USA
e-mail: [email protected]
3
Design Engineer, Pacific Gas & Electric Co., Diablo Canyon Power Plant, CA, USA
e-mail: [email protected]

Introduction/background
Diablo Canyon Power Plant intends to replace the existing Reactor Vessel Closure
Head (RVCH) with a new RVCH and Integrated Head Assembly (IHA) that is
heavier than the existing configuration. During refueling, the new combined RVCH
and IHA structure is stored on a structure composed of four individual stands, 90
degrees apart. Since the head stands are only loaded during refueling outages, they
are not considered to be safety-related structures. The Diablo Canyon design criteria
allow such systems to have inelastic deformation under a seismic event if the behavior
does not adversely impact any safety related structures, systems or components.
Previously, each stand had been modified to raise the elevation of the stored
reactor head in order to allow for access to the bottom of the reactor head. The
modification is composed of steel members that are welded to each other creating
portal frames that are bolted to each of the original short storage stand legs. The
original storage stand legs are welded to concrete embedment plates. The portal
frames are not attached to the concrete slab and only act in bearing. The RVCH rests
on the top of stand and is connected to the stand through (a) shear pins to resist
horizontal forces and (b) bolts to resist vertical uplift. The total weight of the new
RVCH and IHA is heavier than the original assembly, and the available design
margin in the existing analysis is limited. Therefore, conservatisms in the existing
evaluation must be reduced to avoid structural modifications.

Objective
Several analysis techniques are presented to increase the design margins in order
to qualify an existing structure without the need for structural modifications.
These methods include developing higher damping seismic response spectra,
inelastic spectra generation, decoupling of non-linear systems, simplified non-
linear modeling and the use of the 100-40-40 combination method.

141
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Approach
One of the conservatisms in the existing seismic evaluation was the use of 4%
damping Hosgri response spectra because no higher damping value response
spectra were available. NRC Regulatory Guide 1.61 (1973) permits the use of
7% damping for bolted structures under Safe Shutdown Earthquake. The 7%
damping is more appropriate since the reactor head is bolted to the stand, and
each head stand also has bolted connections. Therefore, a simplified
methodology is developed to generate the 7% damping response spectra based
on the available 4% damping response spectra using the relationship given in
Newmark (1971). The seismic response spectra are further reduced in the low
and intermediate frequency ranges by assuming a limited nonlinear (elasto-
plastic) behavior exists in the structural system. This method was proposed by
Newmark (1982) and is also discussed by Gupta (1992).
The RVCH head stand exhibits non-linear characteristic at the interfaces between
the RVCH and head stand, and between the head stand portal frame and concrete
slab. The non-linear characteristic is due to the different stiffness when the
connection is in compression or tension. A methodology is developed to account for
the different compression and tension stiffness to better predict the behavior of the
structure. This method is then simplified by using a decoupling technique to separate
the structure into two models at the interface of RVCH and head stand.
The reduced inelastic seismic response spectra is used in finding the reactions
for an ANSYS stick model of the reactor head, CRDM, IHA, and the bolts
between the head and the stand (noted as Model 1 in Figure 1). The non-linearity
of Model 1 is due to reactor head-to-stand bolt stiffness in tension being smaller
than the head-to-stand bearing compression stiffness. The reactions at the base
of the ANSYS stick model are combined using the 100-40-40 method. Unlike
the Square Root of the Sum of the Squares (SRSS), the 100-40-40 method
maintains the sign of the reaction forces. This allows the analysts to distinguish
between the different structural behavior, i.e., in tension and compression. As
such the 100-40-40 reduces conservatism by qualifying members based on their
actual response in tension or compression. The reactions of controlling
combinations from Model 1 are then applied statically on a 3-D non-linear
model of all the support frames comprising the stand structure (noted as Model 2
in Figure 1). This model depicts the behavior of the system and distributes the
loads between the stand frames. The non-linearity of Model 2 is due to the fact
that the portal frame (W14) is not attached to the concrete slab (i.e., act in
bearing only). The resulting forces and stresses from Model 2 are evaluated
using the design basis allowable criteria and code of record, AISC (1989), for
acceptance.

142
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Figure 1. Reactor Vessel Closure Head (RVCH) Stand.

Conclusions
The existing head stand structure was qualified without the need for structural
modification using simplified non-linear analysis techniques. These techniques
include developing higher damping spectra, inelastic spectra, decoupling of the
structure, simplified non-linear model, and the 100-40-40 seismic combination
methods.

143
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

References
American Institute of Steel Construction (AISC), 9th Edition. Manual of Steel Construction,
Allowable Stress Design.

Gupta, A.K. 1992. Response Spectrum Method in Seismic Analysis and Design of
Structures. CRC Press.

Newmark, N.M., Rosenblueth, E. 1971. Fundamentals of Earthquake Engineering. Prentice-

Hall, Inc.

Newmark & Hall, 1982. Earthquake Spectra and Design. Earthquake Engineering Research
Institute.

U.S. Nuclear Regulatory Commission, Regulatory Guide 1.61, October 1973. Damping
Values for Seismic Design of Nuclear Power Plants.

144
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

NPP seismic protection against shock and

vibration loads (5-2479)

V. Belyaev1, V. Guskov2, Y. Routman3

1
Professor, Research Center for Capital Construction, Saint-Petersburg
2
Dr., JSC Bureau of Special Mechanical Engineering, Saint-Petersburg
3
Professor, JSC Bureau of Special Mechanical Engineering, Saint-Petersburg

The paper presents the results of theoretical and experimental study of a seismic
isolation system based on 3D elastic-plastic dampers (3D EPD) designed for
overall protection of nuclear power plants against seismic, shock and vibration
loads.
Choice of the material for 3D EPD elements, their sections and spatial
configurations provides for standard level of load transfer to the object to be
protected, operational stability of 3D EPD at a given number of loading cycles.
Using 3D EPD makes it possible to create a compact system for the object
protection against active spatial impacts. Together with the above specific
features 3D EPD preserves all good properties of unidirectional plastic dampers:
they preserve their characteristics irrespective of the environment conditions,
their long-duration operation does not require maintenance.
Designing of the 3D EPD depends on the initial 3D EPD yield surface, which
is to be found as the solution of rigid-plastic problems for the 3D EPD strength
model. A new method for these problems solving had been developed. When the
initial 3D EPD yield surface is defined we can evaluate the dynamic movement
of the object protected and the loads acting under given external impacts.
More exact solution for the problem of seismic protection based on EPD
requires to consider the following: reinforcement effect, loading recurrence
(recurrent reinforcement, low-cycle fatigue), elastic operation of nor-plastic-
deformed segments of 3D EPD elements and their cross-sections, elastic
operation of 3D EPD elements when the load has been removed, plastic
deformation influence on changes in the 3D EPD elements geometric parameters.
Thus the 3D EPD types proposed had passed the whole cycle of development:
calculation, dynamic and static tests of elements and pilot samples. Therefore
using the EPD for protection against extreme loads is effective.

145
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

The gearbox for the helium cycle of

10 MW high temperature gas-cooled
reactor (5-2508)

Lushuai Wang, Suyuan YU, Shutang Zhu, Xuanyu Sheng

Institute of Nuclear and New Energy Technology, Tsinghua University
Beijing 100084, China

The direct helium cycle has much high efficiency to generate the electricity for
the power conversion unit of high temperature gas-cooled reactor (HTGR). In
order to validate the helium cycle technology, INET initiates a project to couple
a helium cycle with 10 MW high temperature gas-cooled reactor (HTR-10) to
replace current steam generator. Such a cycle consists of a turbine-compressor
system and a power generator. While the speed for the turbine-compressor
system is 15000 rpm and the speed of the power generator is 3000 rpm. A
gearbox is chosen to connect the turbine-compressor system and the power
generator. The paper studies the special features of the HTR-10 helium cycle as
well as the current gearbox technology worldwide, provides the technical
requirements for the gearbox, and then makes the preliminary design for the
gearbox based on the structural mechanical analysis. The center distance
between the turbine shaft and the generator shaft is 300 mm. An vertical gearbox
with a speed ratio of approximately 5:1 is selected to connect the helium turbine-
compressor system and the power generator to transfer the power of 2500 KW.
The mechanical analytical results show that the gearbox design successful
satisfies the technical requirements and the specification.

146
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Reactor building 3D-model for evaluating

the pressures on concrete regularization
and foundation waterproofing
membrane (5-2514)

Glauco J.T. Mello Junior, Tarcísio de F. Cardoso, Carlos L.M. Prates

GAN.T, Eletronuclear
Rua da Candelária, 65, Rio de Janeiro, CEP 20091-906, Brasil
e-mails: [email protected], [email protected],
[email protected]

Introduction
Angra dos Reis site in Brazil has already 2 operating PWR NPPs. Unit 3, with
identical design to Unit 2, also a 1350 MW PWR, is expected to have its
construction started in 2009. The reactor building of Angra 3 is a complex
concrete structure with several thickness and dimensions. In a general point of
view it is founded on a base plate having a thickness of about 2 m and radius
about 30 m. The top level of this foundation is -0.85 m and the ground level is
6.15 m. This new plant shall be founded directly on a hard sound rock. The first
step is to prepare this rock surface with a concrete regularization and a
foundation waterproofing membrane.

Aim of the work

This study presents a 3D model approach of the corresponding reactor building
to verify the maximum loads acting on this surface. ANSYS Mechanical Release
11.1 is used for this analysis.
Dead weight, permanent and live loads, Safe Shutdown Earthquake (SSE)
combined with Burst Pressure Wave (BPW) from the Feedwater Tank and
temperature are taken into account.
Foundation stiffnesses corresponding to the hard sound rock and to the
waterproofing membrane are represented separately by elements of linear elastic
spring COMBIN14. Stepped foundation is also represented by additional
springs. The springs are in the three orthogonal directions. Rock stiffnesses are
obtained from elastic half-space theory and membrane stiffness is based on test
results. The superstructure is represented by linear shell elements – SHELL63.
The maximum pressure for each load case is obtained for vertical and
horizontal directions that correspond to compression and tangential reaction
loads.

147
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Results and conclusions

The results are compared with a more simplified analysis performed before,
showing a good agreement in global values.
The 3D model permits to show a more realistic pressure distribution at every
foundation specific detail, proving that the results obtained of combined loads,
1192 kN/m2 for compression, 179 kN/m2 for shear effects, are bellow the allowable
limits.

148
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

On the generation of inelastic secondary

system seismic response spectra (5-2526)

Tarcísio de F. Cardoso1, Andreia A. Diniz de Almeida2, João L. Roehl2

1
GAN.T, ELETRONUCLEAR, Rua da Candelária, 65, Rio de Janeiro,
CEP 20091-906, Brasil, e-mail: [email protected]
2
DEC, PUC Rio, Rua Marquês de São Vicente, 225. Rio de Janeiro,
CEP 22451-041, Brasil, e-mails: [email protected], [email protected]

Introduction
In the electric power reactor industry all safety related systems are designed to
resist and to keep the operability during and after a postulated earthquake. The
diversity and the large number of the secondary systems in a NPP lead to the
response spectra methodology for the seismic analysis.
In general the secondary system seismic design is based on floor response
spectra, using the assumption of linear analysis, although, it is useful to evaluate
the plastic reserve due to secondary system ductility.
For piping systems, the evaluation of this reserve can be performed by the
comparison of the linear response spectra to the nonlinear inelastic ones obtained
on particular models of such systems.

Aim of the work

It is suggested a procedure for the generation of in-structure seismic response
spectra for secondary system design, which includes a probabilistic approach
and considers coupling effects between primary and inelastic secondary systems.
First, the ground excitation PSD is transferred to a SDOF model of the
secondary system conveniently attached to the primary system. Then, a
uniformly probable coupled response spectrum is obtained using the first
passage analysis, /5/.
A global ductility factor, relating the plastic to the overall work done by the
seismic external forces on the secondary system, and a specified yielding factor
allows one to obtain transposition factors from elastic to inelastic response
spectra.

Summary of the methodology

The proposed script is developed in frequency domain. A set of computer
programs is developed to be used with the SASSI2000 system modules to

149
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

consider three-dimensional models and their responses for a generic base

excitation, acting in 3 orthogonal directions.
The methodology includes the representation of coupling effect between
primary and secondary systems, and the influence of the secondary system
multiple supports relative displacements, to produce uniformly probable coupled
response spectra, i.e., its ordinates represent the maximum response of a single
degree of freedom (SDOF) system with equal probability of non exceedance
along the usual frequency range of analysis
With the use of the mentioned transposition factors, it is possible to evaluate
the plastic reserve of the secondary piping systems.
The application of this methodology is a practical and consistent solution. It is
practical because it only needs the earthquake PSDF and consistent since it may
produce also coupled response spectra whose ordinates maintain a probabilistic
commitment with the PSDF amplitudes.

Conclusions
Although the great computational efforts, with the proposed methodology one
can achieve the following advantageous:
− Better representation of the damping effects, considered directly in the
soil-structure interaction formulation, because neither the use of modal
damping nor the definition of the Rayleigh coefficients are necessary.
− The choice of the frequency for which the response spectrum is calculated
is oriented by the Transfer Function peak values. It requires a lower
number of calculations points than if all modal frequencies are used
besides that 75 specified ones [US NRC- RG 1.122 – 1978].
− Superposition and combination of different responses can be obtained in
probabilistic ways.
− Probabilistic response spectra, obtained directly from PSD, are much
smoother than those obtained deterministically from time history samples,
and the errors can be evaluated, leading to more analysis reliability.
− Evaluation of the plastic reserve of secondary piping systems is achieved,
by the use of transposition factors.
− The degree of automation, allows the production of response spectra
including modeling refinements, reaching a more realistic analysis,
without additional efforts beyond those already required by the usual
methodology.

150
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

References
1. Almeida, A.D. de. Análise probabilística de segurança de sistemas e componentes
estruturais. 2002. Tese (Doutorado) – Pontifícia Universidade Católica do Rio
de Janeiro, Rio de Janeiro, 2002.

2. Almeida, A.D. de., Roehl, J.L. Uniformly Probable Coupled Response Spectra for
Secondary Systems. Transactions of the 17th International Conference on
Structural Mechanics in Reactor Technology (SMIRT 17). Praga, Agosto 2003.

3. Cardoso, T. de F. Visão geral sobre espectros de resposta sísmica para sistemas

secundários. 2008. Dissertação – Pontifícia Universidade Católica do Rio de
Janeiro, Rio de Janeiro, 2008.

4. SASSI2000 A System for Analysis of Soil-Structure Interaction. Rev. 1 (1999). J.

Lysmer, F. Ostadan, M. Tabatabaie, F. Tajirian, S. Vahdani. Department of Civil
and Environmental Engineering – University of California, Berkeley.

5. Vanmarke, E.H. On the distribution of the First-Passage Time for Normal Stationary
Random Processes. 1975. Journal of Applied Mechanics, 42, pp. 215–220.

151
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Seismic response of a two-degree-of-

freedom system with friction based on
the mass ratio (5-2542)

Akinori Tomoda, Tetsuya Watanabe, Kihachiro Tanaka

Division of Mechanical Engineering and Science, Saitama University
255 Shimo-Okubo, Sakura-ku, Saitama 338-0825, Japan
e-mail: [email protected]

Introduction
Recently, massive earthquakes such as the Niigata-ken Chuetsu-oki earthquake
(2007) or the Tokachi-oki earthquake (2003) have occurred frequently in Japan.
Large-magnitude earthquakes seriously damage not only ordinary houses, but
also structures in industrial facilities. In order to prevent hazardous material
spills and secondary disasters, industrial facilities are required to have high
aseismic performance compared with houses.
On the other hand, the frictional isolator has attracted attention for the
reduction of seismic response in industrial facilities. This system has the effects
of shifting the natural frequency away from the predominant frequency of the
seismic wave and dissipating seismic energy. These effects can improve the
aseismic capacity of a structure in industrial facilities and reduce the cost of
seismic design. When a structure is a multi-dof system with friction, however,
the seismic response of this system can be obtained by only non-linear time
history analysis [1]. A great deal of time is necessary for non-linear analysis. In
general, immediate and easy estimation of the seismic response is required in the
seismic design of industrial facilities. In the previous studies [2]–[4], the authors
proposed a response spectrum for a 1-dof system with friction. This spectrum
can simply estimate the seismic response of a 1-dof system with friction.

Aim of the work

The present study deals with the easy estimation of seismic response for a multi-
dof system with friction using the response spectrums method. The 1-dof linear
system and the 1-dof friction system are used in this method. This estimation is
based on the modal analysis method. The authors also proposed a method of
modal separation to use one slip direction for a piping system with friction [5].
In addition, Hanawa et al. proposed a method that is available for any slip
direction [6]. When we consider the simple 2-dof system with friction, however,

152
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

the method of modal separation for same directional friction force has not yet
been proposed.
In the present paper, the method of modal separation for a 2-dof system with
friction is proposed when the mass ratio of the structure and the support is nearly
zero. The 1-dof system, which is converted to modal space using this method,
calculates the time history response. The time history response on modal space is
then translated to real coordinates. An application of the proposed method is
discussed by comparing the results of modal analysis for a friction system with
the 2-dof non-linear time history analysis.

Essential results
This paper show the one mode of the vector of the friction force can be regarded
as approximately zero when the mass ratio is nearly zero. However, the accuracy
of the proposed method depends on the mass ratio, the frequency ratio and the
friction force. The proposed method is compared with 2-dof non-linear time
history analysis, and the accuracy of the proposed method is calculated using the
sinusoidal wave and seismic waves. The present study defines the area within an
error of 10% as the applicable area of the modal analysis method for a friction
system and shows the range of error over 10%. From this area, we can judge
whether the proposed method is applicable for each parameter.

Conclusions
In the present study, the method of modal separation for a 2-dof friction system
using the ratio of elements of the modal matrix is considered. The proposed
approximation method can convert dynamic equations of a 2-dof friction system
into modal space. Analytical errors of the maximum acceleration at the structure
and support obtained by the proposed method are calculated and compared with
the 2-dof non-linear time history analysis. The results of the present study
revealed that an area of error of over 10% for the modal analysis method for a 2-
dof friction system. Therefore, the range of application of the frequency ratio
and the mass ratio can be easily estimated using this error area.

References
1. Marui, E., Kato, S. Forced vibration with solid friction, Trans. of JSME, Series C, Vol. 49,
No. 443(1983), pp. 1146–1152.

2. Watanabe, T., Suzuki, K. Seismic Response of Piping System with Friction Sliding on a
Plane, Trans. of JSME, Series C, Vol. 64, No. 617(1998), pp. 64–69.

3. Watanabe, T., Suzuki, K. Estimation of Equivalent Damping Ratio using Friction Response
Spectrum, Trans. of JSME, Series C, Vol. 66, No. 642(2000), pp. 87–92.

153
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

4. Tomoda, A., Watanabe, T., Tanaka, K. Study on Friction Response Spectrum Under Long
Period System, Trans. of JSME, Series C, Vol. 73, No. 730, (2007), pp. 1699–1704.

5. Watanabe, T., Suzuki, K. Seismic Response Analysis of Piping System with Friction
Support (Maximum Response Estimation Using Frictional Response Spectrum),
Trans. of JSME, Series C, Vol. 64, No. 619, (1998), pp. 87–92.

6. Hanawa, Y. et al. Vibration Test and Analysis of Response of Piping System with Friction
Support, Trans. of JSME, Series C, Vol. 64, No. 640, (1999), pp. 4611–4617.

154
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Response of graphite dowel-socket

structure under various loads (5-2564)

Han Fengshan, Sun Libin

Institute of Nuclear and New Energy Technology, Tsinghua University,
Room 310, Energy Source Building B, Beijing, 100084, P. R. China
e-mail: [email protected], [email protected]

Keywords: HTR, graphite components, dowel-socket, equivalent stiffness

The graphite components are the main part of the core structures of high
temperature gas-cooled reactor (HTR). Unlike metal components which can be
welded or riveted together, those graphite components should be only connected
by dowel-socket and key-keyway structures to transfer horizontal forces and to
restrict correspondingly horizontal motion, therefore, to ensure the integrity of
the whole core structure.
Among the assembly of graphite components, small clearance remains to
accommodate the thermal and fast neutron irradiation strains. Under dynamic
loads, such as seismic loads, the impact would occur among those graphite
components due to the given clearance. Therefore, in addition to the research of
the overall response of graphite dowel-socket structure under various loads, it’s
necessary to study the influence of different gaps of dowel-socket structure on
the load-deformation response of the graphite structure. Since the final goal is to
investigate the integrity of the whole core structure, it is necessary and important to
get the equivalent stiffness of dowel-socket structure that consists of two graphite
blocks, which is considered as the basic component of the whole core structure.
In this paper, with commercial finite element code ANSYS, the load-
deformation response of the graphite dowel-socket structure is studied, and the
influence of different gaps of dowel-socket structure on the response is discussed
in addition. Besides, the corresponding results and equivalent stiffness of dowel-
socket structure are given.

155
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Internal pressure capacity evaluation of

prestressed concrete containment
buildings considering multiple
aging effects (5-2580)

Daegi Hahm1, In-Kil Choi2, Hong-Pyo Lee3

1
Faculty of Korea Atomic Energy Research Institute
1045 Daedeok-daero, Yuseong-gu, Daejeon, Korea, e-mail: [email protected]
2
Faculty of Korea Atomic Energy Research Institute
1045 Daedeok-daero, Yuseong-gu, Daejeon, Korea, e-mail: [email protected]
2
Faculty of Korea Electric Power Research Institute
65 Munji-ro, Yuseong-gu, Daejeon, Korea, e-mail: [email protected]

Introduction
Since the accident at Three Mile Island nuclear plant in 1979, it has become
necessary to evaluate the internal pressure capacity of the containment buildings
for the assessment of the safety of nuclear power plants [1–3]. According to this
necessity, many researchers including Yonezawa et al. [4] and Hu & Lin [5]
analyzed the ultimate capacity of prestressed concrete containments subjected to
internal pressure. In these studies, the ultimate capacity analyses are performed
for the containments under fresh condition. However, most of nuclear power
plants are exposed to the severe environments such as costal area and ambiance
irradiation. Hence, the aging effects on the structural system caused by these
environmental conditions should be considered for the estimation of internal
pressure capacity in a quantitative manner. Especially in Korea, some
containment structures were built in the late of 1970 or early 1980, so that the
degradation of their structural performance also must be explained in the
procedure of the of the internal pressure capacity evaluation.
Therefore, in this study, we developed the degradation models for the structural
components of prestressed concrete containment buildings, and evaluated the
internal pressure capacity considering multiple aging and degradation factors.
The target containment building types were PWR (Pressurized Water Reactor)
and CANDU (CANada Deuterium Uranium) type containments which are the
most typical reactor buildings in Korea.

Methodologies and results

There exist many degradation and aging factors in the prestressed concrete
structures. For the concrete material, the most degradation factor can be

156
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

classified into physical processes such as cracking, freezing & thawing,

irradiation, fatigue, settlement, and chemical processes such as sulfate &
biological attack, acids, aggressive water, etc. In general, these degradation
factors finally cause a loss of section and tensile/compression strength of
containment wall. We modeled such aging effect and considered for the internal
pressure capacity estimation. In the prestressed concrete containment building,
the loss of prestressing force is recognized another important aging factor.
Hence, we also estimated the amount of prestressing loss for the evaluation.
For the modeling of containment buildings, there exists a burden process
on the modeling of tendons at the hatch area since that its geometry is somewhat
complicated. In this study, we developed a program which can automatically
generate the tendon components considering the hatch areas. The FE-based
general-purpose structural analysis program, ABAQUS [6], was adopted as an
analysis tool and developed models were implemented into the input files. We
modeled PWR and CANDU type containments as 3D FE models. For the
modeling of containment wall, dome, buttress and slab, solid elements were
used. The reinforcement bars and tendons were modeled using embedded
surface and truss elements, respectively. The material nonlinearity of concrete
was implemented by introducing the concrete damaged plasticity model [6]. The
tri-linear plasticity model and piecewise linear stress-strain model were used for
the material nonlinearity of steel rebars and tendons, respectively. In the
preliminary analysis, the critical points under an internal pressure load were a
top of the hatch and a middle of the containment wall areas. Figure 1 shows one
of the results for the ultimate internal pressure capacity evaluation for the PWR
type containment. The elastic capacity almost linearly decreased as the
prestressing loss increased, while it does not vary during the degradation of
concrete strength. It can be also found that the loss of concrete strength effect
only to the elastic stiffness of containment walls.
120
120

100 100

80 80
Pressure (psi)
Pressure (psi)

60 60

40 40
Fresh condition Fresh condition
Prestressing loss 20% Concrete Elasticity 80%
20 20
Prestressing loss 40% Concrete Elasticity 60%
Prestressing loss 60% Concrete Elasticity 40%
0 0
0 1 2 3 4 5 0 1 2 3 4 5 6
Displacement (in) Displacement (in)

Figure 1. Internal pressure capacity evaluation results considering the prestressing loss
and the degradation of concrete strength.

157
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Conclusions
We developed the degradation models for the structural components of prestressed
concrete containment buildings, and evaluated the internal pressure capacity
considering multiple aging and degradation factors. The FE-based general-
purpose structural analysis program, ABAQUS was adopted as an analysis tool
and developed full-3D nonlinear FE models were implemented. The results
show that the elastic capacity almost linearly decreased as the prestressing loss
increased, while it does not vary during the degradation of concrete strength.
More detailed results for the various multiple aging effects will be discussed in
the full paper.

References
1. US Nuclear Regulatory Commission, Office of nuclear reactor regulation. Standard
review plan of the safety analysis reports for nuclear plants, Section 3.8.1,
NUREG-0800; 1987.

2. Amin, M., Eberhardt, A.C., Erler, B.A. Design considerations for concrete containments
under severe accident loads. Nucl Eng Des 1993;145:331–338.

3. Boeck, B.D. A review of containment accidents. Nucl Eng Des 1993;145:279–288.

4. Yonezawa, K., Imoto, K., Watanabe, Y., Akimoto, M. Ultimate capacity analysis of 1/4
PCCV model subjected to internal pressure. Nucl Eng Des 2002;212:357–379.

5. Hu, H.T., Lin, Y.H. Ultimate analysis of PWR prestressed concrete containment subjected
to internal pressure. Int J Pres Ves Pip 2006;83:161–167.

6. ABAQUS/Standard 6.8 – User’s Manual, 2008, Hibbitt, Karlsson & Sorensen, Inc.

158
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Qualification seismic test on control rod

driving mechanism of CEFR (5-2588)

Jing Wen1, Lei Sun2, Hongyi Yang1, Xuede Chen2, Hailong Li1,
Qing Song1, Xiaoxuan Li1, Jiang Qian3
1
China Fast Reactor Research Center, China Institute of Atomic Energy
Beijing, China, e-mail: [email protected]
2
Structural Mechanics Division, Nuclear Power Institute of China
Chengdu, China, e-mail:[email protected]
3
Institute of Structural Engineering and Disaster Reduction, Tongji University
Shanghai, China, e-mail: [email protected]

Background
China Experimental Fast Reactor (CEFR) is the first fast neutron breeder reactor
in China. The Control Rod Driving Mechanism (CRDM) of CEFR was designed
and fabricated by Russian engineers. The CRDM has been qualified strictly by
tests according to Russian nuclear codes, including seismic test. Because of the
difference between Russian codes at that time and Chinese codes or international
nuclear codes, the CRDM was qualified by seismic test according to the Chinese
codes and international codes. The latter seismic test is introduced in this paper.

Aim of the work

The control rod assemblies (CR) of CEFR consist of three types: Safety
Assembly (SR), Compensatory Assembly (CA) and Regulatory Assembly (RA).
The CRDM is the equipment of nuclear safety class 1 and seismic qualification
of class I. The CR should be inserted in the core during and after the earthquake
to make sure that the reactor be shut down safely. The factors that influence the
inserting are very complicated. The CRDM is excited by not only one point but
four points during the earthquake. The moving component of the CRDM
impacts on the inner wall of the guiding pipe. The most part of the CRDM is
immerged in the liquid sodium and fluid structure interaction (FSI) occurs
during the earthquake. The test is performed to validate that the CA can be
inserted in the core during and after the earthquake to shut down the reactor
safely.

159
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Results and conclusions

The qualification seismic test had been performed according to the Chinese
nuclear codes. The original CRDM and full scale CRs was adopted and the
liquid sodium was substituted by water at room temperature in the test. The
earthquake was simulated by multipoint (4 points) displacement time history
excitation. The 5 tests on Operation Based Earthquake (OBE) and one test on
Safety Shutdown Earthquake (SSE) had been finished for every type of CR. The
CRs were validated to be inserted in the core in the time during and after the
earthquake. The longest time to inserted in the core for SA is 0.687 second
(shorter than 0.7 second which is designed) and for CA-RA 1.789 seconds
(shorter than 2.5 seconds which is designed). And no rebound occurred. The test
improved that the seismic design of the CRDM was certified and valid.

References
1. HAF J0053, Guideline of seismic qualification test on nuclear equipments, 1998.

2. HAD102/02, Seismic design and qualification for nuclear power generating stations.

3. IEEE 323, IEEE Standard for Qualifying Class 1E Equipment for Nuclear Power Generating
Station, 2003.

4. IEEE 344, IEEE Recommended Practice for Seismic Qualification of Class 1E Equipment
for Nuclear Power Generating Stations, 2004.

160
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Seismic assessment of the sellafield

B38 mobile caves (5-2615)

Warren Price BEng (Hons), CEng, FIMechE

Lead Technologist Structural & Seismic Modelling
National Nuclear Laboratory, Warrington, UK, e-mail: [email protected]

Introduction
The B38 building at Sellafield originally consisted of six concrete silos with an
overbuilding, and was commissioned in 1964. The first extension, a further six
silos, was commissioned in 1974. A further building extension gave a total of
twenty two silos. The silos were used for the storage of Magnox swarf, and
miscellaneous beta gamma waste under water cover.
In time, storage of waste in B38 ceased and the plant lay dormant for many
years. In the early 1990’s concept designs facilitating the removal of waste from
the silos, dispatching this to downstream plants started to be looked at by BNFL.
This lead to the design of retrieval machines known as Mobile Caves, of which
there are three. These would locate over a silo, remove its roof plug, and by
means of grab, tools and manipulators, remove waste. The waste would be
placed into a skip internal to the Mobile Cave, which would then be removed
through a gamma gate into a 50 Te flask temporarily located on the Cave. The
flask would then be removed by building cranes for further downstream
treatment. Despite being only moderately large in size (approximate envelope of
12 m × 5 m × 6 m), a Mobile Cave has a design mass approaching 400 Te – the
equivalent mass of a 747-400 series jumbo jet fully laden at take off (Ref 1).
In the mid 1990’s seismic analysis work was completed however, in time the
Mobile Cave design evolved significantly, and after a period of mothballing of
the project, it became clear that new seismic qualification would be required.
Hence, the National Nuclear Laboratory, set up during 2008 as part of the
restructuring of BNFL, was requested to provide seismic analysis support to the
project.

Seismic analysis
1. Building structure
A finite element model of the building structure, concrete silos, and rails along
which the Mobile Caves run was utilized. In this model, the Mobile Cave was
represented as a rigid “brick” with the same mass and centre of gravity location
as the Mobile Cave itself. The response spectrum and time history analysis

161
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

output from this work include the three orthogonal seismic accelerations at the
Mobile Cave centre of gravity, enveloped for multiple retrieval locations. The
above provided the inputs to the work described in this paper, which was the
actual detailed qualification of the Cave structure, described below.

2. Mobile cave structure

A detailed finite element model of the Mobile Cave was built using the
proprietary code Ansys (Ref 2). The main structure was modeled using 3D solid
elements, but where appropriate 3D shell, 3D beam, and 3D mass elements were
also utilized. To ensure that the effect of prying is included in the bolt load
analysis, Ansys 3D surface-to-surface contact elements were used between
mating faces. In all locations where a bolted connection is made, co-incident
nodes have been positioned in the model, and these are coupled in the three
translational degrees of freedom to simulate bolt load transfer. By isolating the
various assemblies from the rest of the model in post-processing runs, the bolt
loads and hence bolt stresses were determined.
A separate finite element model of the seismic isolation bearings (SIBs) was
used to confirm that the properties used in the main model were appropriate.
Hand calculations determined appropriate section and material properties
required in the model in order to replicate the known stiffness properties of the
SIBs. Before inserting these calculated values into the main structural model,
simulation of the SIBs in a separate finite element model was used to confirm
that the calculated properties correctly replicated the SIB stiffness
Allowable stresses were specified using ASME III (Ref 3). Defining
allowables for the Mobile Caves in terms of ASME III is difficult, since the
primary intent of ASME III is to assess pressure retaining equipment, piping and
supports. The Mobile Cave structure is neither a pressure retaining structure, nor
a support. The use of ASME III was however, retained in the assessment as there
are parts of the code that can be used in principle to evaluate performance of
both the main structural items, and the bolts holding them together.

Results
The finite element model was used to successfully demonstrate that all key
structural items had stress levels within permissible values. The attachment bolts
and fabrication welds were likewise demonstrated to be acceptable. Due to the
modeling & assessment work described in this paper, the safety case seismic
performance requirements of the Mobile Caves was satisfied. The Mobile Cave
is expected to be commissioned and operational by 2011.

162
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

References
1. Boeing website http://www.boeing.com/commercial/747family.

2. ANSYS Inc, Southpointe, 275 Technology Drive, Canonsburg, PA 15317, USA.

3. American Society of Mechanical Engineers Boiler and Pressure Vessel Code, Rules for
Construction of Nuclear Power Plant Components. Section III.

163
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Response and seismic margin of

Kashiwazaki Kariwa Nuclear Power Plant
building to Chuetsu-oki earthquake (5-3192)

Atsushi Suzuki4, Rikiro Kikuchi1, Takayuki Koyanagi2,Minoru Kanechika3,

Yoshinori Mihara5
1
Nuclear Asset Management Department, Tokyo Electric Power
Company, 1-1-3, Uchisaiwai-cho, Chiyoda-ku, Tokyo 100-8560, Japan,
e-mail: [email protected]
2
Nuclear Asset Management Department, Tokyo Electric Power
Company, 1-1-3, Uchisaiwai-cho, Chiyoda-ku, Tokyo 100-8560, Japan,
e-mail: [email protected]
3
Nuclear Power Department, Kajima Corporation, 6-5-11, Akasaka,
Minato-ku, Tokyo 107-8348, Japan, e-mail: [email protected]
4
Nuclear Power Department, Kajima Corporation, 6-5-11, Akasaka,
Minato-ku, Tokyo 107-8348, Japan, e-mail: [email protected]
5
Nuclear Power Department, Kajima Corporation, 6-5-11, Akasaka,
Minato-ku, Tokyo 107-8348, Japan, e-mail: [email protected]

1. Introduction
The maximum earthquake acceleration observed in the reactor buildings of
Kashizaki Kariwa Nuclear Power Plant during the Niigata-ken Chuetsu-oki
earthquake in 2007 exceeded the design value. However, the structural members
were generally within the elastic range, so it was considered that their integrity
was ensured. This report discusses the causes of the above by comparing the
design force with the actual seismic force by the Chuetsu-oki earthquake.

2. Structure of nuclear reactor building

Reactor building Unit 6 was investigated. Its major structure is of reinforced
concrete, with four stories above ground and three stories under ground. The
height from the base mat is 63.4m and that from ground level is 37.7m. It is of
almost square in plan, measuring 56.6m (NS) x 59.6m (EW).
Its main earthquake resisting walls are the external box type wall and interior
walls, and it is designed as a reinforced concrete containment vessel (RCCV).
The seismic walls resisted the horizontal force that acted on the building during
the earthquake.

164
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

3. Design seismic load of nuclear reactor building

The design dynamic and static seismic forces were both three times larger than
those required by the Construction Standard Act in Japan. The dynamic seismic
force was calculated from earthquake response analysis using the reference
earthquake motion S1-D (195 Gal on the base mat).
The design seismic force ensured a tolerance by adding a safety margin to both
the dynamic and static seismic forces. This margin resultantly became within a
range in which the effects of burying into the ground, taken into account in the
static seismic force calculation, could be ignored.

4. Comparison with seismic force due to Chetsu-

oki earthquake
The seismic force acting to nuclear reactor building Unit 6 during the Chuetsu-
oki earthquake (320 Gal at the base mat) was estimated from an elastic seismic
response analysis using the observation record. The maximum shear stress
during the Chuetsu-oki earthquake became larger than S1-D. However, it was at
the same level or lower than the static seismic force, and it was sufficiently
lower than the design seismic force.
The design seismic forces on the building’s structural members were less than
the allowable forces which have the members within elastic range. Because the
seismic force was lower than the design seismic force during the Chuetsu-oki
earthquake, the building was considered to be within the elastic range.
It was considered that an earthquake ground motion larger than the reference
earthquakes ground motion S1-D and S2-D, had acted. However, the margin
adopted for the static / design seismic force would have maintained the
building’s integrity.

5. Response on shear skeleton curve

The seismic force caused by the Chuetsu-oki earthquake was estimated from
elastic analysis. However, attempts were also made to estimate it from elasto-
plastic analysis. The nonlinear characteristics (skeleton curve) of shear force of
the earthquake resisting wall was determined by the method specified in the
guideline for seismic design of nuclear power plants JEAG4601. The conditions
assumed for the above were the same as those for the elastic analysis. The
results show that the response to the Chuetsu-oki earthquake was below the first
turning point (concrete crack onset point) and was considered to be within the
elastic range.

165
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

6. Conclusions
The response on the base mat showed that the earthquake ground motion caused
by the Chuetsu-oki earthquake at the reactor building Unit 6 was larger than the
design force. However, the structural members of the building were considered
to be within the elastic range.
This is considered to be because a static earthquake ground motion three times
that of ordinary buildings was assumed and a tolerance was incorporated into the
design seismic force. Furthermore, the effect of the tolerance was taken into
account in designing the reinforcing bar arrangement. Thus, the behavior was
considered to be within the elastic range. The above is consistent with the
relationship between the hysteresis characteristics and the response.

166
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

Simulation analysis of reactor buildings on

Niigataken Chuetsu-oki earthquake at
Kashiwazaki-Kariwa Nuclear Power Plant
(5-3193)

Takayuki Koyanagi1, Rikiro Kikuchi2, Atsushi Suzuki3, Minoru Kanechika4

1
Nuclear Asset Management Department, Tokyo Electric Power
Company, 1-1-3, Uchisaiwai-cho, Chiyoda-ku, Tokyo 100-8560, Japan,
e-mail: [email protected]
2
Nuclear Asset Management Department, Tokyo Electric Power
Company, 1-1-3, Uchisaiwai-cho, Chiyoda-ku, Tokyo 100-8560, Japan,
e-mail: [email protected]
3
Nuclear Power Department, Kajima Corporation, 6-5-11, Akasaka,
Minato-ku, Tokyo 107-8348, Japan, e-mail: [email protected]
4
Nuclear Power Department, Kajima Corporation, 6-5-11, Akasaka,
Minato-ku, Tokyo 107-8348, Japan, e-mail: [email protected]

1. Introduction
Observation records were obtained at the reactor buildings of Kashiwazaki-
Kariwa nuclear power plant during the Niigata-ken Chuetsu-Oki Earthquake,
which occurred on July 16, 2007. These records were used to carry out
simulation analyses on representative reactor buildings. This report presents the
framing of the analyses and the analysis results for Reactor Buildings of Unit 1
and Unit 6.

2. Method of analyzing earthquake response

Observation records (acceleration time history) on the base mat slab were input
to the earthquake response analyses for the reactor buildings specifically where
such records were available. In the analysis, linearity was assumed for both the
building and the soil spring. Two horizontal directions (North-South and East-
West) were analyzed independently.
The simulation analysis method is as follows.
First, the dynamic responses of the ground at the embedded part of the
building were calculated from the analysis based on the one dimensional wave
propagation theory.
Second, the dynamic responses were applied to the building for the simulation
analysis in the horizontal direction as the input motions. As a result, the

167
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

responses of the building were evaluated taking into account the soil structure
interaction. With the responses of the building, transfer function between each
floor and the base mat slab was obtained.
Third, the earthquake responses of the building’s individual parts were
obtained by multiplying the transfer function calculated as above by the Fourier
spectrum obtained from the Fourier transform of the observation records for the
base mat slab.

3. Earthquake response analysis model

The analysis model considering soil structure interaction to be used for the
earthquake response analysis was determined to be an embedded sway-rocking
model consisting of a lumped mass system for the building and the ground
spring.
The building model is described as lumped mass model with weight
concentrated at mass point located at each floor, taking into account bending and
shear stiffness. In the building model, auxiliary walls that were thought to
increase stiffness during an earthquake were taken into account in addition to the
main shear walls considered in the initial design. Stiffness of concrete was
evaluated based on the measured compressive strength of test pieces sampled
from actual building walls. The model also took into account the stiffness
estimated in accordance with the design code for reinforced concrete structures
in Japan (Young’s modulus and shear modulus).
The building’s damping constant was determined to be 5%. The soil springs
for the base mat slab employed a constant spring (horizontal and rotational)
based on Tajimi’s vibration admittance theory. The soil springs for the side
plane of the building’s embedded part employed a constant spring (horizontal
and rotational) based on the NOVAK’s method. The ground properties took into
account the strain-dependency of the stiffness and damping based on laboratory
tests. Building-specific decreasing stiffness and including damping factor were
provided in accordance with the strain level of the ground.

4. Outline of earthquake observations

Seismometers were installed on the base mat slab of the building’s lowest floor
and on the intermediate floor in the building’s upper portion in each building.
The seismometers recorded the acceleration time histories as the observation
record. Sampling frequency was 100Hz. A sufficient number of observations
were acquired for the earthquake response analysis.

168
 5. Modeling, Testing and Response Analysis of Structures, Systems and Components

5. Results of simulation analysis

Simulation analysis was carried out using the acceleration time histories
observed at the Reactor Buildings of Unit 1 and Unit 6 as input motions. The
results were compared with the observation records in terms of the distribution
of the maximum response accelerations in building height direction, the wave
form of the acceleration time history, the acceleration response spectrum, etc.
As a result, the simulation analyses using the analysis models that took into
account the realistic conditions during the earthquake showed relatively good
agreement with the earthquake observations.

6. Conclusions
Simulation analyses were carried out on reactor buildings using observation
records at reactor buildings of Kashiwazaki-Kariwa Nuclear Power Plant
obtained during the 2007 Niigata-ken Chuetsu-Oki Earthquake, which occurred
on July 16, 2007.
Those for Reactor Building of Unit 1 were based on the design model. They
employed the sway rocking models incorporating realistic conditions of building
and ground. It also utilized observation records from the base mat slab. As a
result, the following items were found.
• The simulated distribution of the maximum response acceleration
closely followed the trend of observation records and showed good
agreement.
• The simulated floor response spectrum of the intermediate floors in the
building closely followed the frequency characteristics of the
observation records.

169
5. Modeling, Testing and Response Analysis of Structures, Systems and Components

170
 6. Design and Construction Issues

6. Design and Construction Issues

Design, manufacture and construction of fuel and core structure, pressure

vessel; steel and concrete containment structures; other concrete structures
(storage, processing, etc.); piping and major components, fuel cask
design. Capacity, ductility, redundancy and quality considerations. Design
of passive safety systems. System isolation and energy absorption.
Construction management.

171
6. Design and Construction Issues

172
 6. Design and Construction Issues

On the design of pipe supports and

steelwork regarding revised German
nuclear safety standards (6-1587)

Dr. Lutz Lindhorst, Jens Milleder

TÜV SÜD Industrie Service GmbH, Munich, Germany

The paper deals with recent revisions of technical standards for structural
steelwork and supports in the conventional, non-nuclear area and their influence
on the state of the art concerning nuclear safety standards in Germany.
Since several years efforts are made in the European Union to find a common
basis concerning the set of technical standards and to leave the path of individual
standards in the single nations. One of these efforts concerns structural steelwork
in the non-nuclear area, where the EUROCODE 3 /1/ was established in civil
engineering and it is planned to withdraw the valid German DIN 18800 [11/90]
/2/ in the near future, as it was already done with former German standard DIN
18800 [03/81] /3/ some years ago. Both the EUROCODE 3 /1/ and the German
standard DIN 18800 [11/90] /2/ are based on a semi-probabilistic procedure
considering different partial safety factors for actions as well as for material
properties in contrast to the obsolete German standard DIN 18800 [03/81] /3/
which was based on a deterministic procedure using allowable stresses and
global safety factors for the resistances.
In Germany the mechanical engineering parts (including their support
structures) of nuclear power plants are based on atomic regulations (and related
nuclear safety standards) but the civil works of nuclear facilities (including for
example anchorages and dowels) are additionally based on conventional
buildings regulations (and related conventional standards). Many of the existing
supports with importance to safety in German nuclear power plants where
installed in the construction phase of the power plants. This was a time when a
deterministic concept was the basis for the related design calculations. Today the
deterministic procedure is still implemented in German nuclear safety standard
KTA 3205 /4/–/6/ for supports. But the deterministic design concept is today not
more allowed for new proof calculations concerning the power plant buildings
and anchorages. Therefore two different design concepts (deterministic and
semi-probabilistic) meet at the interface between the building and the supports of
mechanical engineering equipment. This is especially to be seen for the building
structure interaction loads acting on anchorages of passive and active
components, because there is no safety factor for actions in the deterministic
design concept but there are different partial safety factors for permanent,
variable and accidental loads in the new semi-probabilistic design concept.

173
6. Design and Construction Issues

Current efforts are made in standardization committees dealing with German

nuclear safety standards KTA to find an interface agreement considering the
state of science and technology. In this paper an overview about the German
nuclear safety standard KTA 3205 part 1 to 3 /4/–/6/ for supports is given with
respect to recent works concerning the revision of part 2 of KTA 3205 /5/.
Furthermore differences between German standards DIN 18800 [03/81] /3/, DIN
18800 [11/90] /2/ and EUROCODE 3 /1/ are regarded. An example of a steel
construction is presented to illustrate the design concepts. The results obtained
show that different solutions concerning stress utilization ratio and interaction
loads can occur. Caused by the new German and European design standards for
steelwork a revision of current German nuclear safety standard KTA 3205.2 /5/
is necessary. In this revision requirements resulting from mechanical engineering
and from civil engineering have both to be taken into account.

References
1. DIN EN 1993-1-1: EUROCODE 3: Bemessung und Konstruktion von Stahlbauten –
Teil 1-1: Allgemeine Bemessungsregeln und Regeln für den Hochbau, Deutsche
Fassung EN 1993-1-1:2005, Normenausschuß Bauwesen (NABau) im DIN
Deutsches Institut für Normung e.V.

2. DIN 18800, Teil 1: Stahlbauten, Bemessung und Konstruktion, 11/90, Normen-

ausschuß Bauwesen (NABau) im DIN Deutsches Institut für Normung e.V.

3. DIN 18800, Teil 1: Stahlbauten, Bemessung und Konstruktion, 03/81, Normen-

ausschuß Bauwesen (NABau) im DIN Deutsches Institut für Normung e.V.

4. KTA 3205.1: Komponentenstützkonstruktionen mit nichtintegralen Anschlüssen; Teil 1:

Komponentenstützkonstruktionen mit nichtintegralen Anschlüssen für Primärkreis-
komponenten in Leichtwasserreaktoren, 06/02, Kerntechnischer Ausschuß (KTA).

5. KTA 3205.2: Komponentenstützkonstruktionen mit nichtintegralen Anschlüssen; Teil 2:

Komponentenstützkonstruktionen mit nichtintegralen Anschlüssen für druck-
und aktivitätsführende Komponenten in Systemen außerhalb des Primärkreises,
06/90, Kerntechnischer Ausschuß (KTA).

6. KTA 3205.3: Komponentenstützkonstruktionen mit nichtintegralen Anschlüssen; Teil 3:

Serienmäßige Standardhalterungen, 11/06, Kerntechnischer Ausschuß (KTA).

174
 6. Design and Construction Issues

Active control of vibrations in piping

systems (6-1658)

Carsten Block, Juergen Engelhardt, Fritz-Otto Henkel

Woelfel Beratende Ingenieure GmbH + Co. KG
Max Planck Strasse 15, 97204 Hoechberg, Germany
e-mail: [email protected]

There are various dynamic load cases that can lead to vibrations in piping
systems. Typical vibration sources are, for example, turbulences in pipelines
with through-flow, pressure surges resulting from actuation of valving,
vibrations of directly connected machines, machine faults or earthquakes.
If the resulting vibrations exceed permissible values remedial measures have
to be taken. Common measures for reducing vibrations in piping systems are
reduction of excitation, detuning, damping or passive tuned mass dampers.
The efficiency of conventional passive methods for reducing vibrations is
limited. The application of passive dampers for example requires a fixed
support. Tuned mass dampers demand a mass ratio of about 10% to reach a good
attenuation. Therefore tuned mass dampers are often unfeasible due to a high
static load. In addition, tuned mass dampers are restricted to one resonance
frequency.
To overcome problems with conventional passive methods an Active
Vibration Absorber (AVA) for vibration reduction of industrial piping was
developed, manufactured and tested. The design and the realization of the AVA
shows Fig. 1. The AVA is based on a 2 DOF active system that is able to reduce
vibrations over a large frequency range.

unit
I unit
 II


Figure 1. Active Vibration Absorber (left: design, right: realization).

175
6. Design and Construction Issues

The AVA’s function is based on the principle that an accelerated inertial mass
generates a reaction force in the supporting structure:
F = − m ⋅x (1)
The reaction mass is connected to the structure by means of a spring. An
actuator is located parallel to the spring. By accelerating the reaction mass, using
the actuator in combination with an appropriate control algorithm, a resultant
force for vibration reduction is obtained.
The AVA was investigated on a mock-up of a piping system. The developed
decentralized controller shows an excellent performance in the considered
frequency range. In addition, it could be demonstrated that the control
performance is not sensitive even to distinct parameter changes of the structure.
Fig. 2 shows typical frequency response functions measured at the mock-up with
and without the AVA.

vertical horizontal
4 2

3.5

3 1.5

2.5
A [-]

A [-]

2 1

1.5

1 0.5

0.5

0 0
5 10 15 20 5 10 15 20
f [Hz] f [Hz]

Figure 2. Typical frequency response functions (left: vertical, right: horizontal).

References
1. Nishimura, I. et al. Active tuned mass damper. In: Smart Mater Struct (1992), 1,
pp. 306–311.

2. Preumont, A. Vibration Control of Active Structures. 2nd Edition, Dordrecht: Kluwer

Academic Publishers, 2002.

3. Engelhardt, J. et al. Active Vibration Absorbers for Industrial Piping Systems. Adaptronic
Congress 2007, 23–24 May, Goettingen Germany, 2007.

4. VDI 3842: Vibrations in piping systems, 2004–2006.

176
 6. Design and Construction Issues

Performance-based design of SSC

wall in fire (6-1675)

Il Hwan Moon1, Nam Yong Jee2, Won Ki Kim3,

Chung Seon Lee2, Suk Tae Yoo1
1
Civil/Architectural Eng. Dept., Korea Power Engineering Co., Inc.
Yongin, Korea, e-mail: [email protected]
2
Dept. of Architectural Eng., Hanyang University, Seoul, Korea
3
Dept. of Architectural Eng., Hoseo University, Asan, Korea

Introduction
This paper describes proposed design method based on the fire performance test
results of stiffened steel plate concrete (SSC) wall with ribs. SSC wall is a
composite structure consisting of surface steel plate with stud and rib instead of
rebar of reinforced concrete wall. In fire conditions, surface steel of SSC wall is
directly exposed to fire attacks, therefore load-carrying capacity of the member
is reduced due to degradation of material property. Although the reduction of
stiffness and strength is expected, a minimum fire resistance rating of 3 hours
should be provided for application of SSC wall in nuclear power plant structure.
In this study, the testing of a loaded SSC wall under ISO fire conditions was
carried out in generic transient state testing way and also a simple design method
was proposed based on fire performance test result to ensure required fire
resistance rating. The design specifications are included in SC structure standard
for nuclear facilities in Korea.

Fire test of SSC wall

Fig. 1 shows arrangement for fire testing SSC wall and the typical test results. In
the testing, axial load is applied to the structure first and then held constant and
the structure is exposed to ISO fire attack. The test is terminated when one of the
specified failure criteria is reached. Horizontal displacement at 185 minutes after
fire exposure is drastically increased in test result and the maximum deflection
exceeds L/20, where L is the span of the specimen. Fig. 2 shows the relations of
vertical displacement and fire exposure time. The axial contraction at 185
minutes exceeds limiting value defined in ISO 834 as h/100, where h is the
initial height in millimeters. Fig. 3 shows the temperature distribution on
unexposed surface and the maximum temperature rise at the locations is 87oC.
This temperature is less than 135oC which is maximum allowable temperature
on unexposed surface of fire barrier wall.

177
6. Design and Construction Issues

Displacement [mm]
-5

-1 0
h o r izo n ta l d is p la c e m e n t S - 2 0 0 0
h o r izo n ta l d is p la c e m e n t C - 2 0 0 0
-1 5 h o r izo n ta l d is p la c e m e n t N - 2 0 0 0
a ve ra g e

-2 0
0 20 40 60 80 100 120 140 160 180 200
T im e [m in ]

(a) Test Set-up (b) Measured Displacement

Figure 1. Test Set-up & Measured Displacement.

I average II-1 average II-2 average III average

100
2 u n e xp o s e d
u n e xp o s e d
s u r f .1
s u r f .2
90
u n e xp o s e d s u r f .3
u n e xp o s e d s u r f .4
80 u n e xp o s e d s u r f .5
0 u n e xp o s e d s u r f .6
70 u n e xp o s e d s u r f .7
Displacement[ mm]

u n e xp o s e d s u r f .8
u n e xp o s e d s u r f .9
Temperature [°C]

60 u n e xp o s e d s u r f .1 0
-2 u n e xp o s e d s u r f .1 1
50

40
-4
30

-6 20

10

-8 0
0 20 40 60 80 100 120 140 160 180 200
0 20 40 60 80 100 120 140 160 180 200 T im e [ m in ]

Time[ min]

Figure 2. Vertical Displacement. Figure 3. Temperature on Unexposed

Surface.

Performance-based design method

SC wall design specifications for the fire conditions consists of the design
requirements for fire conditions, mechanical properties of steel and concrete at
high temperature, temperature distribution at cross-section, finite element
method (or fiber model analysis), segment analysis, simple design method, and
required detail.
Based on the fire test results and the theoretical studies the specifications are
developed for SC wall design under fire condition. In this specifications, SC

178
 6. Design and Construction Issues

wall with the thickness (t) of above 300 mm and the ratio of below H/t(wall
height/wall thick)=13 shall be considered the fire barrier structures satisfying
license requirements if the axial force ratio is less than 40%. There are three
design methods for the fire barrier design of SC wall such as the fire model
analysis, the segment analysis, and the simple design method. The simple design
method is the safe design method. Moreover, the simple design method is
adapted because structural components in the nuclear power plant are typically
thick enough to satisfy a three-hour fire resistance requirement even without a
precise design method (Korea Electric Association, 2008). In proposed simple
design method for SC wall, a damaged depth for the stiffness calculation is
assumed as 100 mm based on the test data of Kodaira et al. using SC wall with a
thickness of 200 mm and the test result of the authors using SSC wall with a
thickness of 300 mm. A damaged depth for the strength calculation is assumed
as 50 mm based on the same test results.

Reference
Korea Electric Association (2008). SN Structures for Nuclear Facilities; SC Structures.
Korea Electric Power Industry Code, KEPIC SNG.

179
6. Design and Construction Issues

Upgrade and modification of fuel handling

equipment in Korea (6-1693)

Chang, Sang-Gyoon
NSSS Engineering project Division, Korea Power Engineering Company, Inc.
Daejeon, Korea

The upgrade and modification of the fuel handling system are currently in
progress in Korea. Twenty (20) Nuclear Power Plants are in operation and four
(6) units are under construction and two (2) units are being designed. The fuel
handling system consists of equipment used for receiving and transporting fuel
assemblies and acts as critical processes in the refueling outage. The refueling
machine, the fuel transfer system and spent fuel handling machine are used for
hoisting and transferring a fuel assembly between the core of the containment
building and the fuel storage rack of the fuel building. The improvements for the
equipment under construction and in operation have been studied to enhance the
operation reliability and efficiency. One of these improvements is to upgrade the
hoisting and transferring capability of the equipment during the refueling outage.
For the upgrade and the modification of the equipment, the Programmable Logic
Controller (PLC) based control system and high speed motor drive system and
advanced operating procedures are introduced. In this study, the scope of the
improvements for the fuel handling equipment is reviewed and the performance
of the equipment during the fuel loading process are analyzed and evaluated. We
can expect the fuel transferring capability of the equipment per hour and
recommend further improvements based on this study. Resulting from applying
the advanced fuel handling equipment, operating efficiency of the plant will be
increased by reducing the refueling time and accompanied by reduction of
radiation exposure during the refueling outage.

180
 6. Design and Construction Issues

Damping values for seismic design of

nuclear power plant SC structures (6-1697)

Wonki Kim1, Seung Joon Lee2, Rae Young Jung3, Moonsoo Kim4
1
Professor, Architectural Engineering Dept., Hoseo University, Korea
e-mail: [email protected]
2
Professor, Architectural Engineering Dept., Ajou University, Korea
3
Senior Researcher, Structural Systems & Site Evaluation Dept.
Korea Institute of Nuclear Safety
4
Senior Researcher, Structural Systems & Site Evaluation Dept.
Korea Institute of Nuclear Safety

Background
Structural damping values for seismic design of nuclear power plant structures
are specified in Regulatory guide 1.61 for reinforced concrete structures of 4%
(OBE) and 7% (SSE), and for steel structures of 3% (OBE) and 4% (SSE), but
not for steel-plate concrete (SC, hereinafter) structures. SC structures have been
being developed in the worlds for long time, but no research investigates the
damping values except that Akiyama et al. concludes the identical damping
value of 5% for both RC and SC structures for nuclear power plants. However,
the experimental tests conducted by Akiyama are static cyclic loading tests and
hydraulic-shaker vibration tests without any mass in the test specimens of both
test types.

Objectives and investigations

This paper describes the research work of experimental testing method, analysis
of test results and proposed damping values for seismic design of nuclear power
plant SC structures.
The concept of this research is to investigate the relative difference in
damping values between RC and SC structures rather than to find out the
absolute values of SC structures, so that tested are 4 specimens of RC-S, RC-M,
SC-S and SC-M where S stands for shear-govern and M for moment-govern. As
described in Table 1, the moment-govern specimens are higher than the shear-
govern ones, but all the specimens are 1,700 mm wide and 1,700 mm long with
wall thickness of 240 mm.

181
6. Design and Construction Issues

Table 1. Descriptions of 4 specimens.

Conducted method is free vibration testing by rupturing a brittle steel plate

which links an actuator and the center of 50 ton mass as shown in Figure 1.
Rupturing load levels on each specimen are controlled by pre-designed tensile
strength of the linking steel plate.

Results
Natural frequencies of 4 specimens are determined from the experimental test
results with respect to the load level as illustrated in Figure 2. It is noticed that
those frequencies are similar to the values in design practice.
Figure 3 shows an example of time vs. acceleration curve measured for
specimen SC-S at rupturing load of 420 kN, together with fitted curves of
exponential function representing free vibration. Consequently, damping value is
determined from the exponential function and its natural frequency. Similar
analyses are performed to determine damping values of 4 specimens at the
different rupturing load.

Figure 1. Test Setup of SC-M Specimen. Figure 2. Natural Frequency of

4 Specimens.

182
 6. Design and Construction Issues

Figure 3. Example of Time vs. Acceleration. Figure 4. Damping values of 4 specimens.

Figure 4 shows comparison of damping values of 4 specimens with respect to

load level, where the load level is the rupturing load divided by designed load
which corresponds to specimen design strength.

Conclusions
By examining the relative differences in damping values of 4 specimens, it is
proposed for SC structure to use the same damping values of 4% as RC at OBE,
but 1% less value than RC resulting in 6% at SSE.

References
1. U.S. Nuclear Regulatory Commission. Regulatory Guide 1.61. Damping Values for
Seismic Design of Nuclear Power Plants. March 2007.
th
2. Hiroshi Akiyama et al. 1/10 Scale Model test of Inner Concrete Structure Composed
of Concrete Filled Steel Bearing Wall.

183
6. Design and Construction Issues

Performance-based fire design of half SC

slabs in nuclear power plants (6-1698)

Wonki Kim1, Nam Yong Jee2, Chung Seon Lee3, Tae Youp Mun4
1
Professor, Architectural Engineering Dept., Hoseo University, Korea
e-mail: [email protected]
2
Professor, Architectural Engineering Dept., Hanyang University, Korea
3
Ph.D. Candidate, Architectural Engineering Dept, Hanyang University, Korea
4
Manager, Project Engineering Dept., Korea Hydro & Nuclear Power Co., Ltd.

Background
One of primary elements in steel-plate concrete(SC hereafter) structures is a half
SC(HSC hereafter) slab which consists of concrete, top reinforcements, bottom
surface steel plate, bottom steel ribs and shear studs. Meanwhile, required fire
resistance rating is 2 or 3 hours based on ISO fire conditions for all the slabs of
nuclear power plant structures in Korea. It is obvious that HSC slabs are weaker
than SC slabs in fire design since bottom reinforcements of HSC slabs are
basically exposed surface steel plate and attached steel ribs only.
Michikoshi(1) et al. performed experimental tests by simulating both end-fixed
condition for HSC slabs subjected to required loading at fire condition as shown
in Figure 1, and found out that 300mm thick HSC slabs meet the requirement of
3 hour rating by relying on end moment capacity.
There is a need for additional fire testing of thicker HSC slabs which are
connected to SC walls as well in order to provide a performance-based fire
design specification of not only both end-fixed condition but also one end-fixed
and simply supported ones.

Figure 1. Cross-section of Michikoshi’s Specimen for Fire testing.

184
 6. Design and Construction Issues

Objectives and investigations

This paper describes research work of experimental fire testing for HSC slabs
connected to SC wall as shown in Figures 2 and 3. It is noticed that all the
specimens are not loaded since it is impossible to apply required load at fire
condition on full-scaled both end-fixed HSC slabs.

Figure 2. Cross-section of HSC Slab and SC Figure 3. Setup of Fire Testing.

wall in Heating Furnace.

Investigated are only temperature profiles on the concrete far from the steel ribs,
and on the steel ribs at the time of 3 hour firing. Tested are 4 specimens having
primary differences in slab thickness of 450mm and 300 mm, and rib shapes as
described in Table 1.

Table 1. Description of 4 specimens.

Specimen Symbol HSC 450/H200 HSC-450/T250 HSC-300/T175 HSC-300/T135

Slab Thickness (mm) 450 450 300 300
Surface Steel Plate Thickness
9 9 4.5 4.5
(mm)
Rib Section H-200×150×6×9 T-250×200×9×14 T-175.5×72×6×9 T-135.5×72×6×9
Stud φ16@300 φ16@300 φ9@150 φ9@150

Results and analysis

Temperature profiles on concrete far from the steel ribs obtained for all 4
specimens are identical to Michikoshi’s Taisei specimen’s one as shown in
Figure 4. It is noticed that such identity is even resulted from the differences in
slab thickness, concrete strength and loading condition. But, somewhat different
temperature profiles are obtained on the steel rib as shown in Figure 5.
Based on the temperature profiles and well-known properties of steel and
concrete at elevated temperatures(2), flexural strengths of HSC slabs are

185
6. Design and Construction Issues

analyzed for negative bending at the end and for positive bending at the mid
span at the time of 3 hour rating by using segment analysis. Such analysis results
for Michikoshi’s specimen are compared with test results resulting in good
agreement.

Figure 4. Temperature Profile on Concrete Figure 5. Temperature Profile on Steel Rib.

far from Steel Ribs.

Conclusions
Based on this research work, Korea Electric Power Industry Code SNG specifies
a segment analysis method and a simplified method for performance-based fire
design for the 3 hour rating of HSC slabs in both end-fixed, one end-fixed and
simply supported conditions.

References
1. Shintaro Michikoshi et al., "Fire Resistance of Steel Concrete Slabs with T-shape
Steel", Journal of Structural and Construction Engineering, No 554, Architectural
Institute of Japan, April 2002.

2. American Institute of Steel Construction, “Specification for Structural Steel Buildings”,

ANSI/AISC 360-05, March 2005.

186
 6. Design and Construction Issues

Serviceability limit state and crack width

analysis of concrete structures in
nuclear power plants (6-1706)

Pekka Iivonen1, Esa Turunen2, Jari Puttonen3

1
ÅF-Consult Ltd, P.O. Box 61, FI-01601 Vantaa, Finland
e-mail: [email protected]
2
ÅF-Consult Ltd, P.O. Box 61, FI-01601 Vantaa, Finland
e-mail: [email protected]
3
Helsinki University of Technology, Department of Structural Engineering and
Building Technology, P.O. Box 2100, FI-02015 TKK, Finland
e-mail: [email protected]

In nuclear power plants typical for reinforced concrete structures is that they are
thick and massive compared to structures in ordinary buildings. The needed
tightness or long-term stiffness of several critical structural components made of
reinforced concrete is the reason that the serviceability is perhaps more stringent
requirement than the ultimate capacity in designing of reinforced concrete
structures. However, their design is largely based on codes developed for non-
nuclear applications, and the challenge is the application of beam theory based
code formulas to complicated multidimensional structures of nuclear facilities,
where the temperature changes assumed may produce substantial part of loading.
A special feature of reinforced concrete is that the effect of temperature changes
depends on the stiffness of the structure, which on the other hand caused by
cracking. Thus, it is important to have a design tool by which it is possible to
update the stiffness of the structure due to cracking and to consider the effect of
stiffness changes in redistribution of stress.
The paper presents the in-house design tool comprising of several independent
programs that are chained to analyse and dimension three dimensional
reinforced concrete structures iteratively. The main parts of the design tool are a
commercially available program based on the general finite element method and
the in-house programmes that define the cracking, update stiffness and calculate
reinforcement bars in different directions. The results of the tool are actual
amounts of reinforcement required in each element of the finite element mesh.
The amounts can finally be transferred to drawing programs. The program can
be used either for dimensioning of reinforcement of a new structure or for
analysis of crack widths of an existing structure. The influence of the order in
which the loads affect on the structure can also be analysed by the tool.
A specific object of the paper is to compare different design codes in the
serviceability limit state design. The design codes compared are the National
Building Code of Finland “B4: Concrete structures” and the European Standard

187
6. Design and Construction Issues

“Eurocode 2: Design of concrete structures”. The calculation formulas of the

different design codes are compared and their differences are demonstrated by
practical test examples and real structures of nuclear power plants. The results
reveal that these codes lead a little bit different reinforcement even if they both
are official codes in Finland presently. For example with the same crack width
requirement, the amount of required reinforcement according to Eurocode 2 is
for bending a bit smaller and for pure tension notably larger than according to
B4. If the comparison is made using the same exposure class, the amount of
required reinforcement according to Eurocode 2 is smaller than according to B4
both for bending and for pure tension, because the crack width requirement of
B4 is often stricter than the requirement of Eurocode 2.

188
 6. Design and Construction Issues

Demands on anchor systems for concrete

structures of nuclear facilities (6-1709)

Rüdiger Meiswinkel1, Franz-Hermann Schlüter2

1
E.ON Kernkraft GmbH, Hannover, Germany
2
SMP, Consulting Engineers for Civil Engineering, Karlsruhe, Germany

With respect to nuclear power plants or other nuclear facilities the safety aspects
of anchor systems are of great importance especially in view of extraordinary
action effects like earthquake actions. In the last two years this importance has
become obviously especially in Germany by the outage of several nuclear power
plants because of not correct installed metal anchors. In the concerned plants
extensive repairs with the exchange of several thousands of metal anchors were
necessary.
Generally the anchoring of mechanical components in concrete structures will
be realized by steel anchor plates with welds for the connection between the
plates and the components. During the erection of nuclear facilities for the so far
known fastening points cast-in fasteners in form of steel anchor plates with
welded studs will be preferred. After the erection of the buildings for
modification measures so called post-installed anchors are the only way for
subsequent connections. Supplementary metal anchors can be used for such
fastening points. Three types of metal anchors seem to be suitable for the
application in nuclear facilities:
- expansion anchor: anchor with friction connection for the anchoring of
tensile forces
- undercut anchor: anchors which develops its tensile resistance from
the mechanical interlock provided by undercutting of the concrete at the
embedded end of the fastener
- bonded anchor (chemical anchors): threaded bar embedded in the
bore holes by an adhesive mortar.
In German nuclear facilities expansion anchors have been used before the
development of the undercut anchors. Nowadays undercut anchors will be
preferred in Germany for such applications. Until now no bonded anchors have
been qualified for the utilization in German nuclear facilities because of the very
high demands on the resistance in cracked concrete. But in recent years bond
expansion anchors were developed to offer a suitable high performance adhesive
anchoring system in cracked concrete.

189
6. Design and Construction Issues

Metal anchors are building products with high quality demands on the
utilization in nuclear facilities. Regarding these demands three aspects have to
be considered:
- anchor product regarding the licensing aspects for building products
which can be used in nuclear facilities
- design regarding the verification format for the different limit states
especially for the ultimate limit states
- installation regarding the requirements during the installation process
including the necessary installation protocol.
In Germany the different demands have been established in the DIBt-guideline
[1] with the specifications for the licensing of metal anchors and for the design
of anchor connections. This guideline represents an extension of the EOTA-
guideline [2] to consider extraordinary action effects typically for nuclear
facilities like earthquake actions. So for example the load bearing of anchors has
to be guaranteed in cracked concrete structures with crack openings of 1.5 mm
in a single crack considering cyclic loading.
Because of the new partial safety concept in combination with the new
definition of requirement categories for nuclear facilities [3] a revision of the
DIBt-guideline has become necessary. In view of this revision started in 2008
latest scientific findings about the anchor behaviour in cracks (see [4]), the
gained experience and new knowledge about installation proceedings as well as
the nuclear specific demands like the limitation of anchor displacements will be
discussed in the contribution.

References
1. DIBt-guideline (1998). Utilization of anchors in nuclear power plants and nuclear
installations – guidelines for assessing anchored fastenings when granting
permits in an individual case in accordance with the building regulations of the
Federal States, Berlin.

2. EOTA (1997). ETAG guideline for the European technical approval for metal anchors
for anchoring in concrete – ETAG 0001. Appendix C, Brussels.

3. KTA-GS-78 (2005). Recommendations for the consideration of the present

constructional standards using KTA-regulations. KTA office, Salzgitter.

4. Hoehler, Eligehausen (2008). Behaviour and testing of anchors in simulated seismic

cracks. ACI Structural Journal Title no. 105–S34.

190
 6. Design and Construction Issues

Establishment on slip coefficient of slip

resistant connection (6-1712)

Hwan-Seon Nah1, Hyeon-Ju Lee2, Kang-Seok Kim3

Environmental & Structural Lab., Korea Electric Power Research Institute
65 Munji-Ro, Yusung-Gu, Daejon, 305-380, Korea, e-mail: [email protected]
1
Principal Researcher, Corresponding author, e-mail: [email protected]
2
Senior Researcher, e-mail: [email protected]
3
Researcher, e-mail: [email protected]

A slip critical joint has various values to adopt the proper slip coefficient on
various conditions of faying surfaces in following codes: AISC, AIJ and
Eurocode 3. However Korean Building Code still regulates the unique slip
coefficient, 0.45 regardless of diverse faying conditions. In this study, the slip
resistance test including five kinds of surface treatments were conducted to
obtain the proper slip coefficients available to steel plate KS SM490A. The
faying surfaces were comprised of clean mill, rust, red lead paint, zinc primer,
and shot blast treatment. The candidates of high strength bolt were torque-shear
bolt, torque-shear bolt with zinc coating, and ASTM A490 bolt. Based on test
results, the specimens with shot blasted surface and rusted surface exhibited ks,
0.61, 0.5 respectively. It is recommended that the specimens with zinc primer
exhibited ks. The clean mill treated surface were prominently lower values, 0.27.
For red lead painted treatment, thickness of coating affected on the determinant
of slip coefficient, it is necessary to establish the minimum of ks of 0.2, with
coating thickness, 65 μm. For 1,000 hours relaxation, the uncoated surfaces
exhibited the loss of clamping force behind 3%, while the coated surfaces within
a certain limited thickness exhibited the loss of clamping with a range between
4.71% and 8.37%.

191
6. Design and Construction Issues

The SWR 1000 containment – civil

design aspects in view of high
robustness (6-1714)

Eberhard Bielor1, Andreas Garg2, Heribert Hansen2, Rüdiger Meiswinkel3

1
AREVA.NP GmbH, Offenbach, Germany
2
HOCHTIEF Construction AG, Frankfurt, Germany
3
E.ON Kernkraft GmbH, Hannover, Germany

Operational lifetime of nuclear power plants of the new generation is normally

provided to be 60 years currently. Taking into account a construction period of
about 4–5 years together with a decommissioning period of about 10–15 years,
time periods of about 80 years are newly requested, where safety-related
requirements on construction members have to be guaranteed.
The containment of the SWR 1000 is characterized by a lot of passive safety
systems and therefore by a robust behaviour in abnormal operational conditions.
This concept will be realized not only in plant engineering but also in civil
engineering: The SWR 1000 provides a passive safety concept in its constructional
members which fulfil the requirements of high robustness.
Due to these requirements on robustness, materials and structural members are
preferred which are visible and therefore easily accessible for monitoring. This
results in basic consequences with regard to the application of e.g. pre-stressing
system (especially combined with a steel liner), tightness of joints using sealing
strips, bituminous waterproofing systems, joints between separated fire compartments.
These are only few provisions which can be taken as examples.
The containment structure of the SWR 1000 is therefore provided without pre-
stressing, using only non-pre-stressed reinforcing bars. Monolithic construction
methods avoid joints, result in a higher bearing capacity of the global
containment structure and avoid also difficulties with joint sealants with regard
to fire protection and pressure differences. Watertight structural members, e.g.
waterproof concrete basements, do not require bituminous sealing.
Construction engineering design with regard to high robustness has to be
supported by corresponding codes and standards. These rules should be based on
a unified safety concept, providing design procedures for robust structural
members and addressing also the global integrity of the reinforced containment
combined with a robust liner construction, penetrations and airlocks.
Parallel to the development of the SWR 1000-containment in Germany, the
new codes DIN 1045 (reinforced and pre-stressed concrete structures) with parts
1–4 and DIN 25449 (reinforced and pre-stressed concrete components in nuclear
facilities) have been issued. A working group has been established in the year
2008 to develop a revision of DIN 25459 (reinforced and pre-stressed concrete

192
 6. Design and Construction Issues

containments) to achieve a harmonization and adjustment to the other codes

mentioned above. These three codes, based all together on the Eurocodes
regarding a unified partial safety concept, represent for the first time in Europe a
self contained and complete set of codes and standards for the design of
reinforced concrete containments.
The design of a robust containment will be demonstrated, representing the
construction of the SWR 1000-containment. Furthermore, the current state of
applicable codes and standards and future developments for new rules are
illustrated, to enable the design of robust containments.

References
DIN 1045 (2001). Concrete, reinforced and pre-stressed concrete structures, Part 1–4.
Beuth-Verlag, Berlin.

DIN 25449 (2008). Reinforced and pre-stressed concrete components in nuclear facilities
– Safety concept, actions, design and construction. Beuth-Verlag, Berlin.

DIN 25459 (in preparation). Reinforced and pre-stressed concrete containments of

nuclear power plants. Beuth-Verlag, Berlin.

193
6. Design and Construction Issues

Guidelines and dataware for life cycle

management for NPP pipeline
supports (6-1774)

S.V. Evropin1, A.E. Obushev1, Yu. K. Spirochkin1, A.V. Petrenko2

1
Engineering Center of Nuclear Equipment Strength (ENES)
Malaya Krasnosel’skaya Str. 8/2, Moscow 107140 Russia
e-mail: [email protected]
2
Atomenergoproject (SPbAEP)
2nd Sovetskaya Str. 9/2a, Saint Petersburg 191036, Russia
e-mail: [email protected]

The present development of the Russian nuclear industry requires an increase of

power, life duration and reliability of design for new nuclear power plants
(NPPs). These goals necessitate the upgrade of existing nuclear industry
guidelines and the development of new methods and technologies for life cycle
management of NPP components. The development of a “Code of Rules and
Guidelines on Support Structures for NPPs with WWER” (SPiR-O-2008) is a
part of such work.
The paper describes a structure and basis for SPiR-O-2008 that is created in
such form in Russia for the first time. Methodic approaches, which form the base
of SPiR-O-2008, cover both design and post-design assurance of capacity of
support structures, namely:
– Division of requirements in dependence on safety classes;
– Limit analysis;
– Taking into account of actual finite stiffness;
– Conformance evaluation by calculations or tests;
– In-service inspection (structural health monitoring and damage detection).
The important feature of SPiR-O-2008 is its development on a base of modern
information technology. The document is created in the form of hypertext and
includes electronic catalogues and 3D models of standard supports and hangers
that can be used directly by designers of pipeline systems.
SPiR-O-2008’s information maintenance is based on product life cycle
management (PLM) – technology recently introduced into the Russian nuclear
industry. It involves the use of an integrated information medium for all stages
of the life cycle and for all cooperating participants. It also considers standards
of information exchange and justification of engineering decisions on the basis
of mathematical modeling. There is a logical link created between the
mathematical model and the in-service structural health monitoring. Data from
monitoring is needed to update the mathematical models of support structures

194
 6. Design and Construction Issues

that take into account their actual state. The last point is important for the
substantiation of decisions concerning extension of life duration, repair or
replacement.
Another specific feature of SPiR-O-2008 is its reconciliation with foreign
Codes and Guidelines (ASME Code, KTA Rules, VGB Guidelines) that provide
a possibility of international cooperation within the frame of Russian NPP
projects. Standard support structures of LISEGA AG (Germany) are considered
as an example of such cooperation in a NPP project with WWER designed by
SPbAEP. The paper describes problems arising in this design in connection with
Russian design practices.

195
6. Design and Construction Issues

A case study on a radiation shielding

structure for the cold neutron guide at
HANARO – focused on a mixed proportion
design and fabrication of heavy weight
concrete for a radiation shielding (6-1791)

Sang-ik Wu1, Young-ki Kim1, Tae Bong Kim2

Development of Systems for Cold Neutron Source Project
1
Korea Atomic Energy Research Institute (KAERI)
2
Daeduk College, 1045 Daedeok-daero, Yuseong, Daejeon 305-353, Korea
e-mail address of main author: [email protected]

This paper aims to describe the key factors regarding the design and
fabrication of a heavy weight concrete as a shielding material. The differences,
when compared to the concrete shielding blocks that are normally stacked
around neutron guides, are not only the need for a seismic integrity as a
structural material, but also a radiation shielding effect for the neutron and
gamma rays from the cold neutron guides. In general, heavy weight concrete is
not available as a normal product. That is why this study was conducted, namely
to fabricate a shielding material that meets the design requirements and technical
standards.
The cold neutron guide should be installed inside a biological shielding that is
assumed to be an absorber for neutron and gamma rays. Based on the suggested
material, a simulation for the shielding effect has been carried out to evaluate the
appropriateness of the proposed design. According to the simulation by help of
the Monte Carlo Method, the shielding materials were designed by using a heavy
weight concrete in 3.5 g/cc density. Because many kinds of instruments are
operating around beam ports, there is a lack of space to accommodate a
shielding structure in the reactor hall. So, the maximum shielding thickness
should be equal or lower than 800 mm. There is no other way except that the
concrete must have a higher than 3.5 g/cc density and a 35 MPa compressive
strength at 28 days.
The radiation shielding effect would be mostly affected by the physical and
chemical characteristics of the concrete. Generally, it is known that the shielding
effect depends on the kinds of aggregates to be composed in concrete. American
National Standard provides several cases as an example of typical compositions
of representative concretes after curing. Magnetite was considered, but it did not
have a proper density as a coarse aggregate, namely at least higher than 4.0 g/cc.
Therefore, barite composed of over 90% BaSo4 was adopted as a coarse
aggregate for the heavy concrete. The mixed proportion design for the heavy

196
 6. Design and Construction Issues

concrete was carried out by focusing on the homogeneous density at any part of
a structure and a quality control. After that, a mockup test to investigate the
possibility of segregation, a cracking by shrinkage and a thermal stress by
hydration heat, and workability was carried out. The mockup test was prepared
with a half scale at the same height and thickness as the shielding structure.
From the results, there is no evidence of a crack induced by shrinkage and a
thermal stress after a three months curing. Also, the heavy weight concrete,
which was made by a commercial batcher plant with a 100 ± 25 mm slump, can
be transferred into a mold at a nearly 60 m length by using normal pumping
equipment. The distance is simulated with the same pouring conditions of the
reactor building.
In order to increase the density of the concrete, it was better that the air
content is lower than the normal case. ACI provides that the air content is
defined as 4.5±1.5% for a normal case and 3.5% for a maximum size of an
aggregate at 20 mm and a mild exposure. Also, in ACI 301, the air content may
be additionally reduced by 1% in case of a compressive strength above 35MPa.
So, it has been determined that the air content at 2.5 ± 1% might cause the
density to be an average of 2% higher than that of the normal case. It has been
found that the heavy weight concrete satisfied the design requirements regarding
its density. And, the other characteristics for a structural material such as the
slump, compressive strength, air content, and chloride content have also satisfied
the design requirements. On the other hand, because this shielding structure was
designed with a thickness of 800 mm, we should consider controlling the
thermal stress induced by the hydration heat up of the mass concrete. Based on
the mockup test results, a thermal cracking might be induced at 1.5~2 days after
a pouring. It turns out that the heat up decreases after 8 days. So, it should be
resolved with protection layers, which are made of a fabric, normally used in
curing concrete, to retard the dry up speed of the water in the concrete just after
a pouring.
Conclusively, a fabrication procedure has been derived from the mockup test
and an analysis. The radiation shielding structure was installed at the designated
place for the neutron guides in August 2008. Although there are many
difficulties in a radiation controlled area, it sufficed that the work was
successively conducted without any technical violation by using the developed
procedures. It is hoped that this study will be a useful practice for other cases
that require a high density heavy weight concrete to satisfy not only a seismic
resistance but also a radiation shielding effect in nuclear facilities.

197
6. Design and Construction Issues

An efficient structural form for concrete

containment structures (6-1806)

L.R. Bishnoi
Scientific Officer (G), Atomic Energy Regulatory Board
Niyamak Bhavan, Anushaktinagar, Mumbai-400 094, India
e-mail: [email protected]

Background
Containment system of a nuclear power plant is an engineered safety feature to meet
one of the basic safety requirements of containing radioactivity. It derives
importance not only amongst the scientific community as the ultimate barrier against
release of radioactivity but also as a readily perceptible and hence psychologically
soothing protection as viewed by the general public. Structural design of
containments is governed by the design pressure value and the leak-tightness
requirement. Other important design loads are those associated with the external
events such as earthquakes, wind, airplane crash, and blast shock waves. Generally a
secondary containment is provided to resist loads associated with the external
events, except earthquake that pervades through every structure, and for other
functional aspects such as limiting the ground level releases during an accident.
Large dry concrete containments are the most popular because of the inherent
advantages of large volume and less maintenance. Since shell type structures
resist pressure loads efficiently, these containments are built as prestressed or
non-prestressed reinforced concrete (RC) shells. Concrete is strong in resisting
compression but very weak in resisting tension. To overcome this deficiency,
either pre-compression is introduced by prestressing to balance the tensile
stresses from pressure loading or large cross-sections are adopted to keep the
tensile stresses low and prevent through-thickness cracking. A metallic liner is
introduced, particularly for non-prestressed RC constructions, to improve leak-
tightness. Major drawbacks of prestressed RC constructions are the design and
construction complexities that add to the cost and time of the project, besides
additional monitoring requirements associated with the prestressing system.

Aim
The aim of this paper is to introduce a structural form that can be used for large
dry RC containments with significant advantage in terms of structural capacity
and leak-tightness. In the current proposal, the shells are so oriented that the
internal pressure loading induces net compressive stresses across the cross-
sections and keeps the inner face of the general region of the containment under

198
 6. Design and Construction Issues

compression that eliminates the need for any prestressing system. However,
inner faces near discontinuities develop tensile stresses. A structural framework
of columns and beams is introduced to support this structural form and to
localize the extent of inner face tensile regions around discontinuities, thus
limiting the need for metallic liner to small local regions. The framework may be
integrated with the outer secondary containment to augment its capability to
resist dynamic missile loading due to airplane crash.

Results
The proposed typical structural form is shown in Figure 1. Analysis results for
internal pressure loading concurrent with self-weight are shown in Figure 2. It
can be seen that the tensile stresses in the inner faces of the primary containment
are localized around structural framework members. The need for improving
leak-tightness is thus limited to these localized zones and this can be achieved by
providing metallic liner in these zones only.

Y
X Z

Z
Y Z
X
X
Y

(a) Inner containment (b) Plan and 3-D view of integrated structure

Figure 1. Proposed structural form for containment structure.

-.357E+07 -.151E+09
MN -.130E+09
.167E+08 -.106E+09 -.116E+09
MX

.370E+08 -.821E+08 -.804E+08

.573E+08 -.579E+08 -.452E+08

MN
.775E+08 -.338E+08 -.101E+08

.978E+08 -.969E+07 .251E+08

Y Y
Z X Z Z
X
X .118E+09 .144E+08 .603E+08
Y
MX
MN
.138E+09 .386E+08 .954E+08
MX
.159E+09 .627E+08 .131E+09

.179E+09 .868E+08 .166E+09

(a) 1st principal stresses (b) Meridianal stresses (c) Hoop stresses
Figure 2. Stresses under internal pressure.

199
6. Design and Construction Issues

Important considerations for design against missile effects of airplane crash are
the global vibration effects, the local structural deformations and the local
perforation. Global vibratory effects are hardly a design issue for containments.
Providing sufficient concrete thickness based on empirical relations ensures
safety against perforation. The structural performance of the outer containment
shell may be augmented against the effects of local deformations by integrating
the proposed structural framework of the primary containment to the outer
containment. Transient analyses carried out for missile loading associated with
airplane crash on cylindrical plane shell segments and stiffened shell segments
indicate that the stiffening effect of the structural framework can augment the
structural performance against local deformations.

Conclusions
An efficient structural form for containment structure is introduced that
enhances its structural and leak-tightness capabilities and augments resistance to
loading effects of airplane crash in an integrated structural complex of the
primary and the secondary containments. It needs to be explored whether the
structural framework could serve as the primary seismic resistance system so
that strengthening the framework without altering the standardized containment
geometry could accommodate any change in site dependent seismicity. Another
area for further work is to explore the effect of edge stiffening on the perforation
resistance of RC shells, slabs and walls.

References
1. Kennedy, R.P. A review of procedures for analysis and design of concrete structures to
resist missile impact effects. Nuclear Engineering and Design, Vol. 37, 1976,
pp. 183–203.

2. Kamil, H., Krutzik, N., Kost, G., Sharpe, R. An overview of major aspects of the aircraft
impact problem. Nuclear Engineering and Design, Vol. 46, 1978, pp. 109–121.

200
 6. Design and Construction Issues

Timber mat protection design for buried

utilities subject to impact loads (6-1809)

W. Johnson, M. Das, N. Gidwani, Jaspal S. Saini

Bechtel Power Corporation, Frederick, MD, USA

The heavy rigging operations necessary at nuclear plant sites during

construction, routine maintenance, and special activities such as steam generator
replacement require assessment of safety related underground utilities for the
effects of postulated accidental impact loads in the form of dropped loads, crane
boom drops, etc. Safety related utilities located directly beneath postulated
impact locations may require protection, which frequently is in the form of
timber crane mats. Sand bags, and soil or gravel mounds have also been used. To
design timber mat protection for a buried utility requires understanding of the
dynamic behavior and response of the coupled system comprised of impacting
element, timbers, supporting soil and buried utility. Calculation of such system
response is complicated by a number of factors, specifically a) the nonlinear
behavior of the crushed timbers which exhibit elastic, near perfectly-plastic
behavior up to a locking strain at which high-modulus strain hardening occurs
[Ref.1], b) the stacking arrangement of timber levels which results in tensionless
interstices on the timber sides, c) beam-on-tensionless-foundation behavior of
the timbers, d) soil compliance at the interface with the timber mat, and e) soil-
structure interaction between utility and soil. No evaluative process purporting to
address the combined system has been reported on in the literature. Quick
reference design aids are suitable for such temporary protective measures.
Indeed, alternative load paths would be opted for, before resort to the otherwise
necessary detailed analytical or FEM approaches.
This investigation develops design tables and charts, based on studies of cross
laid timbers resting on an elastic half-space, and dynamically impacted, end on,
by common impacting elements encountered during nuclear plant maintenance.
The analysis model for the soil/mat is comprised of a bar with locking material
behavior to represent the local dynamic crushing of the top wood timbers
[Ref. 1], timber beams on elastic foundation to represent the underlying layers of
timber, and Winkler springs to represent the supporting soil. This model is used
to relate the rigid impacting element geometric and dynamic properties to the
forcing function at the soil/mat interface. The interface forcing function is then
be used to determine a free-field soil response [Ref. 2] which is applied to a soil-
structure interactive representation of the soil and utility to determine the utility
response [Ref. 3].
Based on parametric evaluations for typical and realistic postulated missiles,
design tables and charts associating ground surface pressures and underground
pressures with impact energies are developed. These representations enable the

201
6. Design and Construction Issues

rapid design selection of the appropriate number of timber layers. An example

illustrating the application of the results to the evaluation of an underground
utility is provided.

References
1. Reid, S.R., Peng, C. Dynamic Uniaxial Crushing of Wood. International Journal of
Impact Engineering, Nos 5–6, 1997, pp. 531–570.

2. Laturelle, F.G. The Stresses Produced in an Elastic Half-Space by a Normal Step

Loading Over a Circular Area Analytical and Numerical Results. Wave Motion,
Vol. 12, 1990, pp. 107–127.

3. McClellan, R.E. Ground Shock Effect of Soil Field Inclusions. The Shock and Vibration
Bulletin, June 1984, pp. 203–208.

202
 6. Design and Construction Issues

Design of modular composite walls

subjected to thermal and mechanical
loading (6-1820)

S.R. Malushte1*, A.H. Varma2

1
Bechtel Fellow and Sr. Principal Engineer, Bechtel Power Corporation
Frederick, MD 21703-8306, USA
Tel. (301) 228-7697, Fax: (240) 379-2811
*
e-mail of corresponding author: [email protected]
2
Associate Professor, School of Civil Engineering, Purdue University
West Lafayette, IN 47906

Keywords: Modular, composite walls, thermal load, pressure, interaction

Modular composite (MC) walls consist of steel plates on the exterior to serve as
stay-in-place formwork and equivalent rebar. This type of construction is
attractive for reducing the schedule and field labor associated with massive
concrete construction projects (e.g., nuclear facilities). This paper presents
analysis and design approach for MC walls subjected to simultaneous thermal
loading and transverse applied loads. The applied loads can be due to accident
pressure or a seismic event. Thermal loading can be due to severe temperature
gradient across the wall thickness. Such gradient can be determined by
performing heat transfer analysis.
ACI 349 provides a simple treatment for determining the effects of
temperature gradient on a conventionally reinforced concrete wall. For MC
walls, similar techniques are not available that account for actual behavior under
thermal and mechanical loads. Based on experimental and analytical research,
the authors have developed an approach for determining the thermally induced
moment on MC walls. Unlike the ACI method, the proposed method does not
require determination of an equivalent (approximate) linear thermal gradient;
rather, the actual thermal gradient, which has a sharp nonlinear profile within the
first few to several inches of the wall thickness, can be directly used for
determination of the thermally induced moment.
For structural analysis of walls/compartments subjected to combined thermal
and mechanical loading, the authors have developed an approach that takes into
account the self-limiting nature of the thermally induced moment. The
subsequent design of the walls is thus not unduly conservative.

203
6. Design and Construction Issues

The effects of design parameters on

the thermal response of an LBE
capsule (6-1821)
Y.H. Kang, M.H. Choi, B.G. Kim, Y.K. Kim
Korea Atomic Energy Research Institute, 1045 Daedeok-daro,
Yuseong, Daejon, 305-353 Korea, e-mail: [email protected]

The development of a SFR (sodium fast reactor) as one of the advanced reactor
systems in Korea requires high temperature irradiation tests of new fuels, claddings,
and structural materials. To characterize the performance of these new materials, it is
necessary for us to have leading-edge technology to satisfy the specific test
requirements such as the conditions of high neutron exposures (~ 200 dpa), high
operating temperatures (390–700 ºC) and a specific chemistry (Na). The existing
design concept of a capsule using Al thermal media, however, is not satisfactory for
these high temperature tests. Thus, literature surveys about the system design
characteristics of various irradiation devices being developed or used in foreign
research reactors (i.e. ATR, MITR, JHR), which are helpful in understanding the
key issues for the on-going R&D programs related to a SFR, were conducted to
develop new design concepts. For the high temperature irradiation tests in the
HANARO reactor, the candidate thermal media as well as internal structural
materials as one of the capsule components should have a high thermal conductivity,
a high density and a very low reactivity which is needed to obtain the required
specimen temperatures, and results a small temperature difference within the
specimens. From an extensive survey of the literature, one of the candidate thermal
media is selected as an LBE (lead bismuth eutectic alloy) for the high temperature
irradiation devices.
Under the current HANARO capsule design practice, in order to evaluate the
relative significance of the various parameters on a thermal response, the
temperature calculations for the concept of a capsule using an LBE were performed
using a finite element analysis program, ANSYS. The concerned design parameters
such as the gap between the holder and the specimens (G1), the gap between the
LBE container and the external tube (G2), and the thickness of the specimen holder,
which are designed to effectively control the temperature of a specimen, are
considered as variables. The analysis model for a circular cylinder with multi
specimens is generated by the coupled-field elements of PLANE223 with a 2-D
structural-thermal field. The results of these studies indicated that the gap between
the LBE container and the external tube can have a great impact on the thermal
response. However, variations in the gap size between a specimen and a specimen’s
holder and the thickness of a holder material seem to have no significant effect on
the specimen temperature, and the LBE capsule concept can be applied to a high
temperature irradiation of new SFR materials in the HANARO reactor.

204
 6. Design and Construction Issues

An investigation on the fuel assembly

structural performance for the PLUS7
fuel design (6-1824)

Sang Youn Jeon, Kyou Seok Lee, Hyeong Koo Kim, Yuriy Aleshin, Alberto
Cerracin, Miguel Aullo Chaves Korea
Nuclear Fuel, Westinghouse Electric Company, ENUSA

The extreme level of fuel assembly bow can be the main cause of
IRI (Incomplete Rod Insertion), adverse effects on the nuclear design, or
handling difficulties impacting nuclear plant performance. In order to better
understand the mechanism of in-core fuel assembly structural performance, a
computer code (SAVAN) and methodology have been developed by ENUSA.
The SAVAN code analyzes the fuel assembly growth and bow using fuel
assembly design characteristics and in-core conditions. KNF, Westinghouse and
ENUSA jointly developed a new fuel assembly growth and bow computer
code (SAVAN2D) for the prediction of in-core deformation behaviour of the
PWR fuel assemblies. The SAVAN2D can be efficiently used to facilitate fuel
design development, core loading pattern optimization, fuel structural behaviour
prediction, and fuel loading/unloading sequence optimization.
The PLUS7TM skeleton and fuel assembly models for SAVAN2D analysis
have been developed based on the test results. The PLUS7TM core model has
been developed using fuel assembly model as a basic model. The out-core
mechanical characteristics of skeleton and fuel assembly and the in-core
structural behaviour of fuel assembly were analyzed using SAVAN2D computer
code and models. The load-deflection characteristics and deflection shapes of the
PLUS7TM skeleton and fuel assembly were compared with the test results to
verify the models. The in-core analysis results were compared with the measured
data to estimate the growth and bow characteristics of the PLUS7TM fuel
assembly. The analysis result shows a good agreement with the test result and
measured data and the PLUS7TM bow analysis results were very depend upon
the magnitude of initial bow and gap. It was concluded that the PLUS7TM fuel
assembly and core models can be utilized for the PLUS7TM out-core and in-
core structural performance analysis.

205
6. Design and Construction Issues

Friction coefficient measurement test

on 13MN class tendon of PC strands
for prestressed concrete containment
vessel (PCCV) (6-1825)

Tomoyuki Kitani1, Akira Shimizu2

1
Water Reactor Engineering Department, Nuclear Energy Systems Engineering
Center, MHI, Kobe, Japan, e-mail: [email protected]
2
Engineering Department, Nuclear Facilities Division,
Obayashi Corporation, Tokyo, Japan

In recent years, the pressure by the DBA of the containment vessel of nuclear
power plants tends to increase because the power generation capacity is getting
larger. Along with it, it is necessary that the tendon capacity used for the PCCV
is to be enlarged.
The tendon capacities used for the existing PCCVs are 10 MN or less class
and the values of the friction coefficient of 10 MN class tendon used to design
were obtained experimentally. However, the values of tendon which capacity is
more than 10 MN class have not been obtained experimentally yet.
Therefore, in order to confirm the friction coefficient of the 13 MN class post-
tensioning tendon which is intend to be used for future PCCVs, the friction
coefficient measurement tests using a full-scale mock-up structure of the PCCV
have been carried out.
The outline of the test is shown below.
(1) The friction coefficient measurement tests for the 13MN class tendon
were carried out using a full-scale mock-up structure of a PCCV
constructed.
(2) The mock-up structure is cylindrical RC structure of 23,960 mm in
internal radius, and the wall 700 mm thick and 2,750 mm high. The
friction coefficient measurement tests were executed using three tendon
sheath ducts embedded in this mock-up structure. As for the tendon
sheath ducts, each duct has 160 mm in internal diameter with zinc
electro-coated.
(3) The tendons used for the tests, whose capacities are 13MN class, are
made with 49 bundled No. 15 strands of ASTM A-416 Grade 1860MPa.
The primary coating of the corrosion preventing material is to be applied
to the tendons before the conducting of the tests.

206
 6. Design and Construction Issues

(4) The tests are to be executed according to the following procedures.

1) The tendon is installed by pushing the strands one by one through
the tendon sheath ducts embedded in the cylindrical wall over 360°.
Then the both ends of the tendon are anchored by the anchor heads
at the both sides of the buttress.
2) The hydraulic jacks are set at the both anchorages of the tendon.
Then the tendon is stressed about 1MN by the jacks on both sides
concurrently.
3) After that, the stressing load of the jack on the active side increases
until 10MN by 1MN. Receiving load at the anchorage of the
passive side is measured at each loading point.
4) The friction coefficient is calculated from the relationships between
the stressing load of the jack at the active side and the receiving
load measured at the passive side.
(5) The friction coefficients for curvature obtained by the tests results for
three tendons are 0.108/rad, 0.121/rad and 0.124/rad when the friction
coefficient for wobble is assumed 0.001/m.

207
6. Design and Construction Issues

Analytical study for failure probability of

PCCV under pressure load after seismic
experience (6-1826)

Tetsuya Okutani1, Yoshihiko Hino2

1
Project Development Department, The Japan Atomic Power Company
1-1. Kanda-Mitoshiro-cho, Chiyoda-ku, Tokyo 101-0053, Japan
e-mail: [email protected]
2
Engineering Department, Nuclear Facilities Division, Obayashi Corporation
2-15-2, Konan, Minato-ku, Tokyo 108-8502, Japan
e-mail: [email protected]

Purpose of the study

Recently, the importance of Probabilistic Safety Assessment (PSA) is increasing
because of growing the social demand to the nuclear safety, and the methods of
PSA are developed and improved. Meanwhile, almost all of the seismic PSA
have been executed as only the level-1 (the evaluation of the core damage
frequency) in the industrial world. The evaluation of the seismic PSA exceeding
the level-1 will be needed in the future. In the seismic PSA exceeding a level-1,
failure probability of containment vessel has to be evaluated. But, there are only
few examples and knowledge about it. In this study, analytical evaluation about
the failure probability of containment vessel is conducted and it aims at
establishment of the evaluation method in preparation for the seismic PSA
exceeding the level-1.

Contents of the study

In the seismic PSA exceeding the level-1, discharge probability of the
radioactive material has to be evaluated. Therefore, it is necessary to take local
failure of containment vessel into consideration to evaluate the failure probability.
In this study, using the FE method, the analyses in consideration of local failure
were conducted. The contents of analyses are shown in the following.
(1) Analytical Model
Analytical model was three-dimensional model in consideration of material
nonlinearity for actual PCCV. In order to evaluate the structural performance
of the PCCV as probability distributions, variations of the elastic coefficient
and the strength of construction material were taken into consideration.

208
 6. Design and Construction Issues

(2) Behavior of PCCV by FE analyses

Using the three-dimensional FE model, the analyses were conducted for
three cases: the seismic case, the severe accident case and the severe accident
after the seismic case. In the seismic case, the seismic response analyses
were conducted for the various levels of seismic load, and the whole and
local behavior of the PCCV were evaluated in each seismic level. In the
severe accident case, the inner pressure load by the sever accident were
applied to the model statically, and the whole and local behavior of the
PCCV were evaluated in each pressure level. In the severe accident after the
seismic case, after the seismic response analysis, where the remaining strain
and the crack of construction material were held, the pressure load of the
sever accident were applied to the model. Then, the whole and local behavior
of the PCCV were evaluated.
(3) Failure probability of PCCV
Evaluating the failure probability of the PCCV, it is important what kind of
condition is to be supposed the failure of PCCV. In this study, by the
evaluation of the PCCV behavior obtained from the FE analysis, various
failure modes were set up and the failure probability were evaluated for each
seismic level and each inner pressure level. In addition, the fragility
assessment conducted for each of the seismic case, the severe accident case
and the severe accident after the seismic case.

Results of the study

By the three-dimensional FE analysis, the whole and local behavior of the PCCV
in process of reaching to the failure condition were able to be clarified, in the
seismic case, the severe accident case and the severe accident after the seismic
case. Moreover, from these behaviors of the PCCV, failure probability of the
PCCV was evaluated for various failure modes. Consequently, the effective
evaluation method was shown in preparation for the seismic PSA exceeding the
level-1.

209
6. Design and Construction Issues

Deformation Contour Liner Strain Contour (Von-Mises)

Severe Accident after the Seismic Case Severe Accident after the Seismic Case
(1.0MPa after 1000Gal) (1.0MPa after 1000Gal)

References
1. Overpressurization Test of a 1:4-Scale Prestressed Concrete Containment Vessel
Model, NUREG/CR-6810 Report, USA, SAND2003-0840P.

2. Pretest Round Robin Analysis of a Prestressed Concrete Containment Vessel Model,

NUREG/CR-6678 Report, USA, SAND 00-1535.

3. Yonezawa, K., Imoto, K., Kato, A., Ozaki, M., Kiyohara, K., Murazumi, Y., Sato, K.
Pretest and Posttest Analyses for Nonlinear Behavior of 1/4PCCV Model
Subjected to Internal Pressure, E-298, The First FIB Congress in Osaka, 2002.

4. Ohba, M., Kawasato, T., Kato, A. et al. Analysis Results of a 1:4-Scale Prestressed
Concrete Containment Vessel Subjected to Pressure and Thermal Loading,
International Standard Problem No. 48 Containment Capacity, NEA/CSNI/R
(2005)5/Vol. 3.

5. Kawasato, T., Ohba, M. et al. Analytical Study for Ultimate Pressure Capacity of PCCV
Considering Thermal Effect, SMiRT19 (H06/3), Toronto, 2007.

210
 6. Design and Construction Issues

Civil engineering experiences from the

oversight of Olkiluoto 3 (6-1850)

Pekka Välikangas, Pertti Pitkänen, Jukka Myllymäki, Heikki Saarikoski

Radiation and Nuclear Safety Authority (STUK), Finland
e-mail: [email protected]

Introduction
This paper presents civil engineering experiences, which Finnish Radiation and
Nuclear Safety Authority (STUK) has received from the oversight work of
Olkiluoto 3. Feedback from the Olkiluoto 3 is presented in order to bring some
useful information to be considered in future projects for nuclear power plants.
Experiences presented in this paper are from nuclear safety authority point of
view.

Background
STUK is overseeing the construction of nuclear power plants by inspections and
supervision of design, component manufacturing and construction at the site.
Quality control and assurance by all players of Olkiluoto 3 has brought lot of
information in a form of non-conformance reports and audit findings. Some of
the non-conformance reports have been required by STUK based on inspection
findings, most of these reports are initiated by licensee and main supplier of the
power plant based on their own quality control and assurance.
Nuclear safety relevant and other essential technical questions are dealt with
correspondence between STUK and licensee of the Olkiluoto 3 in order to
ensure, that all solutions fulfil nuclear safety requirements and from licensee
point of view that these solutions are accepted by STUK. Important part for
efficient dealing of different, mostly complicated technical questions has been a
close cooperation between licensee and supplier of the nuclear power plant so,
that also specialists from STUK have participated in technical discussions.
STUK has also ordered independent studies and research work for ensuring
the quality requirements [1, 2]. A Finnish research programme SAFIR [3, 4] has
an important role for giving the state of art level scientific background in
different phases of nuclear power plant delivery.
STUK has been active on developing its’ own oversight of Olkiluoto 3.
Certain tangible changes to the methods and scope of inspections and
supervision have been made in order to be more proactive. Also the cooperation
with independent inspection organisations has been developed.

211
6. Design and Construction Issues

Essential results
Feedback from STUK’s studies and decisions and from non-conformance
reports of Olkiluoto 3 construction has been collected so, that it can serve in
future projects as well as in the further development of regulatory guides on
nuclear safety [5, 6].
Civil engineering experiences are presented by structures so, that is possible to
understand also the correspondent design criteria and cooperation in delivery
chain. Such structures and construction works are excavation works, common
base slab of nuclear island, prestressed reinforced protective shell of inner
containment, steel liner of inner containment and protective structures against
aircraft crash.
Important issue for civil engineering is cooperation between different
technical domain areas, like cooperation with process and piping design.
Vibration specific questions relating the framework of buildings and vibration
resistance of safety equipment of the nuclear power plant are also mentioned.
Civil engineering is part of the fire protection of nuclear power plant.
Experiences from Olkiluoto 3 fire protection issues are described in this paper as
well.

Summary
Experiences from Olkiluoto 3 civil engineering is the main issue of this paper. In
order to see the total picture, brief description of how these lessons have been
learnt in STUK will be presented. Feedback is collected by structures so, that is
possible to understand also the correspondent design criteria and cooperation in
delivery chain.

Selective references from standards, regulatory

guides and research studies, which are directly
related to the reported feedback/experiences:
1. Saarenheimo, A., Tuomala, M., Calonius, K., Lastunen, A., Hyvärinen, J., Myllymäki, J.
Numerical Studies on Impact Loaded Reinforced Concrete Walls, 2007.
International Association for Structural Mechanics in Reactor Technology. 8 p.
th
Transactions of the 19 International Conference on Structural Mechanics in
Reactor Technology, SMiRT 2007, Toronto, Canada, August 2007.

2. Saarenheimo, A., Tuomala, M., Hakola, I., Hyvärinen, J., Aalto, A., Myllymäki, J.
Impact of deformable missiles of reinforced concrete walls, Proceedings of
CONSEC'07. 5th International Conference on Concrete unser Severe Conditions:
Environment & Loading. Toutlemonde, F. et al. (eds.), Tours, France, June
2007. 8 p.

212
 6. Design and Construction Issues

3. Calonius, K., Hakola, I., Hostikka, S., Kankkunen, A., Lastunen, A., Saarenheimo, A.,
Silde, A., Tuomala, M. Experimental and Numerical Studies on Impacts. Räty, H.
& Puska, E.K. (ed.). 2006. SAFIR. The Finnish Research Programme on Nuclear
Power Plant Safety 2003–2006. Final Report. VTT Research Notes 2363. VTT.
Espoo (2006). Pp. 207–224. http://www.vtt.fi/inf/pdf/tiedotteet/2006/T2363.pdf.

4. Keski-Rahkonen, O., Mangs, J., Hostikka, S., Korhonen, T. POTFIS summary report,
http://www.vtt.fi/inf/pdf/tiedotteet/2006/T2363.pdf. Räty, H., Puska, E.K. (eds.).
2006. SAFIR. The Finnish Research Programme on Nuclear Power Plant Safety
2003–2006. Final Report. VTT Research Notes 2363. VTT. Espoo (2006).
Pp. 335–342. http://www.vtt.fi/inf/pdf/tiedotteet/2006/T2363.pdf.

5. Välikangas, P. QC and Regulatory oversight during construction, regulatory inspection

of design documentation of civil structures and components – practice in Finland
and experiences, Licensing and Regulatory Oversight of New Nuclear Build-
Workshop in Helsinki and Olkiluoto, 1–5 September 2008.

6. Management of safety requirements in subcontracting during the Olkiluoto 3 nuclear

power plant construction phase, Investigation Report 1/06, 10.7.2006. STUK,
Nuclear Reactor Regulation.

213
6. Design and Construction Issues

Seismic motion incoherency effects for

AP1000 nuclear island complex (6-1852)

Dan Mircea Ghiocel1, Dali Li2, Keith Coogler2, Leonardo Tunon-Sanjur2

1
GP Technologies, Inc., 6 South Main St., 2nd Floor, Pittsford,
New York 14534, USA, e-mail: [email protected]
2
Westinghouse, 3050 Northern Pike, Monroeville, Pennsylvania 15146, USA
e-mail: [email protected], [email protected],
[email protected]

The paper addresses the seismic motion incoherency effects on the soil-structure
interaction (SSI) response of the AP1000 nuclear complex. The paper addresses
both theoretical and practical aspects of seismic incoherent SSI analysis. The
paper describes briefly the theoretical basis and specific implementation aspects
related to the stochastic and deterministic incoherent SSI approaches. These
incoherent SSI approaches were benchmarked by Electric Power Research
Institute (Short, Hardy, Merz and Johnson, 2006 and 2007).
Two different structural models of AP1000 NI complex are considered: i) the
AP1000-based stick model (used in the EPRI studies) and ii) AP1000 NI20
finite element model (used by Westinghouse for computation of in-structure
response spectra). The AP1000 NI20 model was assumed for sensitivity studies
with both flexible and rigid basemat. Using AP1000-based stick model,
comparative results are shown for a hard-rock site and a soft soil site. Hard-rock
high frequency and Regularity Guide 1.60 ground spectra were considered for
the two site soil conditions. The recent Abrahamson plane-wave coherency
models for hard-rock and soil conditions were applied. However, it should be
noted that the soil coherence function is not accepted by US NRC at this time.
Only the hard-rock coherence function is permitted by US NRC.
The effect of foundation flexibility on the coherent and incoherent SSI
responses is discussed using the AP1000 NI20 model with flexible-basemat and
rigid-basemat, respectively. Finally, few incoherent SSI analysis recommendations
are stated.
The paper also describes the theoretical basis and key specific implementation
aspects related to the incoherent SSI approaches benchmarked by EPRI (Short,
Hardy, Merz and Johnson, 2006 and 2007) for performing incoherent seismic SSI
analyses for new nuclear plant designs. The implementation of all incoherent
SSI analysis approaches is based on the spectral factorization of the coherency
kernel. In AP1000 SSI studies, we considered both stochastic and deterministic
incoherent SSI approaches. In addition to stochastic simulation approach, three
deterministic approaches were considered: i) linear superposition, or algebraic
sum, of the scaled incoherent spatial modes (AS in EPRI studies), ii) quadratic

214
 6. Design and Construction Issues

superposition of the incoherent modal SSI complex response amplitudes

(transfer function amplitudes) assuming a zero-phase for the incoherent SSI
complex response phase (SRSS in EPRI studies), and iii) quadratic superpostion
of the incoherent modal SSI complex response amplitudes (transfer function
amplitudes) assuming a non-zero phase for the incoherent SSI complex response
that is equal to coherent SSI complex response phase (not used in EPRI studies).
The last implementation is an alternate version of SRSS approach that does not
neglect the complex response phase.
For rigid foundations the incoherency-induced stochasticity of the basemat
motion is driven by the global or rigid body spatial variations (integral
variations) of free-field motion and, therefore, is less complex and random than
free-field motion. The rigid foundation motion has a smoothed spatial variation
pattern since the kinematic SSI interaction is large. Thus, the differential free-
field motions are highly constrained by the rigid basemat, and because of this
(rigid body), the foundation motion complexity is highly reduced in comparison
with the complexity of the local motion spatial variations. For flexible
foundations, the incoherency-induced stochasticity of the basemat motion is
driven by the local spatial variations of free-field motion. The flexible
foundation motion has a less smoothed spatial variation pattern since kinematic
SSI is reduced. Thus, the differential free-field motions are less constrained by
the basemat, and because of this, the (flexible) foundation motion complexity is
similar to the complexity of the local motion spatial variations.
Based on our investigations, we noticed that due to their stochastic modeling
simplicity, deterministic SSI approaches are limited to rigid foundation
applications, as shown in EPRI studies. For flexible foundations, the stochastic
simulation approach is the only choice since it accurately captures the statistical
nature of the local free-field motion spatial variations. For flexible foundations,
the free-field motion local spatial variations are directly transmitted to the
flexible basemat motion. Deterministic approaches are not capable of capturing
the local phasing of the interaction motions. Thus, only the stochastic incoherent
SSI approach was used by Westinghouse for the final incoherent seismic SSI
analysis of the AP1000 nuclear island complex founded on hard-rock sites.
Based on the AP1000 NI seismic SSI studies done so far, we state two
important conclusions:
1) The effects of motion incoherency on the computed SSI reponse are
significant for both the rock and the soil sites. The in-structure
response spectra (ISRS) amplitude reductions were twice larger for
rock sites than for soil sites, as shown herein. We believe that more
study is worthwhile to propose and gain acceptance for the use of soil
coherence function by the US NRC.
2) The basemat flexibility reduces the kinematic SSI under incoherent
waves, and by this slightly decreases the motion incoherency effects
on the computed ISRS.

215
6. Design and Construction Issues

References
1. Abrahamson, N. 2007. Program on Technology Innovation: Effects of Seismic Motion
Incoherency Effects, Electric Power Research Institute, Palo Alto, CA and US
Department of Energy, Germantown, MD, Report No. TR-1015111, December.

2. Ghiocel, D.M., Short, S., Hardy, G. 2009. Seismic Motion Incoherency Effects on SSI
Response of Nuclear Islands with Significant Mass Eccentricities and Different
Embedment Levels. The 20th SMiRT Conference, Paper 1853, Helsinki, Finland,
August 14–18.

3. Short, S.A., Hardy, G.S., Merz, K.L., Johnson, J.J. 2006. Program on Technology
Innovation: Effect of Seismic Wave Incoherence on Foundation and Building
Response. Electric Power Research Institute, Palo Alto, CA and US Department
of Energy, Germantown, MD, Report No. TR-1013504, December.

4. Short, S.A., G.S. Hardy, G.S., Merz, K.L., Johnson, J.J. 2007. Validation of CLASSI
and SASSI to Treat Seismic Wave Incoherence in SSI Analysis of Nuclear
Power Plant Structures. Electric Power Research Institute, Palo Alto, CA and US
Department of Energy, Germantown, MD, Report No. TR-1015111, November.

216
 6. Design and Construction Issues

Out-of-plane shear strength of steel

plate concrete walls dependent
on bond behavior (6-1855)

Sung-Gul Hong, Seoul National University, Korea, e-mail: [email protected]

Kyung-Jin Lee (Principal Researcher, KEPRI)
Dong-Soo, Park (Senior Researcher, KEPRI)
Kyung-Won Ham (Researcher, KEPRI)
Han-Woo Lee (Korea Hydro and Nuclear Power Co)
25-1, Jang-dong Yuseong, Daejeon, Republic of Korea
Tel. +82-42-870-5741, +82-42-870-5749(fax), e-mail: [email protected]

Introduction
Double skin steel plate concrete wall structures (SC walls) have been considered
as one of viable options for speedy construction of nuclear power plants. For
reliable design and construction, structural behaviors of SC walls under service
loading condition including extreme cases need to be investigated by 3-year
research program in Korea.

Aim of the work

As one of research topics on their structural characteristics of SC walls, shear
strength of SC walls under out-out-plane shear focuses on difference of shear
transfer between ordinary reinforced concrete structures and SC walls. Review on
the design formulas proposed by JEAG 4618 requires further understanding of shear
behavior dependent on bond stress generated by studs and development of shear
strength taking into account of difference between ordinary reinforced concrete
members. This study proposes a modified shear strength models based on a new
interpretation of the effects of bond on the development of tensile forces in steel
plate due to arch action as well as the diagonal stress field between plates. The
proposed model focuses on the role of bond on arch action which is considered as a
primary out-of-plane shear transfer in SC walls of short a/d ratios. The diagonal
stress field for truss action combined with a single strut for arch action requires us to
consider reduced effective compressive strength of concrete in struts.

Essential results
To this end an arch action dependent on bond action by shear studs is
experimentally investigated and effectiveness factors of compressive strength of

217
6. Design and Construction Issues

concrete of direct strut for arch action are proposed. The contribution of tensile
strength and compressive strength of concrete to shear strength due to arch
action in short shear span ratio is investigated by assuming diagonal
compression field involving a strut. Series of experimental programs in this
study prepared specimens of different shear-span-to depth ratios, web
reinforcement steel ratio, and spacing of shear studs for verification of proposed
shear strength models with test results. Two different types of SC walls were
fabricated: specimens without ribs and specimens with ribs. In the first year
experimental program, tests for non-ribbed SC wall were performed focusing on
three important parameters: shear span ratio, plate thickness, and shear
reinforcement ratio. Also two different loading patterns were applied: double
curvature systems and single curvature systems within shear span.

Conclusion
The proposed formula show good agreements with the test results including those
from JEAG. With consideration of bond and its effect on strut strength for arch
action the proposed shear strength model is able to predict shear strength of SC
walls. The strength models consider effects of bond strength on the interface
between concrete and plates with determination of the width of strut and
effectiveness factor for strut in biaxial stress state. Also, it is necessary to clearly
define shear span to beam depth ratio for the slope of diagonal strut explaining arch
action irrespective of single or double curvature of moment distribution. Otherwise,
conventional a/d ratio has confused engineers blindly following design formula
without understanding of shear transfer by arch action. Increase in flexural capacity
by ribs requires higher shear strength for ductile failure. Strain distribution in plates
indicates that assumption on average bond stress, μavg , in the proposed model be
reasonable. Experimentally observed strain distribution in shear bars and ribs
showed stress states expected from shear strength model.

References
1. Steel and Concrete structure for earthquake design, edition for structures (2005).
JEAG 4618.

2. Nielsen (1999). Limit Analysis and Concrete Plasticity, Second Edition. WILLY

3. First Report and Plan for Out-of-Plane Shear Experiment Program and Plan (2006).
KEPRI.

4. Oehlers, D.J., Bradford, M.A. (1999). Elementary behavior of composite steel and
concrete structural members. PLANT A TREE.

218
 6. Design and Construction Issues

Development of the simplified fuel

assembly model for the fuel assembly
SSE and LOCA analysis (6-1858)

Kyou Seok Lee1, Sang Youn Jeon1, Hyeong Koo Kim1

1
Korea Nuclear Fuel
493, Deogjin-Dong, Youseong-Gu, Daejeon, 305-353, Korea
Tel: +82-42-868-1185, Fax: +82-42-868-1149, e-mail: [email protected]

Introduction
The load on the fuel assembly under accidents and postulated events like Safe
Shutdown Earthquake (SSE) and Loss of Coolant Accident (LOCA) events shall
not results in permanent deformation that would prevent effective emergency
cooling of the fuel or that would prevent safe reactor shutdown. Under the SSE
and LOCA events, the fuel assembly lateral deflection and grid impact force
between fuel assemblies are obtained by the dynamic transient analysis for the
reactor core finite element model.
The impact behavior between fuel assemblies shows non-linear
characteristics, because fuel assembly shows non-linear dynamic characteristics
and its structural geometry is complicated. Furthermore, since a reactor core
consists of a large number of fuel assemblies, the dynamic behavior of the core
under the postulated events is very difficult to analyze due to the nonlinearity
and huge model size. Therefore, it is necessary that fuel assembly model is
simplified with considering the dynamic non-linear characteristics in core
analysis. Until now many researches for the simplification of fuel assembly
model have been performed.
In this study, a simplified fuel assembly finite element model for the
Westinghouse type 17 × 17 RFA has been developed. To obtain the simplified
model, the optimization algorithm of ANSYS code was used, and the
configuration of the model was determined by the sensitivity study. The
simulations for static test, pluck vibration test, pluck impact test were performed
using the model. The simplified fuel assembly model was verified by
comparison with fuel assembly mechanical test results.

Analysis results and discussions

The object function of optimization problem for the simplified fuel assembly
model was defined as the minimization for the difference between the test
natural frequency of fuel assembly and optimized natural frequency of the

219
6. Design and Construction Issues

model. The constraints of the problem were defined as tolerance range for
frequency at the each mode.
In order to obtain the optimized model, the design variables of model
considered the beam rigidity (I) for fuel rods and guide thimbles, and the
rotational rigidities (RT and RB) at top and bottom end to simulate the interface
rigidities between top/bottom nozzles and core plates. And, the rotational springs
(KT: top grid, KB: bottom grid, KM: mid grid, KI:IFM grid) between each grid
and top and bottom ends were considered to simulate the friction resistance
between grid spring and fuel rod. The configuration of the model is shown in
Fig. 1.
RT Fixed
Top Nozzle
Top Grid
Mid Grid KI
IFM Grid

KT
: :
: : KM
: :

Bottom Grid
KB
Bottom Nozzle
RB Fixed

(a) 17×17 RFA (b) Simplified Model

Figure 1. Configuration of Simplified Model.

The sub-problem approximation method in ANSYS was used for the

optimization of model, and the design variables for beam rigidity and the
stiffness for each rotational spring were optimized. After preliminary optimization,
the fixed boundary conditions at top and bottom end were considered, because
RT and RB were shown too high. The optimized results of the model are shown
in Table 1 and Fig. 2.

Table 1. Comparison of Natural Frequencies with Test Results.

Mode 1st 2nd 3rd 4th 5th 6th

Frequency Test 3.65 7.8 12.3 17.7 24.2 30.5
(Hz) Analysis 3.65 7.89 12.53 18.44 24.38 34.12

220
 6. Design and Construction Issues

1.0
1.0

0.0 0.0
0 20 40 60 180 100 120 140 160 0 20 40 60 80 100 120 140 160

Model Model
Test Test
-1.0 -1.0

a) 1st Mode b) 6th Mode

Figure 2. Comparison of Mode Shapes with Test Results.

In order to verify the model, the simulations for static test, pluck vibration test,
pluck impact test were performed. The results are shown Fig. 3, Fig. 4 and Table
2. Transient analysis for core model was performed too. The maximum impact
force was shown to be 2% differential compare with current design code results.
0.10

Test
Test
Model
Analysis

0.05

0.00
0 20 40 60 80 100 120 140 160

Figure 3. Comparison of Static Analysis with Test Results.

0.8

0.6

0.4
Amplitude

0.2

0.0
0.0 0.1 0.2 0.3 0.4 0.5

-0.2

-0.4
Time(sec)

(a) Test (b) Analysis

Figure 4. Comparison of Pluck Vibration Analysis with Test Results.

221
6. Design and Construction Issues

Table 2. Comparison of Pluck Impact Force with Test Results (unit: lbs).

Items Grid 3 Grid 4 IFM 1 Grid 5 IFM 2 Grid 6

Test 1,255 1610 295 1325 - 910
Analysis 1431 1544 334 1517 284 1309

Summary
The simplified fuel assembly model for fuel assembly SSE and LOCA analysis
has been developed with ANSYS code to evaluate the structural integrity of
17 × 17 RFA fuel assembly under SSE & LOCA events. The model has been
developed using optimization method and design variables have been
determined by sensitivity studies. The optimized model has a good correlation
with fuel assembly mechanical test results and in-reactor impact behavior of
current model.

Reference
Kim, H.K. (2001). Development of core seismic analysis models for KNGR fuel
assemblies associate with 0.3g seismic loads. Nuclear Engineering and Design
212 (2002), pp. 201–210.

222
 6. Design and Construction Issues

Concentration of plastic strain in the steel

liner near the equipment hatch in a
1:4 scale prestressed concrete
containment model (6-1903)

Patrick Anderson, Ola Jovall

Scanscot Technology AB, Lund, Sweden

A substantial part of nuclear reactor containments in US and Europe are

designed with an outer bearing concrete structure and an inner sealing consisting
of a tight-welded steel liner. The liner constitutes the ultimate leak-barrier which
prevents leakage at high internal pressure loads.
It has been shown in containment scale tests that the global displacement
measured at liner failure do not correspond to the critical strain level for the
liner. A general conclusion from these results is that some type of strain
concentration has to take place to get this “early” liner failure. In work carried
out at Lund University non-uniform deformation of concrete and discontinuities
such as penetrations is concluded to be important issues giving concentration of
strain in steel liners (see [1], [2] and [3]).
In year 2000 a 1:4-scale containment model was tested by over-pressurization
at Sandia National laboratories (Sandia 1:4, see [4]). The first leak in this test
was concluded to be caused by tears in the steel liner found near the equipment
hatch. In work made at Lund University, a finite element model of the region
near the equipment hatch show that high localized strains will develop in the
vicinity of the bend line (see [2] and [3]). In this local study the liner anchor
profiles (connecting the liner to the concrete) is assumed to follow the uniform
global expansion measured in the containment test model, i.e. the influence of
non-uniform displacement due to concrete cracking is not included.
In the study presented in this paper the model presented in [2] and [3] is
improved by including the non-linear behavior of concrete. The sub-model is
also improved by using more realistic boundary conditions given from a global
finite element model of the Sandia 1:4. The global FE-model was made by
Scanscot Technology within the ISP-48 project (see [5]).

223
6. Design and Construction Issues

References
1. Anderson, P. Concentration of plastic strains in steel liners due to concrete cracks in
the containment wall. International Journal of Pressure Vessels and Piping,
Vol. 85, pp. 711–719, 2008.

2. Anderson, P. Analytic study of the steel liner near the equipment hatch in a 1:4 scale
containment model. Nuclear Engineering and Design, Vol. 238, pp. 1641–1650,
2008.

3. Anderson, P., Jovall, O. Increased plastic strains in containment steel liners due to
concrete cracking and discontinuities in the containment structure., SMiRT 19,
H01/2, 2007.

4. Hessheimer, M.F., Klamerus, E.M., Lambert, L.D., Rightley, G.S. Overpressurization

Test of a 1 to 4 Scale Prestressed Concrete Containment Vessel Model, NRC
report, Sandia National Laboratories, NUREG/CR-6810, 2003.

5. International Standard Problem 48 Containment Capacity, Phase 1–3, OECD, 2005.

224
 6. Design and Construction Issues

Structural design of replacement

emergency core cooling filtration
system (6-1907)

George Stoyanov, Ashok Kanade, Ravi Jategaonkar

Atomic Energy of Canada Limited, Canada, e-mail: [email protected]

Keywords: modular Finned StrainerTM, AECL, ECC, strainer, seismic, submerged,

hydrodynamic, LOCA, thermal

This paper summarizes the results from recent design experience of ECC
(Emergency Core Cooling) strainers. ECC strainers are designed for use in
emergency situations (LOCA event) to filter debris from the water used to
maintain the cooling of the reactor core. After the incident at Barsebäck-2, a
Swedish BWR, nuclear regulators around the world started re-evaluating ECC
systems in place at operational NPPs. Considerable research has been done to
determine adequate design parameters. Based on new debris definitions it was
determined that present filtering systems do not have sufficient capacity to
provide reliable coolant circulation in case of a LOCA. The existing filtration
surface was found to be very small compared to the one needed per new
research. As a result of this new design solutions had to be implemented and
AECL has developed its modular Finned StrainerTM design. This strainer
consists of porous fins attached to a common header and can be modified to fit a
wide variety of conditions. The design process for replacement of containment
sump strainer is a very complex one due to limited spacing in RB basements and
the need for compact and at the same time adequate in suction surface strainers.
The challenge for complex design was not only space limitation, but also
significant suction pressure and temperature variations. In addition to this,
design requirement for considering a seismic event during or following a LOCA
imposes that seismic design be done for submerged under water conditions. The
performance of the equipment had to be evaluated for all those loads, the major
of which are suction pressure, temperature and seismic in submerged conditions.
Each one of those is challenging on its own, but their simultaneous presence
further added complexity to the problem. The suction pressure major challenge
was the need for relatively large surfaces with limited options for stiffening due
to hydraulic flow limitations to avoid flow blockage. Thermal elongation was
significant and because of the need for long suction trains, modularization and
special consideration to sealing had to be given, together with allowing thermal
expansion within each module. Seismic design under submerged condition
requires that hydrodynamic loads due to fluid – structure interaction be considered
in addition to seismic inertia load due to selfweight. In the area of fluid structure

225
6. Design and Construction Issues

interaction, a number of research studies have been published, however, there

are no prescriptive design standards. The event that causes the most critical
loading combination is a Safe Shutdown Earthquake occurring during a Loss of
Coolant Accident (LOCA) i.e. while the strainer is in a submerged condition.
The ability of the strainer to perform its safety function during and after this
event has been demonstrated by analysis. Its ability to function during and/or
following one safe shutdown earthquake (SSE) event, preceded by a number of
operation basis earthquake (OBE) events has been demonstrated as well. Special
design of interconnecting ducts and pipes to the pumps had to avoid imposing
any additional loads on the pump inlet. Due to congested environment, drilling
limitations and the presence of Reactor Building liner, supporting and anchoring
the equipment was a challenge itself. This paper describes the design experience
of the Finned StrainerTM on a number of NPPs. As a result of approaches used all
challenges were successfully overcome and a reliable and robust design was
produced by AECL.

226
 6. Design and Construction Issues

Assessing the reliability of seismic base

isolators for innovative power plant
proposals (6-1918)

Leone Corradi1, Marco Domaneschi2, Chiara Guiducci1

1
Politecnico di Milano, Enrico Fermi Center for Nuclear Studies (CeSNEF)
Department of Energy, via Ponzio 34/3, 20133 Milano, Italy
e-mail: [email protected]
2
Politecnico di Milano, Department of Structural Engineering
Piazza Leonardo da Vinci 32, 20133, Milano, Italy

Most innovative power plant proposals are conceived so as to eliminate or

reduce considerably, at the design level, the occurrence of internal accidents. As
a consequence, the core damage frequency associated to internal events is
significantly decreased, to the point that values below 10-8 events per year often
are declared. In this situation, external events (such as earthquakes) become
dominant and reducing their damage probability presently is a challenge. In
particular, for seismic events a still unsettled problem is the definition of the
design earthquake, expected to be significantly more severe that nowadays
SSE’s, to be used when so low values for the damage frequency are demanded.
A possibility that presently is under study refers to the introduction of base
isolators, consisting of laminated rubber bearings capable of very large
deformations. Such isolators have been successfully employed in several
instances for buildings, bridges or tanks operating in seismic areas and it is felt
that similar benefits could be gained in the nuclear context as well. The
experience gained so far is significant, but is limited to ground motions
associated to the earthquakes that are presently considered for design purposes.
Whether such devices would maintain their effectiveness for more severe
seismic inputs is a question still open, which is felt to deserve some study. This
paper is intended as a first step in this sense. Rather than facing the problem of
the definition of the design earthquake, a probabilistic reliability assessment is
performed on a typical bearing, with the aim of determining the maximum peak
ground acceleration (compatible with a given design spectrum) that can safely be
undergone.
Analyses are performed on a simplified model which considers rubber as an
incompressible hyperelastic material undergoing large strains and metal disks as
rigid. The model is simple enough to permit the expression of displacements,
stresses within the rubber and at the rubber-metal interface as functions of the
material parameters by means of comparatively simple formulas, for different
values of the vertical and horizontal forces acting on the isolator.

227
6. Design and Construction Issues

Such forces were computed for different ground acceleration time histories,
referring to earthquakes of increasing amplitude and are available as input data.
They are used to define families of fragility surfaces for the isolator, to be used
as a basis for its reliability assessment under different earthquakes. As limit
situations, the following are considered:
− Excessive horizontal displacement
− Limit strength in the rubber
− Strength of the rubber-metal interface bonding.
Such limits are treated as random variables, as the material constants and the
parameters governing the internal dissipation of the rubber. A Monte Carlo
procedure is adopted to evaluate the failure probability of the device for different
peak ground accelerations.
The results presented refer to preliminary computations, only a first step toward
a reliable assessment of the earthquake magnitudes that do not jeopardize the
isolator functionality. Even at this early stage, however, the effectiveness of the
procedure is apparent.

228
 6. Design and Construction Issues

Low-activation concrete design

methodology for reducing radioactive waste.
Categorization of low-activation concrete by
low-activation factor (6-1923)

Ken-ichi Kimura*1, Masaharu Kinno2, Akira Hasegawa3

1
Fujita Corporation, Technical Development Division
2025-1,Ono,Atsugi, 243-125 Kanagawa, Japan
2
Fujita Corporation, 3Tohoku University

Introduction
Concrete is very valuable and inexpensive material, however it can be changed
to be expensive and hard to deal with in use of a reactor after long operation.
One of the counter plans for the above is to use low-activation concrete instead
of the ordinary concrete, that will reduce radioactive waste and could be even
below clearance level in decommissioning and that is very useful in term of life
cycle cost. Instrumented neutron activation analysis showed that Co and Eu were
the major target elements which decide the radioactivity level of reinforced
concrete in decommisioning stage, and a several material were selected as a low-
activation raw material from wide survey of raw materials for concrete (typically
aggregates and cements). With the canditate of raw materials, several low-
activation concrete were proposed for various portion of the reactor, which
reduction ratio were 1/10 to 1/30 which were mainly consist of limestone and
low heat cement or white cement, comparing to the ordinary concrete in ΣDi/Ci
uinit, where “Di” indicates concentration of each residual radioisotope, Ci
defined by IAEA as a clearance level, and suffix of “i” indicates each
radioisotope.

Aim of the work

National funded project for development of low-activation design method for
reduction of radioactive waste below clearance level were started from 2005 to
2009 with aiming (1) development of a database on the content of target
elements, which transform radioactive nuclides, in raw materials of reinforced
concrete, (2) development of calculation tools for estimation of residual
radioactivity of plant components, and (3) development of low-activation

*
Presenting author, e-mail: [email protected]

229
6. Design and Construction Issues

materials for concrete such as cements and reinforcing steel bars for structural
components. So, development for low-activation concrete design reducing
radioactive waste have been conducted in the project mentioned above, and 13
papers were presented in the last SMiRT19. For the applying low-activation
concrete to the real reactor portion, effective evaluation of every uncertainty
between the designed concrete and execution concrete are necessary. So, the
comparison of calculated activation for the mix proportion based on raw data, to
the concrete by the mix trial was conducted in this paper.

Essential results
The projects have conducted several results, such as development of new low
activation cement, additives,and reinforcement bar, material data base including
more than 2000 law materials, low-activation material development system and
activation mapping system. In additions, aout 100 mixture propotions of low
activation concrete by tons of execution experimental works for several types of
low-activation concrete, which were conducted to categolize as conventional,
high performance and boron added, with reduction rate of radioactivity to the
ordinary concrete from 1/10 to 10000.
The materials use for Low-Activation concrete are, white cement and low
heated cement for the cement, pure limestone for the aggregate, and limestone
powder and silica fume for the additives. These materials were selected by the
wide survey of instrumented neutron activation analysis. Pure limestone was
used for fine and course aggregate with the different density. The trial mixings
were conducted for 2006 to 2008, and materials were ordered two times to four
times. Physical properties as a raw material for concrete were tested with the
confirmation of Low-activation by Instrumented neutron activation analysis.
Low-Activation Factor (LAF) is tentatively defined as the ratio of the ΣD/C
for Low-Activation concrete to the ordinary concrete in the assumption of
certain neutron field condition with Clearance level by IAEA. On the other
hands, each trial mixing test produced many samples for several tests, in order to
confirm the proper physical properties as structural concrete in the reactor, as
well as Low-Activation performance. This paper described the relationship of
Low-Activation Factor to the cement content per unit volume of concrete in the
mix proportion, compressive strength of concrete, and dried density. The three
distributions of above relationship to the LAF has similar tendency that higher
LAF become the lower values (the cement content per unit volume of concrete
in the mix proportion, compressive strength of concrete, and dried density). In
additions, the comparison of Low-Activation Factors of measured data by
instrumented neutron activation analysis of executed trial mixing concrete and
those of calculated data by instrumented neutron activation analysis of raw
materials and mix proportion were conducted, and average value of the ratio of
LAF for measured data to that for calculated data is close to 1.

230
 6. Design and Construction Issues

Conclusions
More than 50 kinds of developed Low-Activation concrete were categolized by
defined Low-Activation Factor (LAF). The relationships of LAF to the typical
phisical propertise of concrete, which were the cement content per unit volume
of concrete in the mix proportion, compressive strength of concrete, and dried
density, were evaluated. It was clearfied that these relationships had similler
tendency.
This work is supported by a grant-in-aid of Innovative and Viable Nuclear
Technology (IVNET) development project of Ministry of Economy, Trade and
Industry, Japan.

References
Hasegawa, A. et al. Low-Activation Reinforced Concrete Design Methodology (1) –
th
Overview of the project – 19 International Conference on Structural Mechanics
in Reactor Technology, Toronto, Canada, 2007.

Kinno, M., Kimura, K., Nakamura, T. Ultra-low-activation limestone for neutron irradiation.
ANS Radiation Protection and Shielding Conference, pp. 673–678, Spokane,
USA, 2002.

231
6. Design and Construction Issues

Concrete shrinkage taken into account as

crack width assessment (6-1924)

Etienne Gallitre1, Pierre Alain Naze2, Pierre Labbe3

1
EDF-SEPTEN civil work section manager and reinforced concrete researcher
2
EDF-CNEN civil work section manager and reinforced concrete researcher
3
EDF-DIN special seismic and civil work expert and IAEA adviser

Context
In very large buildings with connected walls, such as large Nuclear Power Plants,
concrete shrinkage strains have to be considered because of elements differential
strains, as required in the new European construction code (EC 2). The fastest
engineering method consists in considering shrinkage as an equivalent thermal
strain, which is in fact computed as internal forces. In EPR, for Flamanville 3
conditions, this first method led to enormous reinforcement ratio in the lower
part, so EDF with its partners proposed a new methodology based on crack width
assessment.

Methodology summary
At first, we compute the shrinkage differential strains, “εs”, depending on
moisture conditions and elements thickness. Then we fix a reinforcement section
as a calculation hypothesis in order to estimate the distance “Ss” between cracks,
which is independent from the loads. Consequently, we can assume that a certain
crack width value “Ws” is consumed by the shrinkage itself, with Ws = εs Ss
Rax (Rax restriction factor). So the available crack width for the other loads is
the remaining crack width. From cracking theory and according to EC2; we can
deduce steel stress σd. So the structure design (reinforcement mainly) can be
undertaken with this allowable limit value σd., in the load combinations where
shrinkage as to be considered.

Conclusions
This new methodology is more physical than the one with a thermal equivalent
load, so safety requirements are satisfied in focusing on crack width assessment,
which is a performance approach. It has allowed EDF to size a reinforcement
ratio, which is compatible with concrete technical rules, especially in the areas
near the raft. Of course this method remains in accordance with durability
hypotheses and other requirements connected to nuclear specificities.

232
 6. Design and Construction Issues

Design of suspended ceilings in main

control room of units 5 and 6 of Kozloduy
NPP (6-1933)

Dimitar Tanev1, Penka Sofronieva2, Marin Jordanov3

1
Structural and Seismic Engineer, EQE Bulgaria AD
H. Smirnenski Blvd. 1, Bulgaria, Sofia, e-mail: [email protected]
2
Structural Engineer, EQE Bulgaria AD
H. Smirnenski Blvd. 1, Bulgaria, Sofia e-mail: [email protected]
3
Structural and Seismic Engineer, EQE Bulgaria AD
H. Smirnenski Blvd. 1, Bulgaria, Sofia, e-mail: [email protected]

In 2007 was realized EQE Bulgaria project for new suspended ceilings in Main
Control Room of Units 5 and 6. The project purpose was to be replaced the old
suspended ceilings with new ones, which are modern with broken up design, on
different levels and with curved outlines. Ergonomic lighting with controlled
brightness was built in the ceilings. It was also improved the air exchange and
the climate in the rooms, thanks to the new built in air-conditioning system.
The bearing structures of the new suspended ceilings were qualified for new
seismic loads.
The project was implemented in short terms and resulted in improved comfort
and ergonomic working environment for the operators in Units 5 and 6 of
Kozloduy NPP.

233
6. Design and Construction Issues

Soil remediation for seismic design of

independent spent fuel storage installation
(ISFSI) pad (6-1935)

Tripathi, Bhasker (Bob) P.

United States Nuclear Regulatory Commission
Mail Stop: EBB-3D-02M, Washington, DC 20555-0001, USA
e-mail: Bhasker.Tripathi@nrc,gov

Keywords: ISFSI Pad, seismic, soil remediation, soil mixing, compaction, grouting

Interim spent fuel storage, using a U. S. Nuclear Regulatory Commission (NRC)

approved Dry Cask Storage System (DCSS) is an acceptable means of spent fuel
management, until the U. S. Department of Energy accepts and stores the spent
nuclear fuel in a high-level waste repository. The DCSS’ for Independent Spent
Fuel Storage Installation (ISFSI) are massive steel and/or concrete structures,
loaded with spent nuclear fuel, and stored (in most cases unanchored) outside on
reinforced concrete pads. The storage cask vendors have specific requirements
for critical soil parameters under the reinforced concrete pads.
Requirements for critical soil parameters under the reinforced concrete pad
foundation supporting the DCSS have to be met to ensure stability of the pad
when challenged by natural phenomena such as seismic events. These requirements
vary for different vendors and are described in the vendor-specific Certificate of
Compliance (CoC) issued by NRC. In instances where the existing soil is
vulnerable to potential liquefaction, and/or settlement because of a design-basis
seismic event, various approaches could be used by licensees to stabilize the
natural soil. The reinforced concrete pad foundation is required to satisfy the
safety objectives of Title 10 of the Code of Federal Regulations (10 CFR) Part
72. Regulatory Guides (RGs), NUREGs, Standard Review Plans (SRPs) and
other guidance documents are available to assist an applicant in complying with
the regulations.
This paper will: 1) provide an overview of selected approaches that could be
used for meeting the seismic demand on the storage pad for ISFSIs licensed
under the provisions of 10 CFR Part 72.210; 2) discuss the design process and
compare the relative merits of these approaches; 3) investigate the potential
effects of soil remediation; and 4) discuss the resulting soil-structure interaction
effects and the seismic input ground motions appropriate for the design of the
foundation.

234
 6. Design and Construction Issues

Improving constructability of the new

generation nuclear construction through
improvements in design efficiency and use
of high-strength reinforcement (6-1937)

Javeed Munshi
Principal Engineer, PhD, SE, PE, Bechtel Power Corp., Frederick, MD 21703

Preliminary indications are that new generation nuclear structures will need
significant amount of reinforcement in order to make them conform to the
current regulatory requirements, Codes and Standards. The concern is that
amount of reinforcement required is likely to cause congestion and consequent
concrete placement problems thus impacting constructability of new generation
nuclear plant structures. The factors that contribute to the reinforcement
requirements include:
1. Creeping loads due to revised seismic hazard criteria/ground motion data.
2. Standard plant concept which requires that a standard plant be designed for
a suite of soil conditions ranging from very poor soil to rock site.
3. 2-Step Method of Analysis which involves using the envelop of seismic
forces (irrespective of the fact that they may occur at different times) from
the soil structure interaction model analysis as static input to structural
analysis model for design of concrete elements.
4. Interpretation of finite element results – because of lack of a rational
methodology, the following process is currently adopted for design of
concrete walls.
i. The design is based on envelop of element forces which may occur at
different times during the analysis.
ii. The out-of-plane bending is assumed to occur simultaneously with the
in-plane forces. Note that out-of-plane bending of wall is generally a
high frequency mode and does occur simultaneously with in-plane shear.
iii. A sectional methodology is applied to design elements in the vertical
and horizontal direction which are assumed to act independently as
columns.
iv. In general, the reinforcements required for in-plane shear and
membrane normal in-plane and out-of-plane forces are determined
independently and added to get the total reinforcement for the walls.

235
6. Design and Construction Issues

The larger design loads combined with many layers of conservatism discussed
above add to the required reinforcement and potential congestion. The paper
discusses factors that will help reduce the overall conservatism in the
analysis/design process. The incompatibility of using the FEM analysis results
with design approaches prescribed in the Codes/Standards is discussed. Based on
this discussion, a rational and transparent analysis/design approach aimed at
reducing the overall conservatism is outlined.
In order to reduce the reinforcement congestion and improve constructability,
use of high-strength reinforcement (Gr 75 as a minimum) is proposed in nuclear
construction. Note that Grade 75 reinforcement is permitted in ACI 318 under
ASTM A615 and has been in use for over 20 years now. The chemistry control
and bend requirements of Grades 60 and 75 are essentially the same. The
elongation requirement, which is used as a measure of ductility, is 7% for Grade
60 and 6% for Grade 75 for No. 9 and larger bars. Note that No. 9 and larger size
bars are predominantly used in nuclear construction.
ASTM A615 Grade 60 reinforcement has been successfully used in nuclear
and defence construction to resist both seismic as well as impact loads. Since
nuclear structures are designed to remain essentially elastic during a design
seismic event, a marginal reduction in elongation requirement [from 7% to 6%
elongation] should not be an issue for use of Grade 75 reinforcement. For impact
type of loads, the expected ductility of concrete elements with Grade 60
reinforcement is given in Appendix F of ACI 349. Although this appendix calls
for use of A706 reinforcement, the allowable ductility values are still based on
ASTM A615 reinforcement used previously. As a result of this, no significant
changes are necessary for use of Gr 75 reinforcement for impact resistance. Note
that European regulations/Codes also permit use of equivalent Gr 75 reinforcement
in nuclear construction.
Use of high-strength reinforcement in nuclear construction is consistent with
recent trends in the construction industry. Several high profile projects using up
to 100 ksi reinforcement have recently been showcased in some major industry
magazines. Furthermore, there is a proposal in ASTM to get ASTMA706 Gr 75
and Gr 80 reinforcement standardized. The ACI Task Group ITG-6 is proposing
a design yield strength of 100 ksi for flexural tension reinforcement, 80 ksi for
compression reinforcement, 60 (or possibly 80) ksi for shear reinforcement, and
100 ksi for confinement reinforcement in columns and shear walls. Note that 100 ksi
for confinement reinforcement is already accepted in the current ACI 318 Code.
The design process improvement outlined in this paper will help reduce the
overall reinforcement requirement thus improving design efficiency and reducing
potential for congestion. The use of high-strength reinforcement will further
improve constructability by allowing larger spacing between reinforcement to
ensure quality concrete placement that is warranted in nuclear construction.

236
 6. Design and Construction Issues

Implementation of high-performance
concrete in the ACR-1000 containment
structure for 100 year design life (6-1969)

Homayoun H. Abrishami1, Medhat Elgohary1, Denis Mitchell2,

John A. Bickley3, R. Doug Hooton4, William D. Cook2
1
Atomic Energy of Canada Ltd, e-mail: [email protected]
2
Department of Civil Engineering and Applied Mechanics
McGill University, Montreal, Canada
3
John A. Bickley and Associates, Toronto, Canada
4
Department of Civil Engineering, University of Toronto, Canada

Introduction
The ACR-1000* reactor (Advanced CANDU Reactor*) is designed for a 100-year
plant life including a 60-year operating life and an additional 40-year
decommissioning period. The ACR-1000 containment structure consists of a
vertical cylindrical perimeter wall, founded on a base slab and hemispherical
dome at the top. The perimeter wall and the dome are prestressed concrete with
post-tensioning tendons in the horizontal and vertical directions as well as
reinforcing bars in both directions. The inside surfaces of the containment
structure are lined with a carbon steel liner. The base slab is a reinforced
concrete structure.
It is evident that the service life performance relies not only on the Ageing
Management Program (AMP), but is also strongly influenced by the design
strategy and material characteristics [1]. It is believed that improved performance
during the design life of a structure can be achieved by implementing durability
design criteria and improving material characteristics. Therefore, for a new
plant, the Plant Life Management Program (PLiM) starts at the design process
stage and continues through the plant operations and decommissioning stages.
Modern building codes are increasingly based on performance specifications
for durability (Performance Based Design) [2]. In the development of the ACR-
1000, particular attention is paid to specifying structural and long-term durability
performance as part of the technical requirements. Many recent innovations in
advanced concrete materials technology have made it possible to produce
modern concrete with exceptional performance characteristics. The concrete

*
ACR-1000® (Advanced CANDU Reactor®) is a registered trademark of Atomic
Energy of Canada Limited (AECL)

237
6. Design and Construction Issues

performance strongly relies on four key factors a) material ingredients, b) mix

design, c) concrete production and (d) curing.

Aim of the work

AECL has been active with an R&D in order to implement this new approach,
including studies on mix design, analysis techniques, and construction procedures
to reduce the risk of cracking at early ages. The long-term performance of the
containment structure is strongly influenced by crack control, particularly at
early ages. Successful implementation of the program recommendations will
enable high performance and time-effective construction of the reactor building
base slab, containment wall and dome. Implementation of a continuous-cast
concrete base slab has already been addressed in an earlier phase of the research
program [3].

Essential results
This paper provides the results of the research and development program for the
containment wall and dome. Laboratory and field-trial tests, together with
thermal and stress analyses for the probable range of concrete mixes have been
carried out in order to predict the mechanical properties, temperature variations,
thermal stresses and risk of cracking for a thick pour concrete containment wall
and dome. Performance criteria for a 100-year service life and the type of
specification required to meet the performance criteria have been established.
Laboratory and field-trial tests, together with thermal and stress analyses for
the probable range of concrete mixes have been carried out in order to predict
the mechanical properties, temperature variations, thermal stresses and risk of
cracking for jump-form construction of the wall and dome. Mechanical and
thermal properties, obtained from the field trials were employed in the finite
element thermal-stress analysis program in order to predict the time-dependent
risk of cracking at early ages of the wall and dome. Figure 1 shows the
temperatures measured near the centers of the 1 × 1 × 1 m insulated cubes that
were cast as part of the field trials.

238
 6. Design and Construction Issues

75

Mix 1 - Reference Concrete

65

55
C
o

Mix 2 - 50% Slag Concrete

Temperature,

45

35
Mix 3 - 50% Fly Ash Concrete

25

15
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0
Time, days

Figure 1. Measured temperatures near center of insulated cubes for different concrete mixes.

Summary
The development program provides the necessary recommendations and guidance
on mix design ranges, analysis tools and construction techniques to build the
containment structure including the wall and dome of the ACR-1000 reactor
building.

References
1. Abrishami, H.H., Ricciuti, R., Elgohary, M. Plant Life Management of the ACR-1000
Concrete Containment Structure. CSNI Workshop on Ageing Management of
Thick-Walled Concrete Structures, Including In-service Inspections, Maintenance
and Repair, Instrumentation Methods and safety assessment in View of Long-
Term operation., NEA/SEN/SIN/IAGE(2008)7, Nuclear Energy Agency, OECD,
Prague, Czech Republic, October 1–3, 2008.

2. Bickley, J.A., Hooton, R.D., Hover, K.C. Performance Specifications for Durable Concrete.
Concrete International, September 2006, pp. 51–57.

3. Abrishami, H.H., Elgohary, M., Mitchell, D., Bickley, J.A., Hooton, R.D., Cook, W.D.
Implementation of Continuous-Cast Concrete Base Slab for Future CANDU
NPP. SMiRT 19, International Conference on Structural Mechanics in Reactor
Technology, August 12–17, 2007, Toronto, Canada.

239
6. Design and Construction Issues

Structural analysis, design and detailing of

reactor vault in prototype fast breeder
reactor (6-1972)

K.V. Subramanian, S.M. Palekar, H.A. Mapari, R. Balaji

TCE Consulting Engineers Ltd
243 Matulya Center A, Senapathi Bapat Marg, Mumbai, India
e-mail: [email protected]

Introduction
Prototype Fast Breeder Reactor (PFBR) is the first Reactor of its kind presently
under construction at Kalpakkam in INDIA. This paper deals with different
aspects related to structural layout, analysis, design and detailing of Reactor
Vault (RV), located in Reactor Containment Building (RCB) of PFBR, bringing out
certain typical considerations made, analysis procedures adopted and constructability
aspects.

General layout of reactor vault

The Reactor Vault housed in the Reactor Containment Building (RCB) form a
major structural system. This Building forms one of the buildings constituting
the Nuclear Island Connected Building (NICB), an integrated structure
constituting of eight buildings, viz Reactor Containment Building (RCB), Steam
Generator Buildings (SGB1 & SGB2), Fuel Building (FB), Rad Waste Building
(RWB), Electrical Buildings (EB1 & EB2) and Control Building (CB) integrated
to form a single integrated structural system. The Reactor Vault comprises of an
inner wall and an outer wall (Circular concentric RC walls), and the Reactor
Assembly (comprising of Reactor Vessel and the core) hanging from the outer
wall. Inner wall is lined with metallic liner, and Biological Shield Cooling
System (BSCS) is provided behind the liner for cooling purpose, (Figure 1). A
Safety Vessel (SV) is provided outside the Main Vessel (MV) to act as a barrier
against sodium leakage from the MV. A roof slab is provided for the Reactor
Assembly, over which, different equipments / operating platforms get supported.

240
 6. Design and Construction Issues

Figure 1. Schematic of Reactor Vault along with liner arrangement.

Structural analysis and design

Structural analysis of the Vaults involve performing analysis under normal
conditions (where sodium is contained in the MV) and under an assumed condition
of leakage in the Main Vessel (MV) and sodium occupies the inter-space between
MV and SV (during which SV will be loaded with leaked sodium). Under such a
condition, the structure can be subjected to seismic events (OBE and SSE).
Condition of overflow of leaked sodium from SV into the inter-space between inner
wall of Reactor Vault and SV is also considered in the analysis.
For performing analysis of the structure under normal conditions, when
Sodium is contained in the MV, Reactor Vault, MV and SV have been coupled
along with the Global FE model of NICB, and coupled model has been used for
performing the analysis.
For performing analysis under condition of MV leakage, a separate isolated
model of Reactor Vault along with the RA (MV with its internals) and SV, as
extracted from the Global FE model, has been considered. Analysis is performed
in-order to determine the additional forces that would be transferred to the civil
structure when a seismic event occurs under the condition of sodium leakage. The
leaked sodium mass is lumped in the walls of SV while performing this analysis,
to represent the effect of extra force that gets transferred to the SV support.
Design is performed as per codal provisions of AERB, India for strength and
crack-width based requirements.
The MV support zone in outer wall of RV comprises of a solid bracket of size
2.3 m (vertical C/S Dim) × 2.3 m (horizontal cross-sectional dimension) running
along the circular periphery of MV to support it’s flange. Reinforced Concrete
(RC) in this area is subjected to highly varying stresses because of the geometry
of the bracket and heavy and concentrated loads transferred from MV under the
action of different loads. Separate FE analysis is performed for the projecting
portion of Outer wall where the Flange of MV is supported, by means of refined

241
6. Design and Construction Issues

FE sub-model developed using 3-D solid elements. Moments and forces across
different cross-sections of the model are determined by summing the element
forces about the center of section considered, and design is performed for the
summed forces and moments, using provisions of ACI 349 for Corbel design.
Thermal Analysis of the vaults have been performed for Heat-of-Hydration
load using a Construction-stage Time-History Analysis and Standard Heat
Transfer equations and the reinforcement required is verified to be less than the
provided reinforcement.
The base supporting arrangement for inner wall of RV (IV) comprises of a
central monolithic support transferring the entire seismic shear forces while the
peripheral supports comprising of mirror-finished bearing pads, free to slide over
each other, so as to allow for thermal expansion of the IV. The supporting
system for the IV has been separately designed for the different forces
transferred at the base of IV.
Structural analysis of liner along with the attached BSCS is performed for
imposed deformation loads under different conditions, and different components
of the liner system and BSCS (including liner anchorage) have been qualified for
safety and stability.

Detailing aspects considered

Detailing had been done in a planned manner to avoid any interference between
any of the EP / Liner anchors, reinforcement bars and embedded cooling pipes
and other services, and congestion has been reduced to the extent possible.
Typical reinforcement detailing provided in critical areas like projecting bracket
zone of outer wall of RV, have been presented in the paper.
A full separation is provided between inner and outer walls of the vault, so
that they behave independently. Expanded Polystyrene is provided in the gap
between the two walls. At the base of inner wall of RV, provisions have been
made such that the Inner wall is free to undergo radial thermal expansion,
achieved by means of Manganese Phosphate coated bearing pads.
Mock-up tests have been conducted for different critical portions of the
Reactor Vault assemblage, to study the possible construction problems that
could be envisaged.

Conclusion
This paper presents different issues and aspects pertaining to analysis, design
and detailing of Reactor Vault for PFBR. Critical loading environments
pertaining to Reactor Vault and Analysis methodologies adopted have been
discussed. Detailing issues have been addressed, and methods adopted have been
discussed in the paper, with detailed sketches showing different critical areas
where typical detailing has been provided.

242
 6. Design and Construction Issues

Development and in-reactor verification of

three types of advanced nuclear fuels for
PWRs (6-1986)

Young Ki Jang1, Kyeong Lak Jeon1, Yong Hwan Kim1, Jae Ik Kim1, Jung Cheol
Shin1,
Man Su Kim , Tae Hyoung Lee2 and Jong Ryul Park2
2
1
Nuclear Fuel Technology Department, Korea Nuclear Fuel, Yuseong, Daejeon,
305-353, Korea, e-mail: [email protected]
2
Korea Hydro & Nuclear Power Co., Ltd.

Keywords: Advanced, Nuclear, Fuel, In-reactor, Performance, Verification.

Three types of advanced fuels for PWRs are being verified in three Korean
nuclear reactors: PLUS7TM for Optimized Power Reactor of 1000 MW class
(OPR1000) and Advanced Power Reactor of 1400 MW class (APR1400), and
16ACE7TM and 17ACE7TM for 16x16 and 17x17 Westinghouse types of plants,
respectively.
Each four lead test assemblies (LTAs) for each fuel type had been loaded to
verify the irradiation performances in the commercial reactors. Four steps for in-
reactor verification were being applied: the first one for assembly-wise
examination in poolside after each cycle, the second one for rod-wise
examination after disassembling in poolside, the third one for the rod
examination in the hot cell test facility in detail, and the final one for the
skeleton examination in the hot cell test facility in detail.
The first leading fuel, PLUS7TM, has completed 3 steps of verification and will
start the final step from this year, while the 16ACE7TM has completed 2 steps
and is waiting for the third and fourth steps until next year and the 17ACE7TM is
being verified in reactor for the third cycle irradiation. The examination results
up to now showed that all these 3 types of fuels were being irradiated
successfully in the reactors.
The designs of these three types of advanced fuels are summarized and in-
reactor performances on three types of advanced fuels are compared in this
paper. In conclusion, all the irradiation performance parameters were within the
expected design limits. In-reactor performances of these fuels are being verified
continuously through the surveillance program during commercial
implementation.

243
6. Design and Construction Issues

Investigation of possible corrective actions

during manufacturing of fast breeder
reactor components towards assessing
the structural integrity (6-2003)

S. Jalaldeen, P. Chellapandi, S.C. Chetal

Nuclear Engineering Group
Indira Gandhi Centre for Atomic Research, Kalpakkam, India
e-mail: [email protected]

The specifications of tolerances for the manufacturing of components are very

important. While tight tolerances pose challenge to even big industries, liberal
tolerance may have impact on design and integrity issues. Further, the actually
achieved dimensions after manufacturing may not meet the initial specified
value. One need to investigate the effect of deviation including functional and
integrity requirements preferably at design stage to avoid problems of over /
under specifying the tolerances or need for corrective action / rejection at a later
stage.
In case of pool type fast breeder reactors, the ovality of large diameter shells
(main vessel / safety vessel / inner vessel / thermal baffle) are important from weld
mismatch considerations at manufacturing time apart from other integrity
requirements including buckling during operations. The verticality of penetration
shells in top shield is important from functional requirements. Another possible
necessity of corrective action is mismatch in circumference of two matching
parts such as a nozzle and a large diameter pipe to be welded with the nozzle. In
case the pipe is manufactured in two halves with a final longitudinal weld, there
is a chance that the circumference of the pipe is more or less than that of the
nozzle to which the pipe is to be welded. All these requirements may call for
corrective action or repair or rejection at manufacturing stage if adequate care is
not taken.
A few possible deviations envisaged are discussed along with corrective
actions that can be taken and the effect of corrective action is investigated. In
case of ovality problem, one need to apply uniform force all around which
means straining the component at manufacturing stage. The resulting additional
stresses and their effect on integrity during operation are investigated. In case of
circumference mismatch, (say pipe is smaller than the nozzle), one can attempt
corrective action by mechanically expanding the pipe near the free end (weld
end) using a mandrill. This means a shear force and bending moment are applied
at the junction resulting in additional stresses than envisaged for operating load.

244
 6. Design and Construction Issues

The resulting creep /fatigue damage due to the additional stresses are determined
for typical examples / components taken as case study.
All these investigations form a basis for deciding the course of action during
manufacturing of major components of 500 MWe Prototype Fast Breeder
Reactor, which is under construction at Kalpakkam. The paper discusses the
details of various possible examples, investigations carried out including finite
element analysis and also brings out the actual manufacturing experiences in few
cases.

245
6. Design and Construction Issues

Structural analysis towards erection of

prototype fast breeder reactor
components (6-2005)

Bhuwan Chandra Sati, V. Balasubramnian, P. Chellapandi, S.C. Chetal

Nuclear Engineering Group
Indira Gandhi Centre for Atomic Research, Kalpakkam, India
e-mail: [email protected]

PFBR is a sodium cooled pool type reactor. The normal operating temperature is
670 K (397°C) for its cold pool components and 820 K (547°C) for the hot pool
components. The hot pool and cold pool are separated by an inner vessel. The
entire primary sodium pool is contained in a 25 mm thick main vessel of 12.9 m
diameter and 13 m height. The main vessel is surrounded by a safety vessel
which is still larger than the main vessel. The reactor has a design life of 40
years. Thus fabrication and erection of the large components are key activities
for the construction of PFBR. The life of reactor is dependent on the structural
integrity of these large components in both normal and accidental conditions. In
view of this the fabrication and erection activities connected with PFBR have to
be performed with high quality and safety requirements.
In PFBR higher safety restrictions are called for in view of the radiation hazards
involved during operation as repairs and rectifications for the primary sodium circuit
components is not possible. The large thin components are manufactured with very
tight tolerances. Hence good care is essential to avoid errors and damage to the parts
during erection. Also the functionalities and specification of the PFBR components
are different from conventional industrial components, specialized techniques for
erection are required. In this paper the erection sequence and methodologies adopted
for the various reactor assembly components such as main vessel, safety vessel,
inner vessel, thermal baffles, grid plate etc are brought out.
In PFBR the safety vessel and main vessel are to be placed in a pit type reactor
vault, having very narrow gap between the vessel and the vault. The erection of
components and the construction of reactor building have to be carried out
simultaneously. Since the components are thin shell structure and cannot be repaired
once erected special attachments and handling fixture are designed for the erection.
The components are analysed for the loads occurring on them during handling process.
In this process the safety vessel (diameter 13.5 m and 16 m height) has already
been erected successfully on the inner reactor vault within a gap of 70 mm.
Further erection of components is planned, analysed for erection loadings and
mockups are performed where ever needed. The total sequence of erection for
the various critical components, options studied, analyses carried out etc. will be
discussed in depth in the paper.

246
 6. Design and Construction Issues

SSI analysis for a reactor building with high

frequency seismic ground motion (6-2040)

Richard Chen, Tarek Aziz, Sudip Adhikari, George Stoyanov, Wei Liu
Civil Engineering, Atomic Energy of Canada Limited, Ontario, Canada
e-mail: [email protected]

Introduction
Current seismic hazard assessments express hazard in terms of Uniform Hazard
Spectrum (UHS). High frequency content is present in the UHS for Nuclear
Power Plants (NPP) in Central Eastern North America. It was found that this
high frequency content in the UHS has significant effects on the seismic
response of a structure when using conventional analysis methodologies. In
several analytical cases, the high frequency content contributes to an increase in
the floor response spectra (FRS), especially for those elevations close to the
ground. In reality, however, it is well known that high frequency content of
ground motion has much lesser damage effect to Structures, Systems, and
Components (SSCs) of a NPP than low frequency content (except functional
performance of some vibration sensitive components, such as relays). The
challenge is how to reflect this reality in seismic analysis of a NPP.
In this paper, a literature review of existing techniques to mitigate the effects
from the high frequency contents of the ground motion is presented. Two of
these techniques, namely, seismic wave incoherence effects and multiple sets of
artificial acceleration time histories (instead of the conventional one set of broad
band envelop time histories), are selected to perform Soil-Structure Interaction
(SSI) analysis for a typical reactor building. A simplified stick model (see Figure
1) is developed and is used in the SSI analysis. A free field UHS with significant
high frequency content is used as the seismic ground motion input. The seismic
wave incoherence effects on foundation and building response are considered
using ACS SASSI, a computer software for SSI analysis. Multiple sets of time
histories are derived from the UHS corresponding to different frequency
contents. The effects on both hard rock condition and soft soil condition are
compared. Floor response spectra at both the base slab of the reactor building
and top of the internal structure are generated. The results from ACS SASSI and
other seismic analysis computer codes (e.g., STARDYNE) are also compared.

247
6. Design and Construction Issues

Conclusions
In this paper, Soil-Structure Interaction analysis of a typical reactor building is
presented using a UHS with high frequency content. Floor response spectra are
generated at both base slab and top of the internal structure. The results are
obtained using ACS SASSI and other seismic analysis computer codes (e.g.,
STARDYNE). It is seen that seismic wave incoherency has major effects on
high frequency seismic response of the reactor building founded on hard rock as
expected. Seismic wave incoherency has lesser effect on high frequency seismic
response when the building is founded on soft soil condition. The application of
multiple sets of time histories instead of the conventional spectrum compatible
set can affect the seismic response of the reactor building for both soil and rock
conditions. However, this concept of using multiple sets of time histories has to
be rationalized.

Figure 1. A simplified stick model of a typical Reactor Building.

248
 6. Design and Construction Issues

Evaluation of local stresses at the vessel

shell to nozzle intersection (6-2117)

Andrzej Strzelczyk, San Ho

Ontario Power Generation, 889 Brock Rd., Pickering, Ontario, Canada
e-mail: [email protected]

Local stresses at the intersection of the pressure vessel shell with a nozzle can be
evaluated by Bijlaard or finite element methods. The first method, which is
based on the classical shell theory, is still very common and described in many
references like [1] and [2]. The second approach can be implemented by using
shell or solid finite element model. This paper compares both approaches and
discusses the limitations of the Bijlaard and shell finite element methods.
The paper demonstrates that for the purpose of ASME Code evaluation, a
more practical approach is an automatic evaluation of the stresses from a solid
element model in which one second-order element through-thickness is used.
The advantages of this type of modeling have been discussed in [3]. The
geometry of the model can be generated by a computer program whose input
data are vessel and nozzle diameters and thicknesses. An example of a source
code of such a program, generating an input deck for Abaqus finite element
code, is provided.
The proposed approach not only allows for fast evaluation of the stresses but
also visualization of stresses and deformation as well as investigation of various
factors affecting the solution. The output produced by the Abaqus program can
easily be post-processed automatically for the need of ASME Code evaluation.

References
1. Wichman, K.R., Hopper, A.G., Mershon, J.L. 1965. Local Stresses in Spherical Shells
due to External Loading. Welding Research Council Bulletin 107, October 20
Update of the March 1979 Revision (original, August 1965), pp. 68–70.

2. Forman, B.F. 1979. Local Stresses in Pressure Vessels. Second Edition, Pressure
Vessel Handbook Pub.

3. Strzelczyk, A., Ho, S. 2007. Evaluation of “Linearized” stresses without Linearization.

Proceedings of PVP2007-ICPVT-11 (2007 PVP Conference in San Antonio).

249
6. Design and Construction Issues

A study on optimization of seismic

strengthening for the plant facilities in
terms of plant management (6-2225)

Masami Oshima1, Takashi Kase1, Kazuyoshi Sekine2

1
Chiyoda Advanced Solutions Corporation, Yokohama, Japan
e-mail: [email protected]
2
Yokohama National University, Yokohama, Japan

Introduction
At present, seismic design of chemical plant facilities mainly complies with “the
Seismic Design Code of High Pressure Gas Facilities in Japan. In this notice the
main purpose of seismic design is to ensure public safety at seismic events. In
case of execution of seismic assessment and seismic strengthening, the code is
applied to the existing plant facilities. If the existing plant is revamped, strict
observance of the code is required. Although another design method should be
proposed to execute the seismic strengthening from the view point of “Social
responsibility of product supply” and “Maintenance of facilities and restraining
of damage cost”, which are ones of the most important seismic risks of plant
management.
In this study, from the view point of “Social responsibility of product supply”
and “Maintenance of facilities and restraining of damage cost”, estimated term
of shutdown and necessity of repair are focused on. And authors study
evaluation method of seismic strengthening cost using damage levels which take
them as the criteria for judgment. Hence damage levels after seismic events
focused on the estimated term of shutdown and necessity of repair are
established, and seismic strengthening cost at each damage level is calculated.
Authors have already proposed the method1) which can provide the
information concerning improvement ratio of seismic performance, which is
focused on the estimated term of shutdown and necessity of repair, at the
selected damage level in case of determining the seismic strengthening cost. In
this paper, the procedure to obtain the relationship between seismic
strengthening cost and critical seismic coefficient on the ground at each damage
level is proposed, and a calculation result of an example of a tower is presented.
Furthermore this approach is effectively applicable for existing nuclear power
plants and their related facilities instead of chemical plant facilities.

250
 6. Design and Construction Issues

Aim of the work

The aim of this evaluation method is the following; if investment amount for
seismic strengthening of the equipment and damage level at a seismic event are
supposed, the portions required seismic strengthening and degree of
strengthening can be determined. Accordingly improvement ratio of seismic
performance of the subject equipment at the selected damage level could be
assumed.
Therefore the damage levels focused on the estimated term of shutdown and
necessity of repair are supposed, and the fault tree, whose fundamental events
consist of damage modes, will be prepared for each damage level. Besides by
means of approximation of the cost required for seismic strengthening and
calculation of the seismic strengthening cost using fault tree, extraction of the
portions to be strengthened, its required degree of the improvement, and its
seismic improvement ratios can be assumed. Consequently the relationship
between critical seismic coefficients on the ground and the cost of seismic
strengthening is presented, and this relationship is applied for evaluation of
seismic countermeasures. And here a self-supported tower on skirts is selected
as the typical example of equipment in the plant facilities.

Essential results
The relationship between design seismic coefficients and seismic strengthening
costs at each damage level is calculated. Examples of applications are presented
as follows:
1) The seismic strengthening costs could be approximated by determining the
expected damage level of the subject equipment and the design seismic
coefficient.
2) In case of a given seismic strengthening cost, the improvement ratio of the
seismic performance could be decided by selecting the damage level.
3) If the seismic strengthening cost and damage level are decided, the critical
seismic coefficient on the ground of the equipment at the expected
damage level could be assumed.

Summary/conclusions
In this paper, the estimated term of shutdown and necessity of repair is focused
on and damage levels are proposed considering them, and an evaluation method
of the seismic strengthening cost based on the damage levels is developed.
Using this evaluation method, in case of a given amount of seismic
strengthening cost, the information of improvement ratios of seismic performance

251
6. Design and Construction Issues

focused on the estimated term of shutdown and necessity of repair can be

provided, if the required damage level in terms of maintenance for the
equipment is selected.
Additionally in this method extraction of dominant damage modes and
appropriate distribution of seismic strengthening cost for plural damage modes,
which are derived from analysis of fault trees, are possible. For an example, it is
verified to be possible to extract dominant damage modes require seismic
strengthening, in case of selecting damage levels and seismic coefficients on the
ground.

Reference
1. Oshima, M., Kase, T. et al. Strategy for Seismic Upgrading of Chemical Plant Taking
Productivity as Criterion of Judgment. Proceedings of SMiRT-18, Paper No.
M01_6, August 2005.

252
 6. Design and Construction Issues

Comparing European and American

codification in the field of NPP
civil engineering (6-2493)

Philippe Bécue, Danièle Chauvel

EDF SEPTEN Civil Engineers

Context
The opportunity to build an EPR in different countries during the next decades
has inclined EDF Engineering Department presently involved in the construction
of Flamanville 3 NPP to compare American nuclear codes in Civil Engineering
to the French practice recently brought up to date for this project.

The needs for a new technical code for EPR

In France, the needs for a new design code for the Civil works of the EPR has
come from the necessity to take into consideration the European Regulation
(eurocodes) and the specificity of this project in term of safety improvement
with respect to Severe Accident. This code entitled ETC-C (EPR Technical
Code for Civil works) has been written so far by EDF.

Design of the containment steel liner

The design of a steel liner covering the inner surface of the containment has popped
up the necessity to consider the blistering as a conséquence of the strongly pre-
stressed concrete wall combined with a high temperature inside due to a hypothetical
Severe Accident. This led the ETC-C to complete the mechanical analysis and
reformulate the acceptance criteria used in the former codes.

Concrete structures design

For the concrete structures a comparison is made between the ETC-C and the
American rules ACI 349 and ASME (section III Division 2).
An application of both reglementations to the determination of the reinforcement
ratio on EPR concrete structures such as walls and decks of the reactor building
has shown significant differences due to the codes applied as well as to the
material properties, the loads and the engineering practice.

253
6. Design and Construction Issues

Conclusion
It is obvious that a new period of nuclear plant construction has created an
opportunity in France to upgrade the nuclear codification taking into account the
recent European reglementation in the fields of concrete and metallic structures.
This article gives good examples of the differences between American and
French standards applied to EPR.

254
 6. Design and Construction Issues

Aspects of the design and construction of a

new feedwater line for Angra 1 Nuclear
Power Plant as a part of the steam
generator replacement program (6-2497)

Milton Kronenberg Francioni1, José Eduardo de Almeida Maneschy1,

Carlos Alberto de Oliveira2
1
Eletronuclear – Eletrobrás Termonuclear S.A.
Rua da Candelária 65, 20091-906, Rio de Janeiro, RJ, Brazil
e-mails: [email protected], [email protected]
2
Nuclear and Energy Research Institute, IPEN–CNEN/SP
Av. Prof. Lineu Prestes 2242, Cidade Universitária, 05508-000, São Paulo, SP, Brazil
e-mail: [email protected]

Angra 1 Nuclear Power Plant Steam Generator replacement plays an important

role in the intents of plant owner, Eletronuclear, for plant life extension and
power uprating. As the feedwater nozzle position is now different from the
original design, it brought the opportunity of modifying piping layout and
support configuration of the main feedwater line inside containment. This paper
presents the work performed and the improvements reached, when compared
with the original design.
The main feedwater line piping has been re-routed to mach up with the
feedwater nozzles attached to the new steam generators. The new routing of the
feedwater lines has made it necessary to re-analyze the piping to demonstrate
qualification to design code requirements. Stresses computed are within the
allowable limits. New piping layout was also designed to avoid thermal
stratification possibility in nozzle and line, and the new material of the piping,
A335 P11, was chosen because of its corrosion resistant properties.
Since a new approach to postulated pipe break, based in present regulatory
requirements, is now used, the concept of Arbitrary Intermediate Break,
considered in the original design, was eliminated. According to the stress
analysis, no intermediate break had to be considered in the new arrangement and
thus pipe whip restraints are now used only at terminal points. Old pipe whip
restraints are unnecessary and may be dismantled for increasing working area in
that plant region inside containment.
The piping design was qualified according to the design code requirements
(ASME code, section III, division 1, subsection NC, class 2 components) and to
the Brazilian regulatory requirements.

255
6. Design and Construction Issues

Analysis of the stress-strain state of

containment depending on temperature
fluctuations in the environment (6-2560)

Alexander S. Kiselev, Alexey S. Kiselev, Victor N. Medvedev,

Valery F. Strizhov, Аlexey N. Ulianov
Nuclear Safety Institute of the Russian Academy of Sciences (IBRAE RAS)
115191, Bolshaya Tulskaya str., 52, Moscow, Russia, e-mail: [email protected]

Introduction
At the present time the design life of containments at several Russian Nuclear
Power Plants (NPPs) is nearing completion that requires both analysis of the
capability to prolong their operation and development of measures to extend
service life of their structures. New NPP units under construction also face these
problems due to the need for both reducing damageability of reinforced-concrete
structures during operation and extending their service life.
The running experience of NPP containments collected over a protracted period
of time has demonstrated that environmental effects (mostly seasonal and daily
temperature fluctuations) are the key factors influencing their stress-strain state.
Taking account of the fact that environmental-temperature fluctuations may
have rather large increments over short periods of time, the temperature-stress
values could be also significant and would lead to degradation of strength
properties of containment materials and, as a consequence, to loss in their
operational functionality.
The aim of presented work is the calculation analysis of changes in the
strain-stress state of NPP containments with VVER-1000 depending on seasonal
and daily environmental-temperature fluctuations.

Results
Two specific cases are considered: 1 – real condition of containment without
thermal insulation at the outer wall surface; and 2 – assumed coating of the outer
containment surface with a foam-concrete layer (heat insulation). Two finite
element models of the containment were developed corresponding to considered
cases. The second model has two additional finite-element layers at the outer
surface of both the cylindrical and the domical parts simulating heat insulation.
The model consisted of 567594 finite elements and 604494 nodes.
The convective heat-exchange boundary conditions were assigned at the inner
and the outer surfaces of the containment model.

256
 6. Design and Construction Issues

In the temperature-field calculations with consideration for both seasonal and

daily temperature fluctuations for normal operating conditions was taken the
constant temperature inside the containment. The temperature outside the
containment was assigned as a piecewise-linear dependence in accordance with
the results of in-situ measurements performed every three hours in the course of
the time frame under consideration (30–50 days).
According to the results of air temperature measurements close to the Kalinin
NPP site maximum mean daily temperatures during summer months equal about
30оC, whereas minimum temperatures in winter months go down to minus 30оC. In
spring, substantial daily air-temperature fluctuations (up to 20оC and more) are typical
for the Kalinin NPP area, and in the nighttime air temperatures can even drop down
to -5оC. Such large daily temperature fluctuations may lead to strength degradation
in concrete due to destructive processes caused by freezing-thawing strains.
The calculations were conducted in the following order. At first, temperature
distributions across section of the containment wall depending on fluctuations of
both seasonal and daily temperatures were calculated. Two specific cases were
considered: 1 – without heat-insulating layer; and 2 – with heat insulation of 10-
cm thick (foamed concrete of γ=600 kg/m3 density) at the outer surface of both
the containment cylinder and dome.
Next, calculations of the containment stress-strain state caused by the thermal
load were conducted followed by those due to the integral effect of operating
loads (dead weight, prestress, operating temperature). The stress-strain state was
analyzed for both studied cases (i.e. without heat-insulating layer and with heat
insulation at the outer surface of the containment cylinder and dome).
The fulfilled researches have showed that the following beneficial effects
might be expected as the result of coating the outer surfaces of containments
with heat insulation:
− diminution of negative environmental impacts (temperature fluctuations,
humidity effects, frequent changes of freezing by thawing, impacts of
deleterious substances etc.) on reinforced concrete would allow extending
service life of NPP containments;
− decrease in variations of geometric containment parameters within the
crane track area due to diminution of the temperature gradient inside and
outside the containment;
− decrease in financial and manpower resources for NPP maintenance and
repair thanks to reduction of the scope of crane-track-adjustment works;
− decrease in creep strain of concrete and consequently a diminution of
losses in tendon efforts from creep strains of concrete; and
− diminution of compressive stresses in concrete at the inner containment
surface and within the metal liner which develop under service-load
impacts (taking account of creep strain of concrete) and in case of a
maximal design-basis accident.

257
6. Design and Construction Issues

Using the stressed frame for blast resistant

fenestration design of full containment
structures (6-2601)

Adeola Adediran1, Henry Ayvazyan2

1
Engineering Supervisor with Bechtel National, Inc.
e-mail: [email protected]
2
Principal Blast Engineer with Parsons

Designing for internal accidental explosions in safety related structures (Nuclear

and Demilitarization structures) are a challenge when one considers that safety
related structures have to be designed for full containment. This means that for
the maximum credible event, the structure not only survives the event but must
survive in such a way that it continues to prevent the leakage of agent or release
of radiation. So if the walls of the full containment structures need fenestrations
for say utility openings, doors and HVAC, in a typical design, this usually
necessitates substantial rebar construction around such penetrations. In some
cases, the walls are designed with embedded pilasters integral to the wall and
placed around these openings. The resulting design is highly prone to congestion
of rebar around the openings and subsequent honeycombing of the concrete
placed. One approach to prevent this inevitable rebar congestion is the use of the
stressed frames. This approach has been more frequently used in the
demilitarization projects but its use is being broadened for other safety related
structures. Several uses of the stressed frame currently exist. The discussion in
this paper is general in nature and does not reflect any one design in current use.
This paper discusses what stressed frames are, how they work, how to design
them and estimates cost savings and more importantly schedule savings expected.

Methodology
In this paper a representative wall, 25 inches thick is designed with the typical
substantial rebar construction around openings. Previous problems of
congestions and constructability are investigated and discussed. Then that wall is
designed using stressed frames similar to that shown in Figure 1. The total
containment criteria for the wall chosen is a deformation of not more than 1
degree support rotations when subjected to a detonation from a net equivalent
TNT weight of 18 lbs. Since the hazards in the room are to be contained in the
room the openings are covered openings. So the frames around the openings
may have a gate, shadow shield, penetration plate or valve attached to them. The
stressed frame is designed not just for the shears that result from the covering
within the frame but are designed for the moments and membrane tension that

258
 6. Design and Construction Issues

occur around these openings when the wall deforms. The uniqueness of the
stressed frame, therefore, is that ability to take the stresses around the opening
with the frame itself and not in the concrete. Since the stress concentrations are
not in the concrete the need for additional reinforcement and the potential for
rebar congestion are avoided. Figure 2 shows the stressed frame installed with
the wall reinforcing.

Figure 1. Stressed Frame around Utility Opening.

Figure 2. Stressed Frame with Wall Reinforcement.

259
6. Design and Construction Issues

A new device for the study of early-age

cracking in massive concrete
structures (6-3140)

Briffaut Matthieu1, 2, Benboudjema Farid1,

Torrenti Jean-Michel1, 3, Nahas Georges2
1
ENS Cachan/CNRS UMR8535/UPMC/PRES UniverSud Paris, Cachan, France
2
Institut de radioprotection et de sûreté nucléaire, Fontenay-aux-Roses, France
3
LCPC / EFB, Paris, France
e-mail: [email protected]

At early-age in massive concrete structures, cracking may occur during

hardening. Indeed, hydration is an exothermic chemical reaction (temperature in
concrete may overcome 60°C). Therefore, if autogenous and thermal strains are
restrained, compressive stresses and then traction stresses rise, which can exceed
the concrete strength (in an elastic finite element calculation). This cracking may
increase significantly the concrete wall permeability.
The restrained shrinkage ring is a good method to determine the concrete
behaviour (strain and cracking) due to the autogenous and drying shrinkage. In
this study, a concrete mix which is representative of a nuclear power plant
containment is tested. This test showed that at 20°C and without drying (cf.
Figure 1), the amplitude of autogenous shrinkage is not high enough to cause
cracking. Indeed, in this configuration, thermal shrinkage does not occur.
Therefore, a new test has been developed to study cracking due to restrained
thermal shrinkage.
This new test is an evolution of the restrained shrinkage ring test which allows
taking into account the autogenous shrinkage and also the thermal shrinkage.
Actually under development, it aims is to predict the behaviour, the cracking at
early age of massive structures (especially the nuclear power plant containment).
The new test principle is to increase the temperature of the steel ring in order to
expand it. In this case, expansion of ring is restrained by the extern concrete
layer, which induced compressive stresses in the ring and therefore tensile
stresses in concrete.
The history of the ring temperature could be calculated by finite element in
order to reproduce similar stress rate that can occur in a “real’ massive wall
(calculated also by finite element simulations, cf. Figure 2).
In order to verify that cracking will really occur, we have modelled the new
test with a finite element code ( Cast3M, developed by the French atomic energy
commission). We use a model developed by Benboudjema and Torreti (2008).
This one is a thermo-chemo-elastic-damage model.

260
 6. Design and Construction Issues

Figure 1. Geometry of the restrain Figure 2. History of temperature measured

shrinkage ring test. in a concrete wall (1.2 m width).

The prediction of cracking needs a numerical resolution due to the complexity of

the behaviour of concrete at early-age. In our model, the following phenomena
are taken into account:
- The evolution of hydration: this is achieved here by the use of a chemical
⎛ E ⎞
affinity (Ulm and Coussy, 1998): ξ = A(ξ ) exp⎜ − a ⎟
~
⎝ RT ⎠
- The evolution of temperature: the energy balance equation, which includes
the release of heat due to the hydration reaction is solved:
CT = ∇(k∇T ) + Lξ
- The evolutions of autogenous and thermal strains; ε au = −κξ1 for ξ > ξ 0
and ε th = −αT 1

- The evolution of Young modulus and tensile strength with respect to the
hydration degree by the use of De Schutter laws (1997):
a
⎛ ξ − ξ0 ⎞
B(ξ ) = B∞ ⎜⎜ ⎟⎟ where B is the mechanical variable
⎝ ξ∞ − ξ0 ⎠
- The description of cracking in tension: elastic damage model (Mazars, 1986)
slightly modified: σ = (1 − D )σ ~ = E(ξ )ε = E(ξ )(ε − ε − ε )
~ and σ
el au th

261
6. Design and Construction Issues

To take into account these phenomena, several parameters have to be identified.

Therefore, several tests were performed and some results from the literature
were used. The results of the mechanical tests are shown in Figure 3 and table 1
(the tensile strength is determined by a Brazilian test).

Table 1. De Schutter laws parameters.

Estat Rc Rt
ξ0 0,115
a 0.449 1.105 0.806

With this modeling we could know approximatively when the concrete ring will
is able to crack so we could be able to improve the temperature history by
decreasing the temperature rise rate and obtain a better permeability evolution.

Mechanical strength evolution

45 5

40 Compressive strength (MPa) 4

Compressive strength

tensile strength (MPa)

De Schutter model
35 4
tensile strength (Mpa)
30 3
De Schutter model
(MPa)

25 3

20 2

15 2

10 1

5 1
0 0
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9

Hydratation degree (-)

Young modulus evolution

45

40
Young modulus (GPa)

35

30

25

20 Young modulus GPa

15 De Shutter model

10 Dynamique modulus

5 (De Schutter model

0
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9

Hydratation degree (-)

Figure 3. Mechanicals characteristics evolutions.

262
 6. Design and Construction Issues

For a first approach, we have take a bilinear evolution of the temperature. At the
beginning the ring temperature is maintained at 20°C. Then, at 25 hours (when
the pic of temperature is attempted; see Figure 2), the brass ring temperature
increases linearly with a rate of 0.35°C/h.
The model results show us (Figure 4: example of damage map for local damage)
that with different sizes of mesh (7 mm and 5 mm) and with two types of
computation (local and non local damage) we effectively obtain cracks.

D= 0

D= 1

Figure 4. Example of damage map ( local model; mesh size: 7 mm).

Although we didn’t take into account the creep in this model, the first numerical
results seem to show that we can effectively obtain cracking with our new test
(for suitable values of temperatures) and that we will be able to study the early-
age behaviour and cracking evolution due to restrained autogenous and thermal
shrinkage.

263
6. Design and Construction Issues

264
 7. Safety, Reliability, Risk and Margins

7. Safety, Reliability, Risk and Margins

Reliability of structures, systems and components. Application of probabilistic

methods to improve design, operation and safety of complex systems.
Probabilistic safety assessment (PSA). Internal and external events. PSA
and margin studies. Development of risk-informed approaches and risk-
based regulations, codes and standards. Human factors engineering.
Failure mode analysis. Severe accidents.

265
7. Safety, Reliability, Risk and Margins

266
 7. Safety, Reliability, Risk and Margins

Adjusting the fragility analysis method to

the seismic hazard input. Part I: The
intensity-based method (7-1567)

Kluegel, Jens-Uwe1, Attinger, Richard1, Rao, Shobha, B.2, Vaidya, Nish3

1
NPP Goesgen-Daeniken, Switzerland, e-mail: [email protected]
2
ABS Consulting, U.S.A, California, e-mail: [email protected]
3
Rizzo & Associates U.S.A, Pennsylviana, e-mail: [email protected]

The integrated oversight process of nuclear power plants in Switzerland contains

a significant amount of risk informed elements. National regulations require a
full scope (all internal and external events) PRA for all power and shutdown
operational modes at level 2. Thus, seismic PRA became an integrated part of
this regulatory framework. The development of a seismic PRA typically involves
the following the steps:
• Development of a site-specific seismic hazard input, usually based on
traditional PSHA
• A detailed fragility analysis for all safety important structures and components
• Human Reliability Analysis for post-earthquake actions
• Development of a plant logic model
• Quantification, sensitivity and uncertainty analysis.
Although the methodology of seismic PRA to some extent has become
standardized, different seismic PRA studies differ largely in their results. The
main reason for this observation consists in the different approaches used for
Probabilistic Seismic Hazard Analysis (PSHA) which defines the seismic hazard
input for the PRA. While older studies used an intensity-based PSHA
maintaining a close link between damaging effects of earthquakes (clearly and
uniquely defined in a probabilistic way by the corresponding intensity scale),
newer studies are based on SSHAC [1] type PSHA using ground motion models
for peak instrumental characteristics of earthquakes for the developing of the
hazard, which are not directly linked to damaging effects of earthquakes. This
causes problems as is shown in this paper, because the traditional fragility
analysis method used for seismic PRA is conditioned to a seismic hazard input
which represents strong, damaging earthquakes. Empirical fragilities for
structures and components are derived from field observations covering only
damaging earthquakes. Fragilities for components derived from tests are also
conditioned to the occurrence of damaging earthquakes, because qualification
tests are based on time histories representing high magnitude seismic events.
Therefore, the seismic hazard input derived from contemporary PSHA methods like

267
7. Safety, Reliability, Risk and Margins

SSHAC [1] is inconsistent to the method of fragility analysis as used for seismic
PRA. In case of the direct implementation of the results of the PEGASOS-project
[2] into the seismic PRA of NPP Goesgen this inconsistency of methods leads to a
potential overestimation of the seismic core damage frequency of up to a factor of
10 as is demonstrated in the paper. Therefore, a methodology has to be developed to
adjust the fragility method to the seismic hazard input as delivered by seismologists,
today. In this paper a procedure is proposed, which is aimed at a direct probabilistic
correction of this inconsistency between seismic hazard input and the fragility
analysis method. The main steps of the procedure are:
• Detailed deaggregation of the seismic hazard into a bivariate distribution
of controlling earthquakes, characterized by their magnitude and distance
for different spectral frequencies and different frequencies of exceedance
(hazard frequencies)
• Development or selection of appropriate models for the conversion of
controlling earthquakes into site intensities in the EMS-98 scale for
different levels of frequency of exceedance (hazard frequencies)
• Development of an enveloping distribution for site intensities resulting
from the distribution of controlling events.
• Calibrating the fragility analysis method to the developed distribution of site
intensities to ensure the same statistical damage prediction by the fragility
analysis method as predicted by the intensity distribution by using
appropriate vulnerabilities for classified structures (according to EMS-98)
• Developing a distribution for an adjustment factor applied to the
modelling factor considered in the traditional fragility method. This
assures a more meaningful damage prediction by the adjusted fragility
analysis and more reasonable PRA results.
The developed procedure fully accounts for available epistemic uncertainty and
random variability by performing an appropriate statistical calibration of PSHA-
results against the developed site intensity distributions. The application of this
procedure for the fragility analysis of NPP Goesgen is demonstrated. In a
company paper a more physical approach is suggested to complement this
statistical approach to the adjustment of fragility analysis.

References
1. Senior Seismic Hazard Analysis Committee (SSHAC). 1997. Recommendations for
Probabilistic Seismic Hazard Analysis: Guidance on Uncertainty and Use of
Experts, NUREG/CR-6372.

2. Probabilistic Seismic Hazard Analysis for Swiss Nuclear Power Plant Sites (PEGASOS
Project). 2004. Vol. 1–6, Nagra, Wettingen.

268
 7. Safety, Reliability, Risk and Margins

Adjusting the fragility analysis method to

the seismic hazard input. Part II: The energy
absorption method (7-1568)

Kluegel, Jens-Uwe1, Attinger, Richard1, Rao, Shobha, B.2, Vaidya, Nish3

1
NPP Goesgen-Daeniken, Switzerland, e-mail: [email protected]
2
ABS Consulting, U.S.A, California, e-mail: [email protected]
3
Rizzo & Associates U.S.A, Pennsylviana, e-mail: [email protected]

The integrated oversight process of nuclear power plants in Switzerland contains

a significant amount of risk informed elements. National regulations require a
full scope (all internal and external events) PRA for all power and shutdown
operational modes at level 2. Thus, seismic PRA became an integrated part of
this regulatory framework. The development of a seismic PRA typically
involves the following the steps:
• Development of a site-specific seismic hazard input, usually based on
traditional PSHA
• A detailed fragility analysis for all safety important structures and components
• Human Reliability Analysis for post-earthquake actions
• Development of a plant logic model
• Quantification, sensitivity and uncertainty analysis.
Although the methodology of seismic PRA to some extent has become standardized,
different seismic PRA studies differ largely in their results. The main reason for
this observation consists in the different approaches used for Probabilistic Seismic
Hazard Analysis (PSHA) which defines the seismic hazard input for the PRA.
While older studies used an intensity-based PSHA maintaining a close link
between damaging effects of earthquakes (clearly and uniquely defined in a
probabilistic way by the corresponding intensity scale), newer studies are based on
SSHAC [1] type PSHA using ground motion models for peak instrumental
characteristics of earthquakes for developing the hazard. These characteristics are
not directly linked to damaging effects of earthquakes. This causes problems as is
shown in this paper, because the traditional fragility analysis method used for
seismic PRA is conditioned to a seismic hazard input which represents strong,
damaging earthquakes. Empirical fragilities for structures and components are
derived from field observations covering only damaging earthquakes. Fragilities
for components derived from tests are also conditioned to the occurrence of
damaging earthquakes, because qualification tests are based on time histories

269
7. Safety, Reliability, Risk and Margins

representing high magnitude seismic events. Therefore, the seismic hazard input
derived from contemporary PSHA methods based on the direct use of ground
motion parameters is inconsistent to the method of fragility analysis as used for
seismic PRA. Hence a methodology has to be developed to adjust the fragility
method to the seismic hazard input as delivered by seismologists, today. Detailed
analysis has shown that the reason for the observed inconsistency consists in
neglecting the energy dependency in traditional fragility formulations which are
based on simple ground motion characteristics. In this paper a procedure is
proposed, which introduces energy (magnitude) dependency into traditional fragility
formulations. The procedure considers that the seismic demand in terms of energy
absorption is completely different for small and high magnitude earthquakes
defining a completely different damaging potential of these two categories of
earthquakes. It is shown that this difference in damaging potential can be related to
the ratio of a function of Arias intensity (or Cumulative Absolute Velocity, CAV)
for the controlling earthquake of the seismic hazard to the same function of Arias
intensity of a reference earthquake used to develop component-specific fragilities
from empirical observations. The main steps of the procedure are:
• Development of scenario-specific controlling earthquakes for different
frequencies of exceedance (hazard frequencies)
• Development of models for the evaluation of Arias-Intensity or
Cumulative Absolute Velocity (CAV)
• Derivation of reference earthquakes for different types of components used
for the development of generic fragilities from empirical observations.
• Development of a mean adjustment factor applied to the modelling factor
considered in the traditional fragility method based on the difference in the
different seismic demand with respect to energy absorption.
The developed procedure fully accounts for available epistemic uncertainty and
random variability. The application of this procedure for the fragility analysis of
NPP Goesgen performed for the implementation of the PEGASOS-hazard [2]
into seismic PRA is demonstrated. In a company paper a statistical approach is
suggested to complement this approach to the adjustment of fragility analysis.
The statistical approach can be used for quality assurance.

References
1. Senior Seismic Hazard Analysis Committee (SSHAC). 1997. Recommendations for
Probabilistic Seismic Hazard Analysis: Guidance on Uncertainty and Use of
Experts, NUREG/CR-6372.

2. Probabilistic Seismic Hazard Analysis for Swiss Nuclear Power Plant Sites (PEGASOS
Project). 2004. Vol. 1–6, Nagra, Wettingen.

270
 7. Safety, Reliability, Risk and Margins

On the treatment of dependency of

seismically induced component failures
in seismic PRA (7-1581)

Kluegel, Jens-Uwe
NPP Goesgen-Daeniken, Switzerland, e-mail: [email protected]

The integrated oversight process of nuclear power plants in Switzerland contains a

significant amount of risk informed elements. National regulations require a full
scope (all internal and external events) PRA for all power and shutdown
operational modes at level 2. Thus, seismic PRA became an integrated part of this
regulatory framework. Experience from past seismic PRA studies at NPP Goesgen
[1, 2] indicates that simplifying assumptions with respect to the treatment of
dependent failures of multiple redundant components strongly affects the results of
seismic PRA and shifts them to the conservative site. Typically, it is assumed that
all components of the same type characterized by the same fragility function and
located on the same floor of a building will fail together once the load capacity of
a single, representative component is exceeded. Therefore, full dependency
between the failures of similar components belonging to different redundancies of
the safety system is usually assumed. As long as seismic risk was treated in a
simplified way and was not found to be a significant risk contributor, such a
simplifying assumption could have been tolerated. For older plants with a low
number of redundancies (e.g. with a two-train safety system) and insufficient
physical separation of redundancies this assumption may even be justified today.
The situation is different for newer European plants (e.g. design of KWU, today
AREVA). Here the treatment of dependency of seismically induced failures of
multiple redundant components is of major importance because of the presence of
multiple redundant safety trains with good physical separation. The question is
also important for new designs. The dependency between seismically induced
failures in the Goesgen seismic PRA is modelled by a β-factor common cause
failure (CCF) model [1]. Generic β-factors have been applied to model the
dependency. This procedure was revised to address the results of the revised
fragility analysis and to incorporate the results of the PEGASOS PSHA study. The
Goesgen NPP has a four-train safety system with an additional two-train bunkered
special emergency safety system. Detailed analysis of the allocation of system
trains and their supporting equipment did show that most of the systems could be
represented by two groups of two components, where each group is expected to
show a different response to the same seismic demand. Therefore, the dependency
between the component failures can be divided into two steps:
• a first step treating the dependency between the two groups,

271
7. Safety, Reliability, Risk and Margins

• and a second step treating the dependency within the two components of
a common group.
Therefore, the problem can be reduced mathematically to the treatment of the
correlation in failure occurrence of two objects using two different β-factor
models in a staggered way. In the traditional fragility formulation the correlation
of failure occurrence [2] is described by the following equation:
β R1β R 2 β S 1β S 2
ρ= ρ R1R 2 + ρ S 1S 2 (1)
β R21 + β S21 β R22 + β S22 β R21 + β S21 β R22 + β S22
Here, ρ R1R 2, ρ S 1S 2 are the response and capacity correlation factor between
object 1 and 2; β Ri , β Si the lognormal standard deviations of the response and
capacity of component, i. The total correlation factor can be set equal to the
required β-factor assuming a quasi-linear dependency model.
For the first step of the procedure (between the two groups) the β-factor can
be derived from the simple assumption that ρ R1R 2 = 0 , while the second term in
equation (2) is maximized. This yields a β-factor of 0.5 between the two groups
of structural similar, redundant components.
For the second step a Monte-Carlo-procedure was developed, which evaluates
the fraction of concurrent failures of two components given the same seismic
load and assuming identical response of both components (first term in equation
(1) is maximized and set equal to 0.5). The correlation factor ρ S 1S 2 is evaluated
as the ratio of concurrent failures of both components related to the total
observed failure number in the sample. The Monte Carlo procedure considers
some minimal seismic diversity of the components caused by the uncertainty of
the mass distribution of similar components by incorporating some Gaussian
noise reflecting this uncertainty. Using the results, the β-factor can be evaluated
from equation (1).
A set of calculations for different boundary conditions with respect to the seismic
load, the combined uncertainty of the fragility function and with respect to the median
capacity of components was performed to derive a set of tabulated β -factors for use
in the Goesgen seismic PRA. The values can also be applied for generic applications.

References
Klügel, J.-U., Rao, S., Short, S. 2004. Challenges to Future Seismic PRA. In: C. Spitzer,
U. Schmocker, V. Dang (eds.). Probabilistic Safety Assessment and Management,
PSAM 2004, Springer, Berlin. Pp. 1232–1236.
Ogura, K., Fujuda, M., Sakagami, M., Ebisawa, K. 2006. Accident Sequence Study for
Seismic Event at the Multi-Unit Site, OECD, CSNI, Working Group on Risk
Assessment, Specialists Meeting on the Seismic PSA of Nuclear Facilities, Jeju
Island, Korea, 6–8 November 2006.

272
 7. Safety, Reliability, Risk and Margins

Time-dependent reliability of reinforced

concrete beams considering variability
in degradation due to reinforcement
corrosion (7-1590)

Yasuhiro Mori1, Kapilesh Bhargava2, A.K. Ghosh3

1
Professor, Graduate School of Environmental Studies, Nagoya University
Nagoya, Japan
2
Architecture & Civil Engineering Division, Bhabha Atomic Research Center
Mumbai, India
3
Health Safety and Environment Group, Bhabha Atomic Research Center
Mumbai, India
Correspondence e-mail: [email protected]

Keywords: corrosion, beams, degradation, reinforced concrete, structural reliability

Reinforced concrete (RC) structures in service may be affected by ageing, which

may include changes in strength and stiffness beyond the conditions that are
assumed in the structural design. Corrosion of reinforcement has been identified
as being the most widespread and predominant degradation mechanisms. The
associated reduction of steel area, spalling of concrete cover and/or the loss of
bond with time result in the loss of safety and serviceability. Therefore, the
evaluation of effects of corrosion damage on RC structures would help in
deciding the possible repair and continued service where their replacement is
economically unfeasible. Structural reliability evaluation will further help in the
efficient allocation of limited resources for periodic inspection and maintenance
of such structures.
In this paper, degradation of flexural and shear strength of RC beams due to
reinforcement corrosion is considered. For corrosion-affected RC beams, a
methodology is presented to generate time-dependent mean degradation function
which represents the ratio of degraded strength at any time from the initiation of
corrosion to the original strength. The various parameters that could affect the
time-dependent strength are identified as material strength of concrete and steel,
dimensions of the beam, covers to the reinforcements and the annual mean
corrosion rate. By considering the variability associated with the parameters
under consideration, the time-dependent variability associated with the
degradation function is evaluated. Simple analytical formulations are presented
for estimating the time-dependent failure probability considering variability in
degradation function as well as the stochastic nature of load processes, and
uncertainties in the initial strength. For illustration, time-dependent failure

273
7. Safety, Reliability, Risk and Margins

probabilities are presented for a corrosion-affected RC beam for flexural and

shear failure modes. It has been observed that the time-dependent failure
probability considering the variability in degradation function evaluated from
analytical formulations agrees well with the failure probability evaluated from
Monte Carlo Simulation. Parametric analyses show that failure probability is
sensitive to the variability associated with the degradation function if it is larger
than about 5%.

274
 7. Safety, Reliability, Risk and Margins

Component degradation effect on seismic

risk of NPP (7-1643)

In-Kil Choi, Minkyu Kim, Jin-Hee Park

Integrated Safety Assessment Division
Korea Atomic Energy Research Institute, Daejeon, Korea
e-mail: [email protected]

Introduction
Ageing related degradation of a nuclear power plant component is an important
aspect to secure the long term safety of the plant, especially for the seismic safety,
since the degradation of the components affect not only the seismic capacity of a
component but the response of a component. This can cause the change the seismic
margin of a component and over all seismic safety of a system. Finally this induces
the reduction of plant level seismic safety and the increment of CDF.

Degradation of safety significant components

Total of 530 components in Korean NPP were investigated and a database was
developed based on the investigation results. As a result of seismic walkdown,
crack and corrosion are the typical aging related degradation which can reduce
seismic safety of safety related equipment. In order to manage aging related
degradation data collected through the seismic walkdown in effective and
systematic database system is established.
KAERI (Korea Atomic Energy Research Institute) and BNL (Brookhaven
National Laboratory) are collaborating to develop seismic capability evaluation
technology for degraded components. To better understand the status and
characteristics of degradation of components in NPPs, degradation occurrences
of components in the U.S. NPPs were identified by reviewing recent publicly
available information sources and the characteristics of these occurrences were
evaluated and compared to the observations in the past.

Component degradation effect on seismic risk

The SPRA is an effective method to estimate a plant level risk quantitatively.
The purpose of a SPRA is to determine the probability distribution of core
damage due to the potential effects of earthquakes (ANS, 2007). SPRA is
performed based on four steps: seismic hazard analysis, component fragility
evaluation, plant system and accident sequence analysis, and consequence analysis.

275
7. Safety, Reliability, Risk and Margins

The component degradation effect on the seismic safety of a nuclear power

plant was performed using past SPRA model. The degradation effect on the
initiating event frequency was estimated according to the degradation level of
safety significant components.
Figure 1 shows the total CDF increasing ratio according to the component
degradation level. As shown in this figure, it can be revealed that the degradation of
major equipments can increase the seismic induced CDF. ECW is the most
influential equipment according to the degrading level. Also the degradation of the
Diesel Generator and the CS Tank also has greatly influenced the total CDF of NPP.

Conclusion
For the examination of ageing related degradation of a nuclear power plant
component, several times of plant walk-down were performed for Korean NPPs.
Total of 530 components were investigated and a database was developed based
on the investigation results. To better understand the status and characteristics of
degradation of SPCs in NPPs, degradation occurrences of SPCs in the U.S. NPPs
were identified by reviewing recent publicly available information sources, and
the characteristics of these occurrences were evaluated and compared to the
observations in the past. KAERI and BNL developed a component aging
database of US nuclear power plants using the NRC documents, NUREG report,
LERs, LRA, and generic comments.
For the evaluation of degradation effect of major equipment which related to
an initiating event, seven components and associated initiating events were
selected. When the seismic capacity of the selected equipments was decreased
from 10% to 50%, the initiating event frequency was determined. Through this
study, it was possible to identify the safety significant component to secure a
long term seismic safety of a plant.
2.50E+00
Instrumentation Tube CS Tank
Battery Charger Battery Rack
2.00E+00 Diesel Generator ECW Compression Tank
LOOP
Normalized CDF

1.50E+00

1.00E+00

5.00E- 01

0.00E+00
0% 10% 20% 30% 40% 50%
Degradation Ratio

Figure 1. Component Degradation Effect on Seismic Induced Core Damage Frequency.

276
 7. Safety, Reliability, Risk and Margins

Acknowledgement
This work was supported by Nuclear Research & Development Program of the
Korea Science and Engineering Foundation (KOSEF) grant funded by the
Korean government (MEST). (grant code: M20702030003-08M0203-00310).

Reference
American Nuclear Society (ANS). 2007. American National Standard; External-Events
PRA Methodology. ANSI/ANS-58.21.

277
7. Safety, Reliability, Risk and Margins

The reactor coolant circuit strength and the

safety and reliability issues (7-1651)

V.P. Semishkin, S.B. Ryzhov, V.A. Mokhov, V.A. Piminov, V.A. Grigoriyev
OKB «GIDROPRESS», Podolsk

The problem issues of performing justifications of the strength and reliability of

the WWER-1000 coolant circuit components at the stages of designing,
operation and extension of the operating period are considered. Distinctions in
the approaches to substantiation of the coolant circuit integrity in the Russian
and western regulatory documents are noted. The report deals with the most
complicated problems by solving which the service life is defined. The basic
differences in the approaches to the deterministic and probabilistic analyses of
strength are formulated. The approaches to safety analysis on the basis of
probabilistic assessments of the leaks and application of the LBB concept are
considered as well.
As the safety analysis must be conservative, the deterministic calculations
from the viewpoint of coolant circuit integrity must be taken to be conservative.
Any kinds of calculation uncertainties are considered so that for problem
statement the properties and characteristics of materials and media, the force and
thermal impacts and simplifications of calculation models and approximate
solution bring about the known worse results. Decreasing conservatism of
calculations for strength within the framework of safety analysis is an important
component of reducing the costs during designing, operation and extension of
service life. Here it is shown that there is available a boundary in the process of
decreasing conservatism and beyond this boundary there occurs growth of
uncertainties and to overcome the latter the considerable costs are required as
compared with benefits from properly decreasing conservatism.
The requirement of the regulatory documents on justification of operating
reliability of the primary circuit components and systems during the design
service life presumes for the reliability indices to be set related to probability of
no-failure operation, failure interval, technical life, service life, etc. For the
primary circuit equipment and pipelines, namely, for the vessels and pipelines,
and also for the systems, i.e. for the reactor, steam generators, pumps, etc. such
factors as damage to the walls, for example, formation of a through macrocrack
or division into the parts and loss of tightness of detachable joints are assumed to
be the basic reliability indices.
At the operation stage in the primary circuit components determining the
coolant boundary in accordance with the different mechanisms of degradation of
the materials there exist probability of formation and growth of the flaws and the
technical survey and metal inspection are used to reveal these flaws. Detection

278
 7. Safety, Reliability, Risk and Margins

of the flaws being inadmissible for further operation in case of their obvious
indication is performed using the non-destructive tests on the basis of both
deterministic and probabilistic approaches. In this case the failure probability is
determined considering statistical spread of data on unsoundness, mechanical
properties and crack growth resistance characteristics. The approaches to the
calculated support of the reactor plant operation are formulated including that one
on the basis of computerized monitoring of residual cyclic life.

279
7. Safety, Reliability, Risk and Margins

Improvement of the seismic fragility

analysis by use of the methods of structural
reliability and safety analysis (7-1655)
Dietrich Klein, Fritz-Otto Henkel
Woelfel Beratende Ingenieure GmbH + Co. KG, Max Planck Strasse 15,
97204 Hoechberg, Germany, e-mail: [email protected]

The key task in a seismic probabilistic safety analysis (PSA) is the fragility
analysis. Seismic fragility of a structure or equipment item is defined as the
conditional probability of its failure at a given value of the seismic input response
parameter. The peak ground acceleration (PGA) is commonly used as input
response parameter. The objective of fragility evaluation is to estimate the ground
motion capacity of the item and its uncertainty. Because there are many variables
in the estimation of this ground acceleration capacity, the fragility is described by
a family of fragility curves to reflect the uncertainty in the fragility estimation.
The mostly used method for fragility analysis is the scaling method. In this method
the family of fragility curves is described by three parameters: the median ground
acceleration capacity Ǎ, and logarithmic standard deviations βR for randomness and βU
for uncertainty. The fragility parameters Ǎ, βR and βU are estimated by an intermediate
random variable, the factor of safety F, which relates the acceleration capacity A to the
earthquake level specified for design ASSE. The factor of safety is the product of
individual factors which describe the conservatism in the design. All factors of safety
are assumed to be log-normally distributed, principally for its calculation convenience.
The use of lognormal mathematics in the scaling method is known to be an
erroneous approach in the tails of the lognormal distributions, even when the
lognormal shape adequately describes the data in the main parts of the distribution.
The probability of failure for seismic events, however, is generally low. To
improve this shortcut, the method of structural reliability and safety analysis as it
is used for reliability analysis of civil structures is proposed. In this methodology
the failure mode is defined by a function (limit state function) of deterministic and
stochastic parameters. The stochastic parameters (basic variables) may be
described by any arbitrary distribution function. The dispersion of the basic variables
is split into the inherent randomness and model uncertainty. The probability of
exceedance of the limit state function is calculated by an iterative approximation
method as function of the peak ground acceleration as variable parameter. The
advantages of this method are the appropriate calculation of the probability of
failure also for low probabilities as it is common for seismic events. Furthermore
this method allows the analysis of a complete structure as system by combination
of the individual failure modes in a minimal cut-set representation.
The application of the method is demonstrated at a pipe run of the rapid
shutdown system of a boiling water reactor as illustrative example.

280
 7. Safety, Reliability, Risk and Margins

The seismic fragility assessment of the feed

water tanks plant using robust prediction
concept of structural response (7-1664)
Pentti Varpasuo, Jukka Kähkönen
Fortum Nuclear Services Ltd, POB 100, 00048 FORTUM, Finland
e-mail: [email protected], [email protected]

Keywords: robust fragility assessment, structural response, uncertainty quantification

The computational task described in this paper is part of the updating effort of the
Seismic Probabilistic Risk Assessment for Loviisa Nuclear Power Plant.
Robustness in the prediction of structural response is an essential requirement for
probabilistic fragility assessment. In order to develop the updated fragility curves
for the feed water tanks in the Loviisa plant, which constitute the key element in
assessing the core melt frequency of the plant, very detailed structural model
incorporating the feed water tanks was developed. The fragility of a key plant
component is defined as the conditional probability of its failure given a value of
the response parameter, such as displacement, strain, stress or stress resultant. The
first step in generating fragility is to develop a clear definition of what constitutes
the failure of a component. It may be necessary to consider several modes of
failure, and fragility is required for each mode. To assess the structural performance,
it is important that robust predictions are made that treat all the uncertainties, from
modeling applicable loads to modeling the structural behavior.
For large tanks, a failure of the support system or a plastic collapse of the pressure
boundary is considered to be the dominant failure mode. Structures can be considered
to fail functionally when the inelastic deformations under seismic loads are estimated
to be sufficient to potentially interfere with the operability of equipment attached to the
structure or fractured sufficiently for equipment attachments to fail. The event and
fault trees should appropriately reflect these failure conditions.
The fragility of large components is directly developed from the seismic
response analysis results. The component fragility for a particular failure mode
is expressed in terms of the ground-acceleration capacity A. The fragility is
therefore the probability at which the random variable A is less than or equal to a
specified value, a. The ground-acceleration capacity is, in turn, modeled as
A = A εRεU (1)
In Eq. (1) A is the median ground-acceleration capacity, εR is variable representing
the inherent randomness about A, and εU is a random variable representing the
uncertainty in the median value. It is assumed that both εR and εU are log-normally
distributed with logarithmic standard deviations βR and βU, respectively.
As a result of the fragility evaluation the curves expressed with the aid of
Equation 1 are given for the feed water tanks of Loviisa plant.

281
7. Safety, Reliability, Risk and Margins

Longevity curves for probabilistic lifetime

analysis (7-1684)

Philippe Monette1, Pierre Joly1, Vincent Roux1

M.K. Ravindra2
1
AREVA NP, Plants Sector, Paris
2
M.K. Ravindra Consulting, Irvine, CA

Introduction and purpose

Seismic fragility analysis has become standard practice for assessing the seismic
safety of NPPs for a range of earthquake magnitudes, extending beyond the
design basis level. The methodology is well established, although it continues to
rely on expert knowledge that is not widespread. Individual component seismic
capacity is described by a family of “fragility curves”, defining the conditional
probability of failure as a function of a ground motion parameter. The
component fragility curves are often reduced to a high-confidence-of-low-
probability-of-failure (HCLPF) capacity for each safety important component.
The combination of component level HCLPF capacities, through boolean
expressions describing safety system dependencies, provides a plant level
HCLPF value that is viewed as the true reference seismic level up to which the
plant can be considered as “safe”. Typically, a plant that is designed to a 0.3g
design basis earthquake can normally be shown to have a HCLPF capacity of
0.5g, thanks to conservatisms accumulated in the seismic analysis design
process.
The purpose of this paper is to study the technical value and practicality of
adopting a similar methodology in an entirely different field, namely the
assessment of plant safety for long term operation in the presence of ageing; the
expected outcome being that one would likely be able to demonstrate a “high-
confidence-of-low-probability-of-failure” life duration largely exceeding the
design life.

Technical approach
In order to achieve the above objective, the concept of “longevity curves” is
introduced. Longevity curves are defined as the conditional probability of failure
of a safety relevant component as a function of its age (see Figure 1). Thanks to
the probabilistic formulation, an objective assessment is made possible of the
time (age) at which a nuclear installation will start seeing its safety level
degrading below its intrinsic, time-independent value, expressed in the normal

282
 7. Safety, Reliability, Risk and Margins

PSA practice as core damage frequency (CDF) or large early release frequency
(LERF).
The component level longevity curves should be propagated through PSA
modeling to obtain a plant level longevity curve or CDF over time. In this
process, plant vulnerabilities associated with a particular aging degradation issue
can be uncovered and possibly resolved through mitigation or component repair
or replacement, so as to increase the plant lifetime horizon.
A number of parallels can be drawn between the well established seismic
fragility analysis methodology and the proposed longevity analysis for plant
aging assessment. They will be extensively discussed in the paper. It is sufficient
here to note the following key features:
− the shape of the curves is similar and expected to be well represented by
Weibull or Lognormal distributions
− code based criteria are usually deemphasized in a PSA-based analysis as
they entail varying levels of conservatism
− uncertainties, being the result of randomness in component lifetime itself
and/or of incomplete knowledge of the parameters, are explicitly quantified
− as in a risk-informed approach, the PSA model must be exercised to
incorporate the safety significance of a particular component failure.
Longevity curves are associated with a particular age-related failure mode,
affecting a long-lived component. They can be derived from a full probabilistic
analysis using time-dependent properties and damage modeling. Alternatively,
they can be estimated through a progressive analysis of design safety factors that
are treated as random variables characterized by their median value and standard
deviations. Such safety factors can be grouped in several categories such as:
− FMD for material and design (physical component properties)
− FLE for loads and environment (time dependent solicitations)
− FDT for defect tolerance (safety margin beyond the appearance of a defect).
The longevity curves are then expressed as:
A = FMD*FLE*FDT*ADesign
ADesign being the design life.
Contributors to each of those factors will be illustrated on particular examples of
degradation mechanisms affecting long-lived components of PWRs.
The incorporation of longevity curves into the PSA and the critical aspect of
acceptable risk increase due to age will also be examined.

283
7. Safety, Reliability, Risk and Margins

Conclusion
The seismic fragility methodology provides a blueprint for developing a
longevity analysis methodology, with necessary adaptations. The present paper
will serve as a proof of concept. Longevity curves could provide a consistent
framework for assessing the plant safety impact of individual component aging
issues.

Figure 1. Mean, Median, 5% Non-Exceedance and 95% Non-Exceedance Longevity Curves

for a Component.

284
 7. Safety, Reliability, Risk and Margins

Reliability and safety analysis of raft

foundations under dynamic loading (7-1688)

Suchibrata Dalal1, Dr. Mahua Chakrabarti2

1
Post-Graduate Student, Department of Structural Engineering, Veermata Jijabai
Technolical Institute, Dr. H.R. Mahajani Marg
Matunga, Mumbai 400 019, India, e-mail: [email protected]
2
Professor, Department of Structural Engineering, Veermata Jijabai Technolical
Institute, Dr. H.R. Mahajani Marg, Matunga, Mumbai 400 019, India
e-mail: [email protected]

Moderately thick plates are frequently encountered in civil engineering. Typical

plate structures on elastic foundations are machine foundations and raft
foundations, concrete pavements of highways and airfields etc subjected to
impulsive and harmonic type of loadings. These structures are predominantly
subjected to dynamic loadings and can be modeled mathematically by assuming
a thick to moderately thick plate resting on an elastic foundation.
Some of the most important types of dynamic loadings, nuclear reactors’ raft
foundations are subjected to are seismic loading due to earthquake ground
motion and impulsive loading such as due to explosion, aircraft impact etc.
These raft foundations have been modeled as thick plates on elastic foundations
and the plates are analyzed using Reissner-Mindlin plate theory which considers
first order shear deformation effects thus leading to better representation of
actual behaviour of thick plates. It is very essential to ascertain in advance how
reliable the raft foundations are during there operational life subject to the action
of dynamic loading while exposed to severe environmental conditions. Dynamic
loads may govern the analysis and design of raft foundations and knowledge of
reliability of these structures beforehand will improve the structural design of
raft foundation so as to sustain dynamic impulsive loads and severe
environmental loads. The present study aims at developing formulations for
ascertaining the reliability of nuclear reactors’ raft foundations subject to
impulsive type of dynamic loading and also deriving methods for determination
of safety margin. The analysis and derivations have considered plate-foundation
interaction. The foundation parameters characterize the compressive strain and
shearing strain in the foundation.
For determining reliability first the modes of failure of the structure have to be
assessed and probability of failure to be determined. Deflections of the structure
have been assumed to be the governing parameter for evaluating failure
conditions. Based on several factors viz. foundation stiffness, transverse shear
deformation, plate aspect ratio, shape and duration of impulsive load, loaded
area, and initial membrane stress etc the maximum deflections will be different
for different loading combinations. To remain safe under the action of dynamic

285
7. Safety, Reliability, Risk and Margins

loads, these maximum deflections should be less than the permissible deflections
for respective load cases as per codes of standard applicable. In this paper two
formulations have been investigated for studying dynamic behaviour of
moderately thick plates. The first formulation is based on the Reissner-Mindlin
plate theory, considering the first order shear deformation effect and including
the plate-foundation interaction and thermal effects. The formulations have been
extended to the case of large deflections of Reissner-Mindlin plates. The second
formulation assumes Mindlin plate theory; the governing equations of dynamic
equilibrium have been derived using Hamilton’s principle and Euler – Lagrange
equation of calculus of variation. Using these two formulations, programs have
been prepared and validated for evaluation of deflection and bending moment.
Failure equations have been developed based on values of maximum deflection
and corresponding bending moment. A procedure has been suggested to
determine the reliability of raft foundations.
The plate-foundation interaction has been considered in the analysis and the
plate is assumed to vibrate and deform along with the foundation simultaneously.
The study assumes flexibility of the raft as opposed to conventional engineering
practice which assumes the raft foundation as fixed leading to larger stress
resultants under dynamic loading conditions.

286
 7. Safety, Reliability, Risk and Margins

Challenges in the application of

probabilistic safety goals for nuclear
power plants (7-1769)

Jan-Erik Holmberg1, Michael Knochenhauer2

1
VTT Technical Research Centre of Finland
P.O. Box 1000, FI-02044 VTT, Finland, e-mail: [email protected]
2
Relcon Scandpower AB, P.O. Box 1288, SE-172 25 Sundbyberg, Sweden
e-mail: [email protected]

The paper will deal with challenges in the application of probabilistic safety
goals, as analysed in the Nordic (Sweden/Finland) project dealing with the use
of probabilistic safety criteria for nuclear power plants (NPP). The project has
relations to an on-going OECD/NEA WGRisk task on probabilistic safety
criteria in member countries.
Safety goals are defined in different ways in different countries and also used
differently. Many countries are presently developing them in connection to the
transfer to the risk-informed regulation of both operating NPPs and new designs.
However, it is far from self-evident how probabilistic safety criteria should be
defined and used. On one hand, experience indicates that safety goals are
valuable tools for the interpretation of results from a probabilistic safety
assessment (PSA), and they tend to enhance the realism of a risk assessment. On
the other hand, strict use of probabilistic criteria is usually avoided. A major
problem is the large number of different uncertainties in PSA model, which
makes it difficult to demonstrate the compliance with a probabilistic criterion.
Further, it has been seen that PSA results can change a lot over time due to scope
extensions or increases of level of detail, typically leading to an increase of the
frequency of the calculated risk. This can cause a problem of consistency in the
judgments.
This paper will give an overview of the current situation with probabilistic
safety goals with emphasis on challenges in applying them. The following items
will be discussed:
• use of safety goals in a strict manner (limiting values) vs. as targets
(orientation values),
• definition of valid subsidiary safety goals at the levels of core damage risk
(for level 1 PSA) and large release risk (for level 2 PSA)
• needs to internationally harmonize probabilistic safety criteria for NPPs
• qualification of PSA for the application of probabilistic safety criteria
• differences/similarities in the use of safety goals for new and operating NPPs.

287
7. Safety, Reliability, Risk and Margins

Integrated soil-structure fragility analysis

method for nuclear structures (7-1771)

Alidad Hashemi1, Tarek Elkhoraibi2

1
Senior Engineer, PhD, PE, Bechtel National,
50 Beale St, San Francisco, CA, USA, e-mail: [email protected]
2
Senior Engineer, PhD, Bechtel National,
50 Beale St, San Francisco, CA, USA, e-mail: [email protected]

The Integrated Soil-Structure Fragility Analysis (ISSFA) method, presented in

this paper, is intended to utilize the recently available computation power with
the established analysis techniques to develop a more reliable foundation for the
performance based seismic design.
Performance based seismic design of structures is the accepted design
methodology required by the Nuclear Regulatory Commission (NRC) for
nuclear safety related projects. One of the essential steps in the Performance
based seismic design approach specified in ASCE 43-05 for the seismic design
criteria for Structures, Systems and Components (SSC) in Nuclear Facilities is
seismic fragility analysis. The goal in fragility analysis of SSC is to estimate the
probability of failure – defined as unacceptable seismic behavior – of SSC as a
function of some ground motion quantity representing the intensity of the
seismic event. The failure of SSC is typically defined in terms of seismic force
(or deformation) demands exceeding their corresponding capacities leading to an
unacceptable seismic behavior. Thus, to determine the fragility function at each
level of ground motion intensity, a realistic estimation of the probabilistic
distribution of the demand and capacity parameters is necessary. While mean
and dispersion of the capacity parameters may be prescribed, controlled and
evaluated by design criteria, determining the probabilistic distribution of the
demand parameters generally involve significant analytical effort.
The SSC seismic fragility functions are combined with the seismic site hazard
curves to obtain the annual failure probability (i.e. performance) for the subject
SSC. The NRC performance goal is stated as 1 in 10000 over the preclosure
period for Yucca Mountain Project in the US. Considering a preclosure period in
the order of 100 years, the performance goal can be interpreted as a target annual
failure probability of 10-6.
Traditionally, simplified methods for estimating both the mean seismic
demand values and their corresponding dispersions are developed in building
and nuclear industries. These methods, while attempting to minimize the
computational effort, typically rely on empirically obtained and prescribed
mathematical relationships for the fragility function and (linear) structural
analysis at limited (often one) ground motion intensity levels, and are typically

288
 7. Safety, Reliability, Risk and Margins

overly conservative resulting in over-design of structures. In this paper, the

fragility functions corresponding to Seismic Margin Analysis (SMA) Method
defined in terms of High-Confidence-Low-Probability-of-Failure (HCLPF)
capacity are used as a benchmark for comparing the results obtained from the
featured ISSFA method.
In the past few years, prompted by newly available computing power,
researchers have explored new techniques for determining seismic fragility
functions in the building industry taking advantage of methods such as nonlinear
Mont Carlo time-history analysis and nonlinear incremental dynamic analysis.
An equivalent effort to make parallel explorations and advancements in the
nuclear industry, however, is lacking. This topic is more challenging for nuclear
industry structures considering that these structures are typically very stiff and
the response nonlinearity when subjected to large seismic motions is mostly
associated with the soil layers and soil-structure-interaction (SSI). Other sources
of nonlinearity in the response of nuclear structures that are typically ignored in
the current practice of fragility analysis are foundation sliding, rocking and
separation between soil and foundation. The framework developed in this study
integrates the aforementioned effects in fragility analysis for nuclear industry
structures.
The input data are the site specific seismic hazard curve and rock motions at
different earthquake intensity levels and representative randomized soil profiles
as well as the structural model of the subject SSC. Latin Hypercube Sampling
technique (LHS) is used throughout the analysis to propagate the input
uncertainties through the SSI analysis. The output products are the fragility
functions for specified limit states that in conjunction with the seismic hazard
curve yield the annual probability of failure (performance) for SSC.
In this paper, the developed ISSFA methodology is outlined and discussed.
Furthermore, a comprehensive example analysis is implemented and a comparison
is drawn with the currently applied simplified SMA method.

289
7. Safety, Reliability, Risk and Margins

Risk-informed implementation of results

from modern seismic hazard analyses
into the design of new buildings of
the existing NPP’s (7-1773)

Sener Tinic, Martin Richner

Nordostschweizerische Kraftwerke AG, Nuclear Power Plant Beznau,
CH-5312 Doettingen, Switzerland
e-mails: [email protected], [email protected]

The modern seismic hazard studies like PEGASOS for the existing NPP-sites
exhibit higher spectral acceleration and uncertainties compared to the former
seismic hazard studies for the same probability level. The informed technical
and scientific community are discussing since a half decade the related topics
like methodologies, assumptions to source characterization, appropriate attenuation
models, site response issues, aggregation of expert models, plausibility checks of
hazard results, etc. It seems that many issues of the seismic hazard will not be
resolved in the short time period. However existing NPP’s continue their
operation and need to demonstrate the fulfillment of the national and international
safety criteria. After the completion of PEGASOS-Project, the Swiss NPP’s
reviewed results together with international experts. The main conclusion of the
review is that the uncertainties could be reduced, for example by means of
additional site investigations. Swiss licensees have already started detailed
geological and geotechnical investigations at the existing plant sites. The results
of site investigations will be implemented in the follow-up project namely in the
Pegasos Refinement Project (PRP). However these investigations, new site response
calculations and the conducting of the PRP using different expert groups with a
high level of detail and appropriate quality will require about four years.
In the meantime, the PEGASOS-results were implemented into the Probabilistic
Safety Assessment of the NPP Beznau [2]. According to the PSA-results seismic
is the dominating risk contributor to the Core Damage Frequency (CDF) and the
Large Early Release Frequency (LERF). Independent of the PSA results, some
new buildings and equipments are being planed to replace the emergency power
supply from a nearby hydro plant. To design the new buildings and new equipment,
the seismic input has to be defined. In this regard this paper proposes a risk-
informed approach to determine the seismic design input. The main goal is to
result in a low contribution of the new buildings and equipment to the seismic
CDF and LERF. The proposed procedure leads to a consistent seismic safety for
several buildings of the existing plant as well as avoids especially for foundations
high construction costs due to a disproportional seismic input.

290
 7. Safety, Reliability, Risk and Margins

References
1. PEGASOS, Probabilistic Seismic Hazard Analysis for Swiss Nuclear Power Plant Sites,
31. July 2004.

2. Richner, M. et al. Insights Gained from the Beznau Seismic PSA including Level 2
Considerations, ANS PSA 2008 Topical Meeting – Challenges to PSA during
the nuclear renaissance, Knoxville, Tennessee, September 7–11, 2008.

291
7. Safety, Reliability, Risk and Margins

PSA Level 2 – Experience with the review

process from the perspective of the
independent evaluator (7-1778)

Karl Götz
TÜV SÜD Industrie Service GmbH, Munich, Germany
e-mail: [email protected]

According to the current German PSA guide dating from August 2005, in the
framework of the safety review, a PSA Level 2 has to be delivered by the operator
every ten years. This analysis has to be made in compliance with the “state-of-
the-art” scientific methods.
In the meantime, several analyses have been performed in Germany – based
on the common understanding that performing a PSA level 2 lies within the
operator’s field of responsibility.– The legal status (regarding the influence of
the results on the supervision and licensing process) remains open but a review
of the analysis by the regulator and its independent experts was carried out.
TÜV SÜD Industrie Service GmbH, Energy and Technology, was approved
by the regulatory authority with the implementation of the mandated reviews for
several NPPs in the south of Germany (both for PWR and BWR types).
Based on mutual understanding the IAEA-TECDOC-1229 “Regulatory review
of Probabilistic Safety Assessment (PSA) level 2” represented the guideline and
evaluation criteria.
The aim of this work – shown in this presentation – was to evaluate a German
PSA Level 2 according to international criteria, to gather experience with the
review process in comparison with the “old way” of a detailed expertise
generally applied to German supervision procedures and, of course, to prove the
quality and the results of the submitted analyses. The most important initial step
of the review process was to bring national demands into line with international
requirements and procedures and to specify IAEA TECDOC 1229 in terms of
appropriate procedures and requirements.
The essential results of our work were
• To demonstrate how to convert international to national procedures
• To confirm that the selected approaches and methods are “state-of-the-
art” in science and technology
• To prove that the qualitative and quantitative results of the analyses are
traceable and understandable

292
 7. Safety, Reliability, Risk and Margins

• To identify the optimization measures that should be implemented

quickly, including hardware changes as well as changes to the operating
rules of the plant.
In conclusion, the review procedure has shown to be successful from the
perspective of the evaluator and the other parties. The high safety level of the
plants was reflected by the results of the analyses. The chosen approach is suited
to assess the analysis carried out with respect to methodical approach and quality.

293
7. Safety, Reliability, Risk and Margins

Probabilistic fracture mechanics: PTS

Screening Criteria for RTNDT, application
of FAVOR code to a German
KONVOI plant (7-1785)

Ralf Tiete1, Norbert Schlüter2

1
AREVA NP GmbH, Erlangen, Germany, e-mail: [email protected]
2
Kernkraftwerke Lippe-Ems GmbH, Lingen, Germany

In Germany the structural integrity and safety of reactor components, like

reactor pressure vessel (RPV) or pipes, is done by deterministic analyses.
Deterministic approaches use conservative assumptions (crack geometry, external
loadings, material properties, etc.) to maximize the safety margins. Reasons for
such conservative assumptions can be missing information (aleatory uncertainties)
or missing knowledge of certain mechanisms (epistemic uncertainties).
A probabilistic analysis uses the same methods (e.g. calculation method for
stress intensity factor, crack propagation) as a deterministic one, but addresses
the uncertainty of required input data or mechanisms. Unlike a deterministic
analysis, which criterion is the achievement of a critical or reference value (e.g.
stress intensity factor reaches lower bound of fracture toughness), a probabilistic
analysis gives a probability of flaw initiation or component failure. Therefore a
probabilistic analysis of a reactor component gives an additional classification of
the integrity (safety) of such a component under more realistic assumptions and
helps quantifying governing parameters for the component failure, which can be
useful for lifetime extension.
An example of such a probabilistic safety analysis is the FAVOR (Fracture
Analysis of Vessels, Oak Ridge) computer program (see Williams et al. (2007)),
which was developed in the US by the Oak Ridge National Laboratory, and is
applicable to the core region of a RPV. FAVOR was applied to the RPVs core
region of a German KONVOI-type nuclear power plant (NPP) with the aim of
defining a Screening Criteria for the nil ductility transition range temperature
RTNDT assuming pressurized thermal shock (PTS) transients.
The application of FAVOR, version 06.1, to a German KONVOI-type nuclear
power plant (NPP) is done with the help of the following adjustments:
- For the one dimensional finite elements stress calculation in FAVOR, the
thermal hydraulic conditions (convective heat transfer and coolant
temperature at inner vessel wall) obtained from KWU-MIX were averaged
in axial and circumferential direction.

294
 7. Safety, Reliability, Risk and Margins

- A distribution of postulated flaws was developed from results of ultrasonic

(UT) examinations of the RPVs core region.
- As most of the UT indications are assumed to be underclad defects, own
calculated stress intensity factors for underclad defects were implemented.
The linear elastic stress intensity factor for an underclad defect was
calculated according to Marie et al. (2008) with a plastic correction
obtained from RSE-M (2000).
Several FAVOR applications with increasing cumulated neutron fluence (E > 1 MeV)
at inner vessel wall according to effective full power years were made, in order
to correlate the maximum reached RTNDT in the circumferential weld or in the
two forgings of the core region with the failure frequency (failure probability per
year). For the Screening Criteria a maximum allowable failure frequency of 10-6
was chosen. This leads to a maximum allowable RTNDT in the circumferential
weld of the core region of 285°C and in the two forgings of the core region of
170°C.
Without changing the computer code of FAVOR the user has to deal with
some possible implicit restrictions or assumptions. For example the implemented
Weibull model for initiation fracture toughness or the lognormal distribution for
crack arrest toughness can lead to an over – or underestimation of flaw initiation
or crack arrest. Therefore it might be useful to adapt the computer code or
develop an own probabilistic fracture analysis tool for the application of interest
and its potential unique boundary conditions.
As conclusion, probabilistic fracture analyses are powerful tools to define
governing parameters for component failure like RTNDT under realistic
conditions. This also helps to define governing parameters for lifetime
extension. To ensure that the application of interest will be done without any
unnecessary conservative assumptions, existing PFM tools like FAVOR may
need to be adapted or an own tool may need to be developed.

References
Marie, S., Chapuliot, S. 2008. Improvement of the calculation of the stress intensity
factors for underclad and through-clad defects in a reactor pressure vessel
subjected to a pressurized thermal shock. International Journal of Pressure
Vessels and Piping 85, pp. 517–531.

RSE-M, Appendix 5.4, 2000. Règles de surveillance en exploitation des matériels

mécaniques des îlots nucléaires REP. AFCEN.

Williams, P.T., Dickson, T.L., Yin, S. 2007. Fracture Analysis of Vessels – Oak Ridge
FAVOR, v06.1, Computer Code: User’s Guide. ORNL/TM-2007/0031, Oak
Ridge National Laboratory.

295
7. Safety, Reliability, Risk and Margins

Effects of AAR on seismic assessment

of nuclear power plants for life
extensions (7-1789)
Julia Tcherner (nee Milman), Tarek S. Aziz
Engineering and Technical Delivery, Atomic Energy of Canada Limited
Mississauga, Canada, e-mails: [email protected], [email protected]

In Nuclear Power Plants, the containment structures provide an ultimate barrier

to fission product releases to the environment. The design has been performed
and the operating strategies are in place to ensure that this final barrier is
effective. It is very important to understand that the containment capabilities and
the margins in performance of the containment to resist beyond design loads and
severe accidents and to improve containment performance are being considered.
Several NPPs are nearing the end of their original design life and as service
life is being extended, containment improvements are performed in order to
meet the Probabilistic Safety Assessments (PSA) related targets. As part of the
PSA work, Seismic Margin Assessment (SMA) is performed as seismic
requirement has increased for some plants.
To perform SMA work, new Floor Response Spectra (FRS) for the Reactor
Building (R/B) containment structure had to be developed, which in addition to
new seismic requirements had to consider current condition of the structure
including possible aging related degradation. Alkali Aggregate Reaction (AAR)
was identified as applicable Aging Related Degradation Mechanism (ARDM)
for one of the plants. Alkali reactive aggregate was used for construction and,
although measures were implemented to mitigate possible reaction, evidences of
AAR were found in some parts of the concrete.
Investigation was performed in order to establish the means of accounting for
AAR in the development of Floor Response Spectra (FRS) for R/B to be used in
SMA work. Review of studies and experiments to determine effects of AAR on
mechanical properties of concrete and evaluation of the effects of restraint
provided by the reinforcing and prestressing steel was considered. The main
findings of the investigation are described in the paper.
Based on this investigation and considering geometry and current condition of
the post-tensioned R/B, it was concluded that no change in modulus of elasticity
to account for AAR is necessary for generation of seismic FRS. However, it was
considered prudent to include a possible reduction in modulus of elasticity,
which was quantified as 15% in order to account for possible local variations in
material characteristics, environment of exposure, and chemical reactivity.
Measured compressive strength of the R/B concrete was higher than the
strength used in original design calculations. Thus, an increase in the modulus of
elasticity needed to be considered for seismic FRS generation for R/B.

296
 7. Safety, Reliability, Risk and Margins

Site-specific ground motion models for soil

sites with thick sedimentary layers (7-1795)

Jochen Schwarz1, Christian Golbs2, Christian Kaufmann1, Gebhard Roth3

1
Bauhaus-Universität Weimar, Earthquake Damage Analysis Center
Marienstraße 13, D-99421 Weimar, Germany
e-mail: [email protected]
2
Seismotec GmbH, Thomas-Müntzer-Straße 35, D-99423 Weimar, Germany
3
EnBW Kraftwerk AG, Kernkraftwerk Philippsburg, Rheinschanzinsel
D-76661 Philippsburg, Germany, e-mail: [email protected]

As so far ground motion prediction models (GMPM) for soil sites might be
misleading if the selection of recordings is restricted to the uppermost 25 m or
30 m of the subsoil (and the average shear wave velocity Vs,25 or Vs,30) and if the
whole underlying geological depth profile is ignored. The concept of subsoil and
geology-dependent ground classes (recently introduced into German Code DIN
4149) offers an alternative approach by the explicit consideration of the
thickness of sedimentary layers and their effect on the site amplification as well
as on the shape of site-specific spectra. As a whole, six respectively seven site-
specific subsoil classes are distinguished. Despite the fact that alongside the
river Rhine in Western and South-Western Germany, the thickness of the
sediments is reaching several hundred metres, site-specific data and
corresponding attenuation relationships are missing due to lack of earthquakes
and strong-motion recordings.
Within a series of comprehensive instrumental site studies supported by
USGS, Californian strong motion stations are classified with respect to their
ground classes considering the characteristics of the uppermost 25 m of the
subsoil overlaying the geological depth profile (Lang & Schwarz, 2006). The
commonly used subsoil class “soft soil” (C) is replaced and differentiated by
three classes (C-R, C-T, C-S), where C-R stands for soft soil above rock (with
high amplification factor in a small plateau range), C-S for layers with more than
100 m thickness, and C-T for a transition range. Ground class C-S is connected
with reduced site soil amplification factors (which can also be derived from site
response analysis). Taking profit from an accompanying study (see Schwarz et
al., 2007), as a whole 484 records from Californian “soil site” strong motion
stations are considered. From this basic or primary dataset (DS I) spectral
attenuation relationships are elaborated with different regression methods; the
one with the smallest standard deviation is used.
The attribute of “site-specific” implements two further elements: the
deaggregation of Probabilistic Seismic Hazard Assessment (PSHA) for the
relevant design hazard level (i.e. 10-5/a for N.P.P.) and the selection of recorded
ground motion for the mean or modal (controlling) magnitude-distance combination.

297
7. Safety, Reliability, Risk and Margins

According to these principles, the design ground motion (DGM) is regarded as

site-specific if ground motion data for the local depth profile are considered,
exclusively, and if the database corresponds with the deaggregated hazard from
PSHA results. In the low seismicity areas of Central Europe and in cases where
DGM for very low probability rates of exceedance is required, small magnitude
and near field events (often neglected in common GMPM) are of importance.
Therefore, further sets of data (DS II) are taken from aftershock recordings in
Turkey during in 1999/2000 (DS II) and – as an innovative element of the whole
approach – from seismic instrumentation at reference site(s) in Germany
contributing to a third dataset (DS III).
While the number of records in dataset DS I is permanently growing due the
events recorded from Californian stations, for dataset DS III a longer time
window is required, not at least due the low level of the seismic activity. The key
and linking element of the presented procedure is the instrumental subsoil
classification making the different datasets comparable and unique. Ongoing
studies are related to a further refinement of the selection criteria by comparing
not only the peak and amplitude level of H/V-spectra but also the frequency-
dependent shape by a cluster analysis of all stations related to the target site and
its depth profile.
The whole procedure can be characterized as a „single-station“ approach
considering records from stations of comparable subsoil profiles under the
assumption of similar site amplification effects. The impact of the datasets, their
combination and composition on the ground motion prediction models (GMPM)
are studied. Besides the realistic description of site-specific ground motion it
will be discussed to which extent the procedure is contributing to a reduction in
the uncertainty of ground motion models.

References
Lang, D.H., Schwarz, J. (2006). Instrumental subsoil classification of Californian strong
motion sites based on single-station measurements. Eighth U.S. National
Conference on Earthquake Engineering, San Francisco, California, April 18–22,
2006.

Schwarz, J., Lang, D.H., Kaufmann, Ch., Ende, C. (2007). Empirical ground-motion relations
for Californian strong-motion data based on instrumental subsoil classification.
th
9 Canadian Conference on Earthquake Engineering, Ottawa, Ontario, June
25–29, 2007.

298
 7. Safety, Reliability, Risk and Margins

Seismic performance assessment for

safety-related nuclear structures (7-1818)

Yin-Nan Huang1, Andrew Whittaker2, Nicolas Luco3

1
Postdoctoral Research Associate, State University of New York at Buffalo
221 Ketter Hall, Buffalo, NY 14260, USA, e-mail: [email protected]
2
Professor, State University of New York at Buffalo, USA
e-mail: [email protected]
3
Research Structural Engineer, United States Geological Survey, USA
e-mail: [email protected]

Seismic Probabilistic Risk Assessment (SPRA) was developed in the early

1980s and subsequently accepted by the United States Nuclear Regulatory
Commission (USNRC) to be used in nuclear power plant (NPP) Individual Plant
Examination of External Events (IPEEE). The most widely used SPRA
procedure is the Zion method, which was first developed and applied in the
Oyster Creek probabilistic risk assessment and later improved and applied in
1981 to estimate seismic risk for the Zion Plant (Pickard, Lowe, and Garrick et
al. 1981). In the Zion method, the component fragility curves are defined in terms
of ground-motion parameters (generally peak ground acceleration), although the
failure of a component has a much improved correlation to structural response
parameters, such as floor spectral acceleration and story drift.
Procedures for seismic performance assessment of buildings have been
developed in the ATC-58 project and proposed in the 50% draft Guidelines for
Seismic Performance Assessment of Buildings (ATC 2008) (termed the draft
ATC-58 Guidelines hereafter). These procedures use fragility curves that are
defined using structural response parameters. The procedures in the draft ATC-
58 Guidelines cannot be used directly for performance assessment of NPPs
because the methodology does not accommodate accident sequences, event trees
and fault trees but provides the robust technical basis needed to develop an
alternative procedure for seismic probabilistic risk assessment for NPPs.
This paper introduces a new procedure based on the Zion method and the
methodology presented in the draft ATC-58 Guidelines for seismic performance
assessment of NPPs (Huang et al. 2008, 2009). The new procedure improves the
Zion method by using nonlinear response-history analysis and structural response-
based fragility curves. This procedure is used to evaluate seismic risk of a
sample NPP reactor building of conventional and base-isolated construction. The
impact of the implementation of base isolation on the seismic performance of the
sample NPP is identified.
The proposed procedure includes five steps. Step 1 involves the characterization
of the seismic hazard. The procedure allows the seismic hazard to be defined
using a user-specified intensity of earthquake shaking, a user-specified scenario

299
7. Safety, Reliability, Risk and Margins

of earthquake magnitude and distance or a time-based representation considering

all possible earthquakes. The final products of intensity – and scenario-based
assessments are the probability of unacceptable performance of the NPP to the
specified intensity and scenario of earthquake shaking, respectively. The final
product of a time-based assessment is the annual frequency of unacceptable
performance of the NPP. In this paper, only the results for the time-based
assessment of the sample NPP are presented since the annual frequency of
unacceptable performance is the most widely used index for risk assessment of
NPPs.
The seismic risk for the sample conventional and base-isolated NPP reactor
buildings was evaluated using the procedure described above with a focus on the
secondary systems in the sample NPP. The mean annual frequency of
unacceptable performance of the base-isolated NPP is nearly orders of
magnitude smaller than that of the conventional NPP.

References
Applied Technology Council (ATC). (2007). Guidelines for seismic performance assessment
of buildings. ATC-58 50% Draft, Applied Technology Council, Redwood City,
California.

Huang, Y.-N., Whittaker, A.S., Luco, N. (2008). Performance assessment of conventional

and base-isolated nuclear power plants for earthquake and blast loadings.
MCEER-08-0029, Multidisciplinary Center for Earthquake Engineering Research,
State University of New York, Buffalo, NY.

Huang, Y.-N., Whittaker, A.S., Luco, N. (2009). A seismic performance assessment

methodology for safety-related nuclear structures. Paper submitted for review
and possible publication, Nuclear Engineering and Design.

Pickard, Lowe, Garrick, Inc., and Westinghouse Electric Corporation, Fauske & Associates,
Inc. (1981). Zion Probabilistic Safety Study. prepared for Commonwealth Edison
Company, Chicago.

300
 7. Safety, Reliability, Risk and Margins

Development of a reliability data handbook for

piping components in Nordic nuclear power
plants (7-1837)

Anders Olsson1, Relcon Scandpower, Vidar Hedtjärn-Swaling2,

Relcon Scandpower & Bengt Lydell3, Scandpower Risk Management Inc.
1
Relcon Scandpower AB, Östra Förstadsgatan 34, 212 12 Malmö, Sweden
e-mail: [email protected]
2
Relcon Scandpower AB, Englundavägen 13, 172 25 Sundbyberg, Sweden
e-mail: [email protected]
3
Scandpower Risk Management Inc., 4 Houston Center, 1331 Lamar, Suite
1270, Houston, Texas 77010, USA, e-mail: [email protected]

The Nordic PSA Group (NPSAG) has undertaken to develop a piping reliability
parameter data handbook for use in risk-informed applications that involve the
consideration of structural integrity of piping systems. The scope of the
handbook is to establish high quality reliability parameters that account for the
Nordic and worldwide service experience with safety-related and non-safety-
related piping systems in a consistent and realistic manner.
While the work to develop the handbook is expected to be finalized during
2009, the planning for its preparation has been underway for well over ten years.
An important step towards the handbook development project has been the
international cooperative effort through the OECD Nuclear Energy Agency to
create an event database (OPDE) on the service experience with piping in
commercial nuclear power plants; an event database, which provides the
necessary input to the work with the NPSAG handbook.
The paper will demonstrate the progress made since the initiation of the R-Book
project in 2005. The paper also summarizes the results and insights from a pilot project
to define the content and outline of the proposed handbook. Comments and
recommendations for the R-Book development process were solicited from Nordic
and international experts. Detailed information about technical considerations for how
to derive realistic pipe failure rates from the available service experience data is
documented in SKI Report 2008:01 [1] (January 2008, available from
http://www.ski.se). Work is currently underway to produce the R-Book and a first
edition is scheduled for release to project sponsors during the second half of 2009.
The paper and the presentation at SMiRT 20 will mainly be focused on
presentation of results and insights made when analyzing different process
piping’s and experiences learned from working with the OPDE database.

Reference
1. Olsson, A., Lydell, B. Reliability Data for Piping Components in Nordic Nuclear Power
Plants “R-Book” Project Phase I. SKI Report 2008:01 (January 2008).

301
7. Safety, Reliability, Risk and Margins

Benchmark exercise on risk-informed

in-service inspection methodologies
(7-1841)

Kaisa Simola1, Luca Gandossi2, Alejandro Huerta3

1
VTT Technical Research Centre of Finland, PB 1000, FI-02044 VTT, Finland
e-mail: [email protected]
2
European Commission, JRC-IE, PB 2, NL-1755 ZG Petten, The Netherlands
e-mail: [email protected]
3
OECD Nuclear Energy Agency, 12 Boulevard des Iles,
92130 Issy-les-Moulineaux, France, e-mail: [email protected]

In 2005, the Joint Research Centre of the European Commission (JRC) together
with the Nuclear Energy Agency of the OECD (NEA) launched a project for
benchmarking various risk-informed in-service inspection (RI-ISI) methodologies.
The project, called RISMET, had more than twenty participating organizations
from Europe, U.S., Canada and Japan. The JRC acted as the technical
coordinator of the project, and the NEA provided secretariat support.
The overall objective of the project was to apply, for the first time, various RI-
ISI methodologies to the same case, i.e. selected piping systems in one nuclear
power plant, with the idea of verifying whether they would lead to significantly
different results. Also, a benchmarking exercise would ideally result in the
identification of those phases in a methodology with the greatest potential to
affect the outcome, and might suggest areas for further improvement.
Four systems from the Swedish PWR Ringhals 4 were selected for the
benchmark exercise. The following criteria were used for selecting these
systems: All safety classes should be covered; a variety of degradation mechanisms
should be covered; good coverage of risk categories should be achieved; systems
with a significant increase or decrease in the new inspection program
(before/after applying RI-ISI) should be included; and balance between initiating
and mitigating systems should be ensured. Based on these criteria, the following
systems were suggested by Ringhals and approved by the project team as the
scope of the exercise:
• Reactor coolant system
• Residual heat removal system
• Main steam system
• Condensate system.

302
 7. Safety, Reliability, Risk and Margins

The following approaches to define the ISI program were considered in the
benchmark exercise:
• Swedish regulatory requirements (“SKIFS”)
• PWROG methodology
• PWROG methodology adapted to Swedish regulations (“PWROG Swedish”)
• EPRI methodology
• Code Case N-716, “streamlined RI-ISI”
• ASME Section XI (deterministic).
The application results were evaluated by five groups concentrating on the
following issues: 1) Scope of application; 2) Failure Probability Analyses; 3)
Consequence analyses; 4) Risk ranking, classification and selection of
segments/sites to be included in inspection programs; and 5) Regulatory aspects.
The evaluation included the identification of differences in the RI-ISI applications,
the analysis of the importance of identified differences, and the comparison
between RI-ISI and “traditional” inspection programs.
Even if the scope of the benchmark was limited to four systems, the variety
regarding safety class, potential degradation mechanisms and pipe break
consequences ensured a good coverage of issues for a comparative study. The
risk-informed methodologies showed some significant differences and resulted
in slightly different risk ranking and selection of inspection sites. However, the
results of the benchmark indicated that the risk impact of these differences is
small, and the RI-ISI approaches identify safety important piping segments that
are ignored by approaches not using the probabilistic safety assessment (PSA).
The results of the benchmark exercise RISMET improve the knowledge on
differences in approaches and their impact on plant safety, and promote the use
of risk-informed ISI.
This paper summarizes the results of the RISMET benchmark exercise.

303
7. Safety, Reliability, Risk and Margins

CANDU pressure tube degradation and

probabilistic safety criteria (7-1847)

Alok Mishra∗, Bengt Lydell

Scandpower Risk Management, Inc.
4 Houston Center, 1331 Lamar Street, Suite 1270, Houston, TX 77010, USA
e-mail: [email protected], [email protected]
Michael Kochenhauer
Relcon Scandpower AB, Sundbyberg, Sweden
e-mail: [email protected]

The aging management and plant life extension processes for commercial nuclear
power plants have evolved over several decades. As nuclear power plant age,
degradation of structure, systems and components can be expected to occur. The
understanding of the effect of the age related degradation is important to ensure
safe operation of the nuclear power plant. The wealth of service experience data
from aging research programs in combination with plant-specific data provides a
good foundation for risk-informed applications to assess the potential effects of
aging management on plant safety and reliability. Aging or degradation
phenomena may lead to time-dependant changes in engineering properties that can
impact the ability of plant systems, structures and components (SSCs) to respond
to anticipated and unanticipated challenges during routine and upset plant operations.
Probabilistic Safety Assessment (PSA) is an accepted tool for risk-informed
decision-making. A key consideration of PSA applications is to demonstrate
compliance with regulatory and corporate safety objectives during the whole
lifecycle of a NPP, including the extended lifetime. Demonstration of compliance
with applicable safety objectives is required for the NPPs that are subject to
plant life extension. Incremental changes to core damage frequency (CDF), large
early release frequency (LERF), ΔCDF and ΔLERF provide measures of how
aging could affect plant safety.
The pressure tubes in CANDU reactors are critical components and its failure
has consequences of concern. These cold worked Zr-2.5% Nb pressure tubes
may degrade due to ageing mechanisms like creep and delayed hydride cracking.
It is important to periodically monitor the serviceability of these tubes over the
service life. This paper discusses the estimation of the allowable conditional
failure probability of the pressure tubes for meeting the risk informed regulation
criteria of the US-NRC regulatory guide RG. 1.174.

∗
Corresponding author

304
 7. Safety, Reliability, Risk and Margins

Results and insights from interim seismic

margin assessment of the Advanced
CANDU Reactor (ACR) 1000®
reactor (7-1849)
T. Ha, U. Menon, T. Ramadan, P. Santamaura, M. Elgohary
Atomic Energy of Canada Limited
2251 Speakman drive, Mississauga, Ontario, Canada, L5K 1B2
Tel: 905-823-9060, Fax: 905-403-7386, e-mail: [email protected]

The ACR-1000® reactor developed by Atomic Energy of Canada Limited (AECL)

is a 1200-MWe-class light-water-cooled, heavy-water-moderated pressure-tube
reactor, which has evolved from the well-established CANDU® line of reactors.
It retains the basic, proven, CANDU design features while incorporating
innovations and state-of-art technologies to ensure fully competitive safety,
operation, performance and economics.
The objective of this paper is to describe the seismic margin assessment performed
for the ACR-1000 reactor at full power operation. The design basis earthquake
(DBE) is 0.3g peak ground acceleration (PGA). The seismic margin was assessed,
and potential seismic failure modes as well as weak component links/functionality
leading to severe core damage and widespread fuel damage were indentified.
The Level I internal event at-power PSA models were reviewed and the systems
required to bring a plant from a normal operation to a safe shutdown were identified
in the seismic safe shutdown equipment list (SSEL). In the first approach, seismic
capacities of the items on the SSEL have been developed using the ACR seismic
design criteria and qualification criteria, past seismic experience and recent seismic
probabilistic safety analyses and seismic margin assessments. The plant responses to
seismic events were modelled in seismic event trees, from which the accident
sequences potentially leading to severe core damage and widespread fuel damage
were identified. These accident sequences determined a combination of the failures
of frontline systems. There exist the dependencies between frontline systems and
their support systems, and among support systems. These dependencies were
included appropriately using system dependency matrices. Then the accident-
sequence seismic capacities were estimated from the seismic failures of structures or
components resulting in frontline systems and their support systems in terms of high
confidence of low probability of failure (HCLPF). The plant HCLPF capacities for
severe core damage and widespread fuel damage were then determined.
This assessment shows that the ACR-1000 design can reasonably achieve a
seismic margin in terms of the plant HCLPF that is equal to or exceeding 0.5g
PGA. Therefore, the ACR-1000 design is inherently capable of safe shutdown in
response to a strong magnitude earthquake.

305
7. Safety, Reliability, Risk and Margins

Application of CFD code PHOENICS for

simulating CYCLONE SEPARATORS
(7-1867)

Anu Dutta, B. Gera, Pavan K. Sharma+, R.K. Singh,

A.K. Ghosh, H.S. Kushwaha
Reactor Safety Division, Health Safety and Environment Group
Bhabha Atomic Research Centre, Trombay, Mumbai, India- 400085
+
e-mail: [email protected]

Keywords: separator, CFD, IPSA, PHOENICS, two-phase flow

Cyclone separators are widely used in the field of air pollution control, gas–solid
separation for aerosol sampling and in many industries like power plants, sand
plants etc. In cyclone separators the air flow enters the cyclone through a
tangential inlet, generates a swirling flow that forces entrained particles radially
outward and leaves via an axial outlet pipe at the top of the cyclone. The
rotational fluid motion is generated from the energy obtained from the fluid
pressure gradient. This rotational motion causes the particle to separate relatively
fast due to the strong acting forces. The cyclone separator is very useful
engineering equipment with no moving parts and virtually no maintenance. It
enables particles of micrometers in size to be separated from a gas moving at
about 15 m/s without excessive pressure-drop.
This work presents a computational fluid dynamics (CFD) calculation to
investigate the flow field in a tangential inlet cyclone which is mainly used for
the separation of the moisture from an air stream. Three-dimensional, steady
state Eulerian simulations of the turbulent gas–droplet flow in a cyclone separator
have been performed. Numerical simulation was carried out using CFD code
PHOENICS for the given geometry of separators available in literature. The
IPSA (Inter-Phase-Slip Algorithm) method has been utilized which entails
solving the full Navier-Stokes equations for each phase. The turbulence was
modeled with standard k-ε turbulence model. The liquid droplet was modeled as
particle of size 10 μ and density 1000 kg/m3. The volume fraction of moisture
was 1% at inlet and outlet volume fraction was predicted with CFD. The results
were in good agreement with the reported results. This knowledge can be further
extended for other two phase flow applications in nuclear industry.

306
 7. Safety, Reliability, Risk and Margins

Seismic fragility of a civil engineering

structure (7-1871)

Ajai S. Pisharady, Prabir C. Basu

Civil & structural Engineering Division, Atomic Energy Regulatory Board
Niyamak Bhavan, Anushaktinagar, Mumbai, India
e-mails: [email protected], [email protected]

Introduction
Evaluation of seismic fragility is an integral part of seismic probabilistic safety
assessment. Seismic fragility of civil engineering structures are generally evaluated
by analysis. A nuclear power plant has many safety related civil engineering
structures, with different structural configurations. These configurations vary from a
common framed architecture to complex shear wall type design.
A structure is considered as system, an assemblage of a number of elements.
Seismic fragility for a particular failure mode is determined for each element.
Strength based approach is common for fragility evaluation of element. Weakest
link approach is generally adopted to determine the seismic fragility of the overall
structure, wherein it is assumed that the structure/component fragility is same as
that of the weakest element. Civil engineering structures are complex structures
with high degree of indeterminacy. It may not be rational to apply weakest link
approach for this type of structures. Herein the failure of one element may not
cause global failure of the structure. Seismic fragility evaluation should account
for this consideration.

Aim of the work

Seismic fragility, taking into consideration of global failure of structure, can be
evaluated by means of deflection based approach. It is considered that a building
frame could be designed to a given level of seismic forces – as long as the
building could endure the distortions involved. One approach is to consider
lateral storey drift as structural response parameter. Another approach is to adopt
a displacement based analysis technique like pushover analysis. This analysis
identifies critical elements of the structure which would result in the formation
of a mechanism leading to failure of the structure. The latter approach seems to
be a new one.

307
7. Safety, Reliability, Risk and Margins

In the paper, seismic fragility of civil engineering structure is evaluated adopting

three methods:
Method – 1: Strength based approach,
Method – 2: Deflection based approach considering storey drift as
structural response parameter, and
Method – 3: Based on pushover analysis.
Using an example problem of a representative bay from a RCC framed structure
of a NPP, the paper compares the seismic fragility evaluated adopting the above
three approaches.

Essential results
The paper will describe determination of seismic fragility curves by the three
methods with numerical examples. Comparative study of the results obtained
from three methods will also be presented.

Summary
Evaluation of seismic fragility of civil engineering structures should take
account for consideration of global failure. This can be accomplished by
adopting a global failure parameter as the basis for deriving seismic fragility or
by adopting a displacement based technique like pushover analysis. Out of the
three methods described in the paper, method based on pushover analysis
technique is observed to be the rational approach for determining the seismic
fragility of a civil engineering structure.

References
1. Kennedy, R.P. et al. Probabilistic seismic safety study of an existing nuclear power
plant. NED, 59(1980), pp. 315–338.

2. Kennedy, R.P., Ravindra, M.K. Seismic fragilities for nuclear power plant risk studies.
Nuclear Engineering and Design, 79(1984), pp. 47–68.

3. Basu, P.C. et al. Seismic fragility of a reinforced concrete frame, International Symposium
on Seismic Evaluation of Existing Nuclear Facilities, Vienna, Austria, 25–29
August 2003.

308
 7. Safety, Reliability, Risk and Margins

Thermal-hydraulic analysis for accidents

in OPR1000 and evaluation of uncertainty
for PSA (7-1878)

Tae-jin Kim*, Yun-je Cho, Goon-cherl Park

Seoul National University
Gwanak-599 Gwanak-ro, Gwanak-gu, Seoul, Korea 151-742
e-mail: [email protected]

Probabilistic Safety Assessment (PSA) is a conceptual and mathematical tool to

evaluate numerical estimates of risk for nuclear power plants (NPPs) and can be
used to calculate the probability of damage to the core as a result of sequences of
accidents identified. After the first comprehensive application of the method,
reactor safety study, WASH-1400, PSA has become a standard tool in safety
evaluation of not only NPPs but also industrial installation. When PSA is
performed, thermal hydraulic analysis is necessary to obtain the basic data, from
which system success criteria for construction of event tree and the allowable
outage time for human reliability analysis are determined. Up to now, this
analysis has been undertaken with various system codes such as RELAP,
RETRAN, MELCOR and MAAP4. However, it is well known that deterministic
assumptions and input values have been often applied to the analysis even if
most of them are best-estimate codes. To acquire more realistic result, the
analysis with nominal value and realistic assumptions needs to be carried out
and the uncertainties from the result of the analysis are needed to be essentially
quantified.
The aim of the present study is to develop a best-estimate thermal hydraulic
analysis methodology applicable to PSA as well as able to quantify uncertainty.
In the present study, Optimized Power Reactor 1000MWe (OPR1000), which is
the standard nuclear power plant in Korea, was selected as the objective power
plant. As the frozen code, MARS code was chosen, which is best-estimate code
and has been developed at Korea Atomic Energy Research Institute (KAERI) by
consolidating and restructuring the RELAP5/MOD3.2 code and COBRA-TF
code. Korea Hydro and Nuclear Power (KHNP) Cooperation has already
performed PSA of OPR1000 and made the accident sequence table in which the
accidents have ranked along to the frequency of occurrence. Thus the accidents
in the table as mentioned are required to be analyzed with the input made based
on the realistic assumptions. Moreover the uncertainties from the results of
analyses should be quantified. To do what are aforesaid, Phenomena Identification
and Ranking Table (PIRT) of each accident is necessary since we cannot
consider all parameter for reason of calculation time and cost every analysis.
PIRT for each accident has been already made by the group of expert and is

309
7. Safety, Reliability, Risk and Margins

desirable to be reconstructed if needed. With its own range and distribution, each
candidate parameter in the PIRT was simulated in MARS code. On this
occasion, it was assumed that the range had 95% confidential interval and
acceptable assumption is applied only when the information about the
distribution of parameter does not exist. A number of calculations by MARS
code were performed repetitively with varying the input value of certain
parameter within its uncertainty range. The peak cladding temperatures (PCTs)
from the calculation results, with which it was determined if the core was
damaged, were used to construct the response surface and quantify the
uncertainties. Conventionally, there are several methods to quantify the
uncertainties; Monte Carlo Method (MCM), Latin Hypercube Sampling Method
(LHSM), Response Surface Method (RSM), etc. In the present, MCMs are the
most widely used means for uncertainty analysis. However, after full consideration
of time and cost for the present study, RSM is most suitable to perform the study
since there are more than ten parameters for each accident and it takes too much
time to use MCM to carry out uncertainty analysis. The regression equation for
PCT was obtained by RSM and the randomly sampled values from the range of
each parameter were substituted for the equation. As a result, the distribution of
PCT of each accident was gained and it was used to assess the core damage
frequency (CDF) from the PSA which is already performed.

310
 7. Safety, Reliability, Risk and Margins

Evaluation of the seismic damage index of

structures using fuzzy logic (7-1890)

Adrian Vulpe1, Mircea Ştefanovici2, Claudiu-Răducu Strugariu2

1
Department of Structural Mechanics, Faculty of Civil Engineering,
Technical Univ. of Iaşi, Romania, e-mail: [email protected]
2
Department of Mathematics, Technical Univ. of Iaşi, Romania
e-mails: [email protected], [email protected]

This paper presents an explicit fuzzy logic based method for assessment of
seismic damage of the structures. Uncertainties in earthquake ground motion
parameters and in structural parameters modeling behaviour during earthquake is
expressed linguistically by fuzzification. The number of this parameters and the
choice of the membership functions for the parameters and the damage index are
generally analysed.
The fuzzy rule base for assessment of seismic damage of the structures is
formed for n = 9 input parameters, which take values in the set M = {1,2,3}.
These numerical values are attached to the linguistic values of the parameters by
the level of contribution to damage of structure. To the each system of nine
values of input parameters (x1, x2,...,x9 ) is attached a value of output f in the set
of integers {1, 2, 3, 4, 5}, witch conventionally represent the five levels of
damage, eventually with a percent degree of membership to this level. Then the
response (output) f can take two values ∈ {1, 2, 3, 4, 5} with the same
probabilities pi and qi (equals to the degrees of the membership to two adjacent
classes). If more than two parameters takes, independently, two adjacent ∈ {1,
2, 3} each of them with the complementary probabilities (percents) pi and qi = 1-
pi, then the single output (response) f take more than two values with the
probabilities (percents) equals with the sum of the products of corresponding
probabilities (by the law of alternatives or total probability). A generalised
control rule utilizing weight functions for the input parameters is proposed.
The damage index is estimated finally by applying the centroid defuzzification
method, which express the fuzzy linguistic variable by a crisp value. A program
in MATLAB language for the logical inference is also presented. The proposed
MATLAB program can be easily adapted for other fuzzy parameters or other
number of parameters if their membership functions are defined.
A numerical example verifying this proposed method and a couple of final
remarks close the paper.

311
7. Safety, Reliability, Risk and Margins

References
Cărăuşu, Al., Vulpe, A. 2001. Updating Fuzzy Models for Seismic Risk Assessment.
Transactions SMiRT 16 International Conference. Washington D.C. Paper 1740.
Division M. P. 1–8. August 12–17. 2001.

Cărăuşu, Al., Vulpe, A. 2005. Fuzzy Logic Methods for Seismic Damage Assessment and
Control. Transactions SMiRT 18 International Conference. Beijing. China.
SMiRT18-M03-6. P. 3907–3916. August 7–12. 2005.

Deb, S.K., Kumar, G.S. 2004. Seismic Damage Assessment of Reinforced Concrete Buildings
th
Using Fuzzy Logic. 13 World Conference on Earthquake Engineering. Paper
Nr. 3098. Vancouver. B.C. Canada. August 1–6. 2004.

Klir, G.J., Yuan, B. 1995. Fuzzy Sets and Fuzzy Logic: Theory and Applications. Prentice-
Hall Inc. Upper Saddle River. CA.

Vulpe, A., Cărăuşu, Al. 2008. Fuzzy Logic Models for Seismic Damage Analysis and
Predictions. 14th World Conference on Earthquake Engineering (14 WCEE).
Beijing. China, October 12–17, 2008.

Wilson, C.M.D. 2005. Fuzzy Control of Magneto-Rheological Dampers for Vibration

Reduction of Seismically Excited Structures. Dissertation. FAMU-FSU College
of Engineering. Florida State University.

312
 7. Safety, Reliability, Risk and Margins

A procedure for the computation of seismic

fragility of equipment components
in NPPs (7-1904)

Silvia De Grandis1, Marco Domaneschi2, Federico Perotti2

1
ENEA, Centro Studi Brasimone, Brasimone (BO), Italy
2
Department of Structural Engineering, Politecnico di Milano, Milan, Italy

A numerical procedure for computing the seismic fragility of equipment

components in nuclear power plants is presented, focusing on the effect of the
introduction of a base-isolation system encompassing HDRB devices (High
Damping Rubber Bearings). It is assumed that: 1) no significant interaction exist
between the dynamic behaviours of the building and of the equipment, so that a
two-stage analysis is possible; 2) the dynamic behaviour of the reactor building
can be linearized.
The procedure is based on the use of the Response Surface Method for
modelling the influence of the selected random variables on the building
response; given the system linearity the building performance can be described
in terms of dynamic amplification. More precisely the ratio a/ag of the peak
acceleration at the component supports to the peak ground acceleration is
considered for the non-isolated reactor building.
A Central Composite Design is used as experimental strategy; to account for
random excitation, at each experimental point the dynamic analysis is repeated
for a number of realizations of the seismic input. From the results, the mean and
variance of the extreme value of the dynamic amplification R are computed; two
Response Surfaces are built, modelling both the mean and the variance of R
(“dual RS” approach), which are obviously function of the problem random
variables. The system performance function can be subsequently expressed as
the difference between a given amplification ratio a/ag and the value of R.
Once R is defined, in probabilistic terms, the computation of the probability of
exceeding a given amplification factor Pexc(a/pga) can be obtained, via Monte
Carlo Simulation, for all amplification values in the selected range.
A procedure for refining the RSs is also proposed, based on the computation
of the seismic risk in terms of annual probability of failure for a prototype site
and for a given value af of the support acceleration leading to collapse. By
denoting with pPGA(pga) the PDF of the annual extreme of the PGA, the risk is
equal to the convolution of the conditional probability Pexc(af /pga) and the pdf
pPGA(pga). By investigating the integrand function in the convolution, the PGA
range delivering the largest contributing to the total risk is defined: from this the
amplification range in which the RS’s must be refined is stated.

313
7. Safety, Reliability, Risk and Margins

An example of application is shown regarding the analysis of a preliminary

design of the auxiliary building of within IRIS reactor (∼335 Mwe pressurized
light water reactor) under development by an international consortium which
includes more than 21 partners from 10 countries, led by Westinghouse Electric
Company. Installation in a site characterized by a low-to-average seismicity
level is assumed. Fragility analysis is performed focusing on critical components
inside the reactor vessel; the results show the practical applicability of the
method, leading a reasonable balance between computational effort and degree
of refinement.
When the introduction of the isolation system is considered, resulting in a
drastic reduction of horizontal peak accelerations inside the building, attention is
more focused, in the fragility analysis, on the behaviour of HDRB devices. A
limit value of the relative displacement of the isolators is assumed and the
annual probability of exceeding this value is addressed with the same criteria as
above mentioned. The adopted linearized model of the HDRB devices is
validated by comparison with the results obtained, in the dynamic analysis of the
building, by means of a refined force-displacement hysteretic model taken from
the literature and implemented in the building FE model. The results obtained
are discussed, especially in light of the effectiveness of the HDRB devices and
of the role played by the vertical excitation, which is not affected by the isolation
system.

314
 7. Safety, Reliability, Risk and Margins

Detailed plant seismic walkdown of the

Armenian NPP – Unit 2 (7-1949)

Victor V. Kostarev1, Alexei Berkovski1, Leontiy Chaloyan2,

John Stevenson3, Jan Sedlachek4
1
CKTI-Vibroseism, Russia, e-mail: [email protected]
2
Armenian NPP, Republic Armenia, e-mail: [email protected]
3
Stevenson Engineer, 44125 Midwest Avenue Cleveland OH, US
e-mail: [email protected]
4
S&A, Czech Republic, Plzen, Czech Republic, e-mail: [email protected]

The Armenian Nuclear Regulatory Authority (ANRA) and the Armenian

Nuclear Power Plant (ANPP) requested to the International Atomic Energy
Agency (IAEA) to provide an engineering service to be performed within the
framework of the TC Project ARM/9/014 and in relation to one of the critical
tasks of the seismic safety upgrading programme of the ANPP-Unit 2,
corresponding to the final detailed plant seismic walkdown. This task has been
committed to and performed completely by CKTI-Vibroseism (CVS) team
consisted of S&A, CR and CVS experts with the project review by Dr. John
Stevenson.
The ANPP is located in one of the highest seismic zones with ZPGA level of
0.35 g and seismic capacity of the plant was and still is a key issue of ANPP
restart-up in 1990s and actual operation.
Current project was aimed to:
(a) evaluating the seismic capacity of systems, structures and components
(SSCs),
(b) screening out those SSCs from further consideration because of their
adequate seismic capacity, and
(c) proposing the necessary upgrades (either “easy fixes” or “non-so-easy-
fixes”) as required and appropriate.
Paper will introduce the main findings of ANPP seismic re-evaluation and
upgrades and results of walkdown and SMA CDFM analysis.
In result of investigation of all previous activity and acquired data,
comprehensive walkdown of 2589 items included in the Safe Shutdown
Equipment List (SSEL) and analysis of HCLPF values for the most important
components of the ANPP it was concluded that there is a good reason to believe,
that the problems of ANPP seismic re-evaluation can be successfully resolved
within a limited period of time if advanced analysis and test techniques are used,
and also if necessary administrative solutions will be accepted.

315
7. Safety, Reliability, Risk and Margins

A quantitative method for RI-ISI

assessment (7-1975)

Ilkka Männistö, Otso Cronvall, Kaisa Simola

VTT Technical Research Centre of Finland, P.O. Box 1000, 02044 VTT, Finland
e-mails: [email protected], [email protected], [email protected]

The applications of risk-informed approaches to develop in-service inspection

(ISI) programs are becoming increasingly common in nuclear power plants.
When moving from a deterministic ISI program to a new risk-informed selection
of inspection targets, the safety authorities ask for the assessment of the risk
impact of the change. The risk impact evaluation is especially important in cases
where RI-ISI program is used to justify a reduction in the total number of
inspections.
Assessing risk impact of a RI-ISI inspection program requires three components:
modeling the probability of pipe failures, assessing the consequences of pipe
failures and modeling the effect of inspections in detection of degradation before
failure. This paper presents a method for modeling each of these components and
combining the results for a full quantitative assessment of any RI-ISI program.
The estimation of piping failure probabilities is based on a combination of
probabilistic fracture mechanics (PFM) calculations to simulate crack growth
and a discrete-time Markov process for modelling the inspection activities. The
PFM analyses are performed with a tool based on the deterministic fracture
mechanistic analysis code VTTBESIT, developed by the Fraunhofer-Institut für
Werkstoffmechanik (IWM), Germany and by VTT. VTTBESIT was modified
by adding probabilistic capabilities to the code. In the PFM tool the following
randomised input parameters are used: exponential distribution for initial crack
depth, exponential distribution for initial crack length and Poisson distribution
for thermal load cycle frequency. Monte Carlo simulations are run by sampling
from the probability distributions of the random parameters and using the
respective crack growth (for instance fatigue or stress corrosion cracking)
equation. The simulation ends either when the crack depth reaches the outer pipe
surface, or the time cycles reach the end of plant lifetime, here taken as 60 years.
Each run is then a 60-year simulation with the crack depth calculated at 1-year
intervals, corresponding to the frequency of regular maintenance outages, and
conditional on the existence of an initial flaw. The annual crack depth information
for each simulation is transferred to the second phase of the analyses.
In the second phase, the analyses are based on Markov processes. First, the
PFM simulation results for crack growth are used to construct transition
matrices, where the states of the Markov process correspond to crack penetration
depths in the material, and the transition probabilities from a lower state to
higher states (deeper cracks) model the crack growth. The transition matrix can

316
 7. Safety, Reliability, Risk and Margins

be time dependent to take in to account different PFM calculations as a source

for the transition matrix at different times throughout the plant lifetime. This
makes it possible to have different crack growth characteristics in each, for
example, 10 year interval of the plants lifetime. Inspections are taken into
account with another transition matrix, where the probability of detection (POD)
is a function of the crack size. The effects of inspections are included in the
model as transitions from a crack state to a flawless state.
Results of a standard PSA analysis are used to find the risk consequences of
any pipe failure. When the piping failure probabilities calculated in the Markov
process are weighted with consequences of pipe failure, the full risk impact of
any inspection strategy (selection of inspection targets, ISI intervals, ISI capability)
can be assessed. One significant advantage of the approach is that for analysing
different inspection strategies, the time-consuming PFM calculations need not to
be re-run. Only the Markov analyses, that can be run very fast, are performed.
The model is based on a number of assumptions and limitations, related for
instance to the randomised parameters of the PFM model. The absolute values
are thus subject to large uncertainties. The Markov property assumptions made
for the transition from one flaw depth state to another can be questioned, since
for some degradation mechanisms, it might be justified to assume a “memory”
(the future growth depends not only on the present flaw size, but also on how
this size has been achieved). Further, the assumptions in the inspection process
are highly simplified, assuming that detecting a flaw returns the component to a
flawless state. Despite these simplifying assumption it is believed that the
modelling approach gives reasonably realistic results for the comparison of
alternative inspection strategies and for risk impact assessment.
The developed method and results are illustrated by applying them to a
selected piping system in an existing Finnish NPP.

317
7. Safety, Reliability, Risk and Margins

Evaluation for run-out distance distribution

of rocks falling from slopes (7-1979)

Masato Nakajima1, Hidetaka Nakamura2, Hitoshi Tochigi1

1
Central Research Institute of Electric Power Industry
1646 Abiko, Abiko-shi, Chiba-ken, 270-1194, Japan
e-mails: [email protected], [email protected]
2
Japan Nuclear Energy Safety Organization
Kamiya-cho MT Bldg., 4-3-20, Toranomon, Minato-ku, Tokyo, 105-0001, Japan
e-mail: [email protected]

The Japanese “Examination Guide for Seismic Design of Nuclear Power Reactor
Facilities” was revised in 2006. In the revised guide, it is described that the slope
failure must be recognized as one of the potential phenomena triggered by
earthquake occurrence. Although a large number of studies have been made on
deterministic evaluation of slope stability, only few attempts have been made at
effects evaluation of post failure of slopes due to earthquake ground motions.
For this purpose, it is important to analyze the factors which determine
movements of rocks falling from slopes. The objective of this paper is to specify
the factors which affect the run-out distance of rocks falling from slopes, and
slope dimension, slope inclination and rock size are considered in this analysis.
We adopt two approaches: one is stone falling test using shaking table and the
other is numerical simulation using the analytical method for discontinuous
body.
In order to conduct the test, the concrete slope, on which the steel box
containing stones is attached, is established on the shaking table. The height of
the slope is 1.2 meter. We conduct the following tests:
Test 1: Three hundreds of stones extracted of the same kind rocks are
prepared. The stones are classified into the two groups: the stone size of the first
group is between 20 mm and 40 mm, and that of the second is between 40 mm
and 80 mm. First, every stone is fallen from the slope top (the edge of the box
bottom), individually. Secondly, three hundreds of stones are fallen from the
box, simultaneously. The histograms regarding the stone travel distance from the
slope toe are obtained in a case where we change the slope inclination and stone
size, respectively.
Test 2: By conducting the shaking table tests, the mass of stones falling from
the slope top is measured corresponding to the distance from the slope toe. The
size of the shaking table is 3.0 meter times 3.0 times, and sine wave is input as
ground motions. Slope dimension, inclination, stone size are chosen as
experimental parameters. In the current test, the slope model which side walls
are installed to restrict of the stone movements in the slope traverse direction is
called as the two dimension slope, and the slope model with no side walls is

318
 7. Safety, Reliability, Risk and Margins

called as the three dimension slope. Meanwhile, we perform the numerical

simulations by using two-dimensional Distinct Element Method (DEM) 1) in
order to compare the results from Test 2.
The following results are obtained from the experiments and the numerical
simulations:
(1) The falling stone mass from the slope of 60 degrees inclination is larger
than from the slope of 41 degrees inclination. The maximum distance of
falling stone from the slope of 41 degrees inclination is larger than from
the slope of 60 degrees,
(2) The histogram obtained from the numerical simulation, which represents
the relationship between the distance from slope toe and the mass of
falling stones, is compared with the histogram from the Test 2. The mode,
which is the peak of the histogram, is almost equal and the maximum
distance differs, each other.

Reference
1. Cundall, P.A. A Computer Model for Simulating Progressive, Large-scale Movements
in Blocky Rock Systems. Symposium on rock mechanics, Nancy, Vol. 2,
pp. 129–136, 1971.

319
7. Safety, Reliability, Risk and Margins

Reliability analysis of slope stability at

nuclear power plant site (7-1982)

Weijun Wang1, Carl Costantino2

1
Geotechnical Engineer, US Nuclear Regulatory Commission
Mail Sop T-7E18, Washington, DC 20555-0001, e-mail: [email protected]
2
Engineering Consultant, 4 Rockingham Road, Spring Valley, New York, NY 10977
e-mail: [email protected]

It is well known that great uncertainty and variability exist in subsurface

materials at nuclear power plant site regarding the layer uniformity and soil/rock
engineering properties, as well as seismic loadings to be considered. Those
uncertainties and variability not only affect the stability of the subsurface
materials and foundations, but also affect the stability of the slopes that are
usually present at the site, especially when seismic loadings are involved. This
study focuses on the impact of the uncertainties of soil and seismic loads on the
slopes stability analysis and how to determine the reliability of slope stability
analysis results in term of factor of safety. In this study, uncertainties involved in
the slope stability analysis were first identified, and then a procedure to cope
with those uncertainties in the analysis was developed. After conducting
sensitivity study, the parameters that have greater influence on the slope stability
analysis results are identified. Probability analysis method that considers the
variation of parameters caused by uncertainties and involved in the slope
stability analysis was proposed with the use of reliability index as the reliability
measurement of the factor of safety of slope stability. The proposed procedure
was applied to slope analyses for a specific site using computer software
GeoSlope. The analysis results illustrated that 1. The seismic loadings and soil
internal friction angle parameters affect the slope stability analysis results more
than other parameters involved in the analysis; 2. The proposed reliability
analysis method can give a good indication of the degree of reliability of the
factor of safety of slope stability; and 3. The degree of uncertainty of
engineering properties of subsurface material and seismic loading conditions
affect the reliability of slope stability analysis. Based on the results of this study,
it is concluded that the uncertainties of site subsurface materials and site specific
seismic loadings should be considered in the site stability analyses and their
impacts on the analyses results need to be studied. This paper proposed a method
to evaluate the reliability of site slope stability analysis results in term of factor
of safety and this concept may be applicable to other site safety analyses.

320
 7. Safety, Reliability, Risk and Margins

Seismic damage assessment by

probabilistic seismic demand models
applied to NPP structures (7-1993)

Adrian Vulpe1, Alexandru Carausu 2

1
Faculty of Civil Engrg., Technical Univ. of Iasi, 67 D. Mangeron Blvd, Iasi, Romania
2
Dept. of Mathematics, Technical Univ. of Iasi, 11, Carol I Blvd, Iasi, Romania
e-mails: [email protected], [email protected]

Introduction
This paper extends one of our contributions to the SMiRT 19 Conference, as
well as of other earlier papers of ours [1 – 3] dealing with the probabilistic
modeling of the seismically induced damages in structures, and also with the
seismic fragility models for NPP structures. A couple of contributions to the 13
WCEE Conf. (Vancouver, August 2004) were based on such approaches
involving the probabilistic demand and capacity analysis, or probabilistic
performance-based demand models. Authors like Y.K. Wen, C. Alin Cornell, S.
Janković and B. Stojadinović, J.W. Baker, K. Mackie, a.o. proposed models and
methods going along such approaches for RC frame buildings but also for
nuclear facility structures. Such investigations are based on the (pre)standards
elaborated by FEMA and PEER in several reports starting with FEMA 350 &
356 (2000), PEER 2003/08.

Performance based seismic demand models

The probabilistic estimation of the demand on a structure from a given
earthquake is essentially based on one (or several) intensity measure(s) – IM,
and it consists in evaluating the probability that the earthquake will cause a
certain level of demand in the structure as a function of such IMs. In our paper
[4] we investigated several formal relationships between the engineering demand
parameters (EDPs) and the seismic hazard parameters (like IMs) met in PBSD
models. The conditional probabilities and the total probability theorem are basic
notions that occur in PBSDMs. A typical performance based seismic demand
model involves an equation of the form

λ EDP ( z ) = ∑ P(EDP > z | IM = xi ) ⋅ Δ IM ( xi )

all xi
(1)

321
7. Safety, Reliability, Risk and Margins

where the EDP is compared with the level z, IM is an intensity measure with xi
as one of its possible values and the annual frequency of exceeding it given by
λ IM ( xi ).

PBSD models applied to NPP structures

In the present paper we also investigate some mathematical aspects of the PBSD
models but we discuss some ways to apply them to seismic damage assessment
and to performance based evaluation of NPP structures. Damage indices for
concrete and steel, at FEM level, were presented by four Korean authors in their
contribution [8] to the 13 WCEE, for the EQ damage assessment of NPP
containment structures. We suggest the use of other damage indices besides
those due to Roufaiel & Meyer and Miner. As regards the P-B evaluation of NPP
structures, we investigate how the notion of confidence ratio (approached in [3])
can be used to a higher extent for the risk reduction ratio. This ratio is expressed as
γ Dˆ PH ⎛ k 2 ⎞
λ con = = exp ⎜ − K x β UT + β ⎟ (2)
φC ˆ
⎝ 2 b UT ⎠

where PH is the probability of exceeding a spectral acceleration hazard, D is the

demand, C is the capacity, k is a slope of the log-log plot of PH , β UT is the total
standard deviation due to uncertainty. This expression (2) is based on FEMA
350 report, but it can also be used to express the risk reduction ratio of FEMA
350 in terms of the confidence ratio by

RR = 1 exp ⎛ k 2 β 2 ⎞
⎜ 2 b2 UT ⎟ (3)
k/b
λcon ⎝ ⎠

Conclusions
Providing a method to compute a level of risk reduction implied by FEMA-356
structural evaluation provisions can allow for a proper calibration and rational
use of these provisions for nuclear facility structures. Some ways to follow are
sketched in the conclusions to paper [9] by Orbović et al.

Selected references
1. Carausu, A., Vulpe, A. (1996). Fragility estimation for seismically isolated nuclear
structures by high confidence low probability of failure values and bi-linear
regression. Nuclear Engineering and Design, 160, pp. 287–297.

322
 7. Safety, Reliability, Risk and Margins

2. Vulpe, A., Carausu, A., Vulpe, G.E. (2001). Earthquake induced damage quantification
and damage state evaluation by fragility and vulnerability models. Transactions
SMiRT 16 International Conf., Washington D.C., August 12–17, 2001, Paper
1650, Division M.

3. Vulpe, A., Carausu, A. Assessment of seismic damageability by probabilistic seismic

demand models and confidence ratios Applied to NPP structures. Transactions
SMiRT 19, Toronto, August 2007, Paper 1819.

4. Mackie, K., Stojadinović, B. (2004). Improving Probabilistic Seismic Demand Models

Through Refined Intensity Measures. Proc. 13WCEE, Vancouver, Paper # 1556.

5. Janković, S., Stojadinović, B. (2004). Probabilistic Preformance-Based Sesimic Demand

Model for R/C Frame Buildings.” Proc. 13WCEE, Vancouver, Paper # 1547.

6. FEMA 350 (2000). Recommended Seismic Design Criteria for New Steel Moment-
Frame Buildings. SAC Joint Venture (SEAOC, ATC, CUREE).

7. PEER Report 2003/08. A Technical Framework for Probability-Based Demand and

Capacity Factor Design (DCFD) Seismic Formats. Eds. by F. Jalayer, C.A. Cornell.

8. Cho, H., Koh, H.-M., Hyun, Ch-H., Shin, H.M. (2004). Seismic damage assessment of
nuclear power plant containment structures. Proc. 13WCEE, Vancouver, Paper
# 2972.

9. Orbović, N., Stojadinović, B., Bouchon, M., Wiley, T. (2004). Seismic performance-
based evaluation of nuclear facility structures. Proc. 13WCEE, Vancouver,
Paper # 254.

323
7. Safety, Reliability, Risk and Margins

Safety margins in mechanical integrity

assessments for passive NPP
components (7-2014)

Ardillon Emmanuel1, Meister Eric2, Faidy Claude2

1
EDF-R&D/Industrial Risk Management Department
6, quai Watier, 78401 Chatou, France, e-mail: [email protected]
2
EDF-SEPTEN, 12-14, avenue Dutriévoz, Villeurbanne, France
e-mail: [email protected]

NULIFE (NUclear plant LIFE management) is a European excellence network

comprising nuclear utilities. It was launched in 2007, and aims at promoting
methods to evaluate and manage the lifetime of NPPs, particularly the lifetime of
materials and passive components. In this regard, the issues related to safety and
risk assessment methodologies are an important concern; they become increasingly
significant with the ageing of NPP components and with the industrial objectives
of lifetime extension. They are investigated in the framework of NULIFE expert
group n°4 called “Safety and Risk”.
One particular concern of both regulators and operators is the evolution of
safety margins versus time for the components subject to ageing phenomena: the
requirement of the regulator is to keep the same safety level, and therefore the
same safety margins, during the whole component lifetime, and this requirement
is a prerequisite for any possibility of lifetime extension. Therefore, it is
necessary to identify what safety margins are affected by this requirement.
This paper presents the various meanings of “safety margin” that have been
identified at EDF for the safety margins at the component level. The margins
considered are basically those appearing in the mechanical integrity analysis of
passive components of NPPs.
Firstly, the importance and significance of the use of safety margins is
mentioned. As the first sense of this term denotes the overall safety factor, the
limits of the use of overall safety factors are exhibited on a basic example. The
mechanical integrity analyses of components rely on a mathematical model of
the failure phenomenon, including possible degradation causing structural
ageing. This model characterizes the structural state, which is compared to a
limit (unacceptable) state: this is the failure criterion. Various types of “safety
margins” appear. There is a distinction between implicit and explicit margins.
These two types of margins contribute to the conservatism of the analysis, and
therefore to the structural safety. Explicit margins are the safety factors or
coefficients, that can be an overall safety factor or multiple partial safety factors
applied to several input parameters of the analysis. Implicit margins are

324
 7. Safety, Reliability, Risk and Margins

encompassed in the characteristic values representing the input variables of the

physical model: they are taken as pessimistic values.
Secondly, an understanding of the requirement to keep the same safety
margins is given. It may apply to the safety coefficients. But even these
regulatory coefficients should be considered as a formal reference and other
equivalent sets of factors may be adopted under certain conditions of statistical
knowledge. And the way to perform this equivalence of safety factors is of
course to resort to probabilistic methods, that provide a rational and rigorous
treatment of the uncertainties and explicitly refer to safety (reliability levels).
Finally, what is important is that the safety level has to remain unchanged; and
the appraisal of the safety level should at least include probabilistic highlights.
Finally, a current example of such methodological evolution including
probabilistic considerations is given on the case of the reactor pressure vessel.
RPVs are submitted to a risk of fast fracture due to the following characteristics:
• Existing manufacturing cracks
• Steel embrittlement due to cumulative neutron irradiation,
• Possible Pressurized Thermal Shocks (PTS), occurring under certain
accidental loading conditions.
Two populations of cracks are considered: detected cracks, with known dimensions
and location, and non detected cracks, whose characteristics are not known. The
current deterministic treatment of this population leads to overly conservative
assumptions, that may endanger the objectives of lifetime extension adopted by
EDF. Therefore, an alternative, semi-probabilistic approach, has been developed.
It takes advantage of the better statistical knowledge of the input variables like
fracture toughness, RTNDT shift, fluence and especially the non detected crack
population. In the deterministic analysis, this population is represented by one
single crack (“generic crack”) with the most pessimistic location and dimensions.
Some assumptions are performed about the distribution of these parameters. It is
therefore possible to compare at a given age the RPV reliability of the so-called
generic crack, and the reliability of the population of non detected cracks, which
is of course much better. Consequently, it is also possible to calculate the age at
which the two reliabilities are equivalent. This gives the possible RPV lifetime
extension. This approach is summarized in this paper.

325
7. Safety, Reliability, Risk and Margins

Load factor in case when separating

aleatory uncertainty and epistemic
uncertainty (7-2023)

Takeshi Ugata
Specialized Structural Group, Taisei Corporation
1-25-1, Nishi-Shinjuku, Shinjuku-ku, Tokyo, Japan
e-mail: [email protected]

Aim of the work

The uncertainty exists in the loads acting on a structure. In general, this uncertainty
consists of contains both the aleatory uncertainty and the epistemic uncertainty.
The former is virtually unavoidable and, the latter is caused by the limited ability
and/or the imperfect information and knowledge. Thus, the load factor derived
from the aleatory uncertainty is may be unchangedinvariable, while that derived
from the epistemic uncertainty can be decreased when more reliable model or
method is applied. This paper shows two examples of calculationng of the load
factor when the aleatory uncertainty and the epistemic uncertainty can be are
separated.

Results
When aleatory uncertainty and epistemic uncertainty are separated, load factor is
shown as followings;
a. Case when both load and resistance have log-normal distribution
When both the load and the resistance have log-normal distribution, the load
factor γ is described as γ = γ r ⋅ κ u . Where, γ r is the load factor considering
aleatory uncertainty only, κ u is coefficient for by the epistemic uncertainty,
and then, both are shown as below;
μS ⎧ ⎛ ζ S ,r ⎞ ⎫ ⎡ ⎧ ζ S ,u ⎫ ⎤
γr = exp ⎨α S ⋅ βT ⋅ ζ S ,r ⎜1 − ⎟ ⎬ , κ u = exp ⎢α S ⋅ βT ⋅ ζ S ,u ⎨a − ⎬ ,
Sn ⎩ ⎝ 2 ⎠⎭ ⎣ ⎩ 2 ⎭⎥⎦
respectively.

326
 7. Safety, Reliability, Risk and Margins

b. Case when the uncertainty is described by logic tree

When we evaluate the seismic hazard curves, we often use the logic tree
analysis to cope with the for epistemic uncertainty. The logic tree analysis
gives the fractile hazard. Then, we tried to calculate the load factors from the
fractile hazard. The load factors considering the aleatory uncertainty only is
close to the 50 percentile hazard, while the load factors considering both the
aleatory uncertainty and the epistemic uncertainty is close to about 70
percentile hazard.

Conclusions
There are several problems in separating the epistemic uncertainty from the
aleatory uncertainty. However, if eachboth uncertaintyies can be separated in
load factor, the total amount of the load factor can be decreased by using precise
model and/or advanced analysis method make load factor decrease and. It is very
useful for seismic design of nuclear power plant facilities.

327
7. Safety, Reliability, Risk and Margins

Research associated with the July 2007

NCO earthquake at the Kashiwazaki-Kariwa
nuclear power plant (7-2064)

Greg Hardy1, Robert Kassawara2

1
Simpson, Gumpertz & Heger Inc.
4000 MacArthur Blvd., 7th Floor, Suite 710, Newport Beach, California, 92660, USA
e-mail: [email protected]
2
Electric Power Research Institute
3420 Hillview Avenue, Palo Alto, California, 94304, USA
e-mail: [email protected]

The Tokyo Electric Power Company’s (TEPCO) Kashiwazaki-Kariwa (KK)

plant is the largest nuclear power plant in the world, with a total output of 8,212
MW. The KK plant is 16 kilometers away from the epicenter of the 2007
Magnitude 6.6 Niigataken-Chuetsu-Oki (NCO) offshore earthquake, which took
place at 10:13 a.m. on July 16, 2007. Ground motion recordings at the basemat
of the seven boiling water reactors at the site revealed that the S2 seismic design
level had been significantly exceeded during the event.
Restarting a nuclear power plant following an earthquake that exceeds the
plant seismic design basis entails a number of tasks to verify that damage has
been identified and evaluated and that the plant is in a safe condition to resume
operation. TEPCO has been undertaking these efforts continuously since the
earthquake occurred. EPRI has conducted several research projects in support of
TEPCOs efforts related to technical issues associated with the assessment of the
plant seismic response, the performance of an independent peer review of the
TEPCO seismic walkdown/evaluation program and a review of the seismic
margin for selected KK plant critical components.
Based on EPRI’s independent visual review performed, KK safety-related
structures, systems and components (SSCs) performed very well in response to
the NCO earthquake. No visible damage to the representative safety-related
components reviewed could be detected. This was attributed, among other
factors, to the rugged seismic design practice for the KK plants, particularly for
the supports and anchorage.
Instances of damage were identified associated with some non-safety related
(NSR) SSCs. While the results of these NSR failures and damage may not have
had a critical safety related ramification, some key upgrades have been
recommended which could prevent issues that occurred following the earthquake
related to communications, fire protection, and available services.
The focus of this paper is to summarize the numerous technical efforts that are
being conducted related to the seismic response and the seismic margins

328
 7. Safety, Reliability, Risk and Margins

associated with the Kashiwazaki nuclear plant’s safety related components

which survived the massive earthquake. The engineering community can learn
significant lessons from this recent earthquake’s effects to the world’s largest
nuclear plant, particularly in light of the fact that many nuclear plants (world-
wide) are currently facing increases in the seismic hazard to the plant as well as
being required to assess the seismic risks to the plant using either seismic margin
methods or seismic probabilistic risk methods.

329
7. Safety, Reliability, Risk and Margins

Seismic risk analysis utilizing the PGA and

PGV simultaneously as ground motion
measures (7-2389)

Sei’ichiro Fukushima1, Takayuki Hayashi2

1
Tokio Electric Power Services Co., Ltd., 3-3-3 Higashi-Ueno, Taito-ku,
Tokyo 110-0015, Japan, e-mail: [email protected]
2
Tokio Marine and Nichido Risk Consulting Co., Ltd., 1-2-1 Marunouchi,
Chiyoda-ku, Tokyo 100-0005, Japan, e-mail: [email protected]

Objective
In seismic probability safety analysis, ground motion intensity is usually expressed
by a single index such as peak ground acceleration (PGA), spectral acceleration
for a specified period, or peak ground velocity (PGV). Limiting the number of
indices, however, gives greater uncertainty in the estimation of annual failure
probability that is given by convolving seismic hazard curve and seismic
fragility curve, since information except for ground motion intensity is missed.
Authors proposed seismic hazard analysis utilizing PGA and PGV simultaneously
as ground motion input measures (Fukushima et al. 2007). In this study, seismic
fragility analysis utilizing PGA and PGV is conducted and advantage of vector-
valued risk analysis is illustrated by comparing it with single-valued risk
analysis.

Methodology and results

Seismic fragility of the structure is expressed by the seismic fragility plane that
gives the failure probability under the given combination of PGA and PGV. In
this study, the Monte-Carlo simulation was employed in order to obtain the
conditional failure probability. Random variables were some characteristics that
control the shape of the target response spectrum of input ground motion.
Using generated ground motions, model structure, 7-storey RC building, was
analyzed so that the median and log-normal standard deviation of response for
each combination of PGA and PGV were obtained. Moreover, failure probability
at each PGA-PGV bin was calculated from the stochastic response and capacity.
Relationship between PGA-PGV bin and failure probability was smoothened to
express in the form of fragility plane.
From the fragility planes, it can be seen that both PGA and PGV are adequate
as ground motion measure for limit state of slight damage, since the inclinations

330
 7. Safety, Reliability, Risk and Margins

of hazard plane in both direction are identical. On the contrary, for limit states of
moderate damage, severe damage and collapse, PGV can be better measure than
PGA. This is observed in inclination of fragility plane. The severer the limit state
is, the stronger this tendency is.

Conclusions
In this paper, the seismic fragility analysis using PGA and PGV as ground
motion measure is proposed. In constructing the procedure, the effects of
response spectral shape on the ratio of PGA to PGV is examined so that the
numerous input ground motions for Monte-Carlo simulation can be generated.
By applying the method to model structure that is 7-story RC frame building,
followings are obtained.
By expressing probability characteristic value s as a function of PGA and
PGV, seismic fragility plane can be obtained.
Though both PGA and PGV are adequate as ground motion measure for slight
damage, PGV is preferable for severer damage in which the natural period of
structure gets larger due to inelastic behaviour.

References
Bazzurro, P., Cornell, C.A. 2002. Vector-valued probabilistic seismic hazard analysis
th
(VPSHA). Proceedings 7 U.S. National Conference on Earthquake Engineering.
Boston. MA.

Fukushima, S., Hayashi, T., Yashiro, H. 2007. Seismic hazard analysis based on the joint
probability density function of PGA and PGV. Transaction of 19th SMiRT. Paper
No. M03/1.

Sakai, Y., Yoshioka, S., Koketsu, K., Kabeyasawa, T. 2001. Investigation on indices of
representing destructive power of strong ground motions to estimate damage to
buildings based on the 1999 Chi-Chi earthquake, Taiwan. Journal of structural
and construction engineering. Transaction of AIJ. Pp. 43–50. (In Japanese)

Sato, I., Yashiro, H., Ota, K., Fukushima, S. 2006. Fragility curves for any damage state
based on capacity index. Proc. of 100th Anniversary Earthquake Conference.
CD-ROM.

Sato, Y., Fukushima, S., Yashiro, K. 1995. Study on the index of seismic motion for fragility
analysis. Summaries of technical papers of Annual Meeting AIJ. B-1. Pp. 21–22.
(In Japanese)

Shimomura, T., Takada, T. 2004. Joint pdf of ground motion intensity and duration time based
th
on PSHA. 13 World Conference on Earthquake Engineering. Paper No. 1233.

331
7. Safety, Reliability, Risk and Margins

Seismic isolation of the IRIS NSSS

building (7-2399)
M. Carelli1, M. Ahmed1, A. Maioli1, M. Forni2, A. Poggianti2, F. Bianchi2,
G. Forasassi3, R. Lo Frano3, G. Pugliese3, F. Perotti4,
L. Corradi dell’Acqua4, M. Domaneschi4
1
Westinghouse Electric Company, Pittsburgh (PA), USA
2
ENEA, Bologna, Italy
3
Department of Mechanical, Nuclear and Production Engineering
University of Pisa, Pisa, Italy
4
Department of Structural Engineering, Politecnico di Milano, Milan, Italy

The safety-by-designTM approach adopted for the design of the International

Reactor Innovative and Secure (IRIS) resulted in the elimination by design of
some of the main accident scenarios classically applicable to Pressurized Water
Reactors (PWR) and to the reduction of either consequences or frequency of the
remaining classical at-power accidents initiators. As a result of such strategy the
Core Damage Frequency (CDF) from at-power internal initiating events was
reduced to the 10-8/ry order of magnitude, thus elevating CDF from external
events (seismic above all) to an even more significant contributor than what
currently experienced in the existing PWR fleet.
The same safety-by-designTM approach was then exported from the design of the
IRIS reactor and of its safety systems to the design of the IRIS Nuclear Steam
Supply System (NSSS) building, with the goal of reducing the impact of seismically
induced scenarios. The small footprint of the IRIS NSSS building, which includes
all Engineered Safety Features (ESF), all the emergency heath sink and all the
required support systems makes the idea of seismic isolation of the entire nuclear
island a relatively easy and economically competitive solution. The seismically
isolated IRIS NSSS building dramatically reduces the seismic excitation perceived
by the reactor vessel, the containment structure and all the main IRIS ESF
components, thus virtually eliminating the seismic-induced CDF. This solution is
also contributing to the standardization of the IRIS plant, with a single design
compatible with a variety of sites covering a wide spectrum of seismic hazards.
The conceptual IRIS seismic isolation system is herein presented, along with a
selection of the preliminary seismic analyses confirming the drastic reduction of
the seismic excitation to the IRIS NSSS building. Along with the adoption of the
seismic isolation system, a more refined approach to the computation of the
fragility analysis of the components is also being developed, in order to reduce
the undue conservatism historically affecting seismic analysis. The new fragility
analysis methodology will be particularly focused on the analysis of the isolators
themselves, which will now be the limiting components in the evaluation of the
overall seismic induced CDF.

332
 7. Safety, Reliability, Risk and Margins

Insights gained from the Beznau Seismic

PSA (7-2405)

Martin Richner1, Sener Tinic1, Mayasandra Ravindra2, Robert Campbell2,

Farzin Beigi2, Alejandro Asfura3
1
Nordostschweizerische Kraftwerke AG, Nuclear Power Plant Beznau
CH-5312 Doettingen, Switzerland,
e-mails: [email protected], [email protected]
2
ABS Consulting, 300 Commerce Drive, Suite 200, Irvine, CA 92602, USA
e-mail: [email protected]
3
APA Consulting, 1583 Stratton Circle, Walnut Creek, CA 94598, USA
e-mail: [email protected]

Keywords: seismic PSA, seismic LERF, importance of seismic to risk

PSA studies performed for Light Water Reactors (LWRs) have shown dominating
risk contributions from seismic events. There are several reasons for this finding.
New Probabilistic Seismic Hazard Analyses (PSHAs) calculate a higher seismic
hazard than perceived in the past. In addition, earthquakes represent a common-
mode attack on all safety systems including the containment. On the other side,
refined methods are currently available in the area of seismic PSA that enable
assignment of higher seismic capacities to structures, systems and components.
In this paper, the most recent results and insights gained from the Beznau
Seismic Level 2 PSA study are shown. Beznau nuclear power plant is the oldest
operating pressurized water reactor (PWR) worldwide. The plant was backfitted
extensively during the last two decades by the construction of additional and
seismically more robust safety systems.
The paper first presents the most important characteristics and methods
applied in the actual Beznau Seismic PSA study. The study represents a Level 2
study that fully considers containment integrity and that quantifies the Large
Early Release Frequency (LERF) for seismic events. The paper also shows the
important risk contributors with respect to the Core Damage Frequency (CDF)
as well as with respect to LERF. One main conclusion of the study is that the
seismic capacity of the containment represents a key role with respect to the
seismic Large Early Release Frequency LERF. In addition, the calculated results
indicate that a seismic design of the reactor building of Advanced Light Water
Reactors (ALWRs) of 0.5 g HCLPF may be too low even in areas of low to
moderate seismicity.
Finally, conclusions are drawn with respect to the risk contribution from
seismic events.

333
7. Safety, Reliability, Risk and Margins

Thermal-hydraulic analysis for accidents

in OPR1000 and evaluation of uncertainty
for PSA (7-2478)

Tae-jin Kim*1, Yun-je Cho1, Ho-Gon Lim2, Goon-cherl Park1

1
Seoul National University, Gwanak-599 Gwanak-ro,
Gwanak-gu, Seoul, 151-742, Korea, e-mail: [email protected]
2
Korea Atomic Energy Research Institute,
150 Deokjin-Dong, Yuseong-Gu, Daejeon 305-353, Korea

Probabilistic Safety Assessment (PSA) is a mathematical tool to evaluate numerical

estimates of risk for nuclear power plants (NPPs) and can be used to calculate
the probability of damage to the core as a result of sequences of accidents
identified. After the first comprehensive application of the method, reactor
safety study, WASH-1400, PSA has become a standard tool in safety evaluation
of not only NPPs but also industrial installation. When PSA is performed,
thermal hydraulic analysis is necessary to obtain the basic data, from which
system success criteria for construction of event tree and the allowable time for
human reliability analysis are determined. Up to now, this analysis has been
undertaken with various system codes such as RELAP, RETRAN, MELCOR and
MAAP4. However, it is well known that deterministic assumptions and input
values have been often applied to the analysis even if most of them are best-
estimate codes. To acquire more realistic result, the analysis with nominal value
and realistic assumptions needs to be carried out and the uncertainties from the
result of the analysis are needed to be essentially quantified.
The aim of the present study is to develop a best-estimate thermal hydraulic
analysis methodology applicable to PSA as well as to quantify uncertainty. In
the present study, Optimized Power Reactor 1000 (OPR1000), which is the
standard nuclear power plant in Korea, was selected as the objective power
plant. MARS code was chosen, which is best-estimate code and has been
developed at Korea Atomic Energy Research Institute (KAERI) by consolidating
and restructuring the RELAP5/MOD3.2 code and COBRA-TF code. Korea
Hydro and Nuclear Power (KHNP) Cooperation has already performed PSA of
OPR1000 and made the accident sequence table in which the accidents have
ranked along to the frequency of occurrence. Thus the accidents in the table as
mentioned are required to be analyzed with the input made based on the realistic
assumptions. Moreover the uncertainties from the results of analyses should be
quantified. To do what are aforesaid, Phenomena Identification and Ranking
Table (PIRT) of each accident is necessary since we cannot consider all
parameter for reason of calculation time and cost every analysis. PIRT for each
accident has been already made by the group of expert and is desirable to be

334
 7. Safety, Reliability, Risk and Margins

reconstructed if needed. With its own range and distribution, each candidate
parameter in the PIRT was simulated in MARS code. On this occasion, it was
assumed that the range had 95% confidential interval and acceptable assumption
is applied only when the information about the distribution of parameter does
not exist. A number of calculations by MARS code were performed repetitively
with varying the input value of certain parameter within its uncertainty range.
The peak cladding temperatures (PCTs) from the calculation results, with which
it was determined if the core was damaged, were used to construct the response
surface and quantify the uncertainties. Conventionally, there are several methods
to quantify the uncertainties; Monte Carlo Method (MCM), Latin Hypercube
Sampling Method (LHSM), Response Surface Method (RSM), etc. In the
present, MCMs are the most widely used means for uncertainty analysis.
However, after full consideration of time and cost for the present study, RSM is
most suitable to perform the study since there are more than ten parameters for
each accident and it takes too much time to use MCM to carry out uncertainty
analysis. The regression equation for PCT was obtained by RSM and the
randomly sampled values from the range of each parameter were substituted for
the equation. As a result, the distribution of PCT of each accident was gained
and it was used to assess the core damage frequency (CDF) from the PSA which
is already performed.

335
7. Safety, Reliability, Risk and Margins

Estimation of leak and break frequencies

for probabilistic safety analyses of
piping systems (7-2529)

Rainer Gersinska1, H. Grebner2, Jürgen Sievers2, Leopold Weil1

1
Federal Office for Radiation Protection (BfS), Postfach 10 01 49, D-38201
Salzgitter, Germany, e-mail: [email protected]
2
Gesellschaft für Anlagen- und Reaktorsicherheit, Schwertnergasse 1, D-50667
Köln, Germany, e-mail: [email protected]

In the framework of the BfS-project SR 2608 the details on the estimation of

leak and break frequencies in piping systems contained in the report on methods
and data for the Probabilistic Safety Analysis (PSA) has been updated and
extended by GRS. Based on the hitherto existing regulation new methodical
aspects were introduced. The statistical method based on the evaluation of the
German operational experience for piping systems with different diameters was
updated by the inclusion of structure reliability models based on fracture
mechanics calculation procedures. The example of application of the statistical
estimation method for leak and break frequencies of piping systems with a
diameter of 50 mm out of the volume control system of a German pressurized
water reactor contained in the PSA data volume was updated. To this end the
operational experience considered so far (up to 1995) was extended with respect
to cracks, leaks and breaks in the volume control system of German PWR up to
the year 2006 and the operating time included (191 years of operation) was
accordingly extended to 341 years. Under these conditions new calculations of
leak and break frequencies have been performed and the results have been
compared with the previous values.

336
 7. Safety, Reliability, Risk and Margins

A temperature characteristic diagnosis

algorithm of the abnormal signal simulation
analysis module by using probabilistic
techniques (7-2546)

Kil-Mo Koo1, a Young-Man Song2, Kwang-Il Ahan2, Kil-Nam Oh3

1
Thermal-Hydraulic Safety Research Division, KAERI, 150 Dukjin-dong,
Yuseong, Daejeon, 305-353, Korea, e-mail: [email protected]
2
Synthesis Safety Research Division, KAERI, 150 Dukjin-dong, Yuseong,
Daejeon, 305-353, Korea, e-mail: [email protected], [email protected]
3
Department of Information and Communication,
Gwangju University, 52 Huodek-ro Nam-gu Gwangju, 503-703, Korea
e-mail: [email protected]

The circuit simulation analysis and diagnosis methods are used to assess instruments
in detail when they give apparently abnormal readings. The simulations can be useful
for investigating what the signal and circuit characteristics would look like for a
variety of symptoms that can result from very high temperature environment
conditions. Instrument circuits are first modeled and tested using specific circuit
simulation program. Then degraded temperature conditions are introduced by
modifying the instrument circuit models. The response characteristics of the simulated
instrument circuit to degraded temperature conditions provide the basis for diagnostic
information. The checklists list the steps necessary to obtain information from
instrument loops which may be degraded, but for which the detector should still
be providing valid signals. The role of the circuit simulation is to determine the
diagnostic steps that can be taken to evaluation if the temperature condition is real or a
result of instrument loop degradation. In this paper, a new simulator, ASSA module,
through an analysis of the important circuits modeling under temperature accident
conditions has been designed. It has a special function that means probabilistic
techniques for a diagnosis algorithm of circuit-component degradations including
a temperature characteristic data of the basis for diagnostic method. We present
probabilistic techniques that make synergy use of available process information for
diagnosis and detection of component fault in a circuit-component system. We begin
by describing the motivation for using probabilistic techniques for systems diagnostics
and then define probabilistic expressions that embody the diagnostics knowledge
of interest. We show that a combination algorithm of a Bayesian expression with
the solution to the Chapman-Kolmogoloff equation contains the diagnostic
information of interest while explicitly making use of available process
information obtaining the probability density function corresponding to feasible
circuit-component transitions by an adaptive Kalman filtering.

337
7. Safety, Reliability, Risk and Margins

Fragility functions for seismic performance

assessment of safety-related reinforced
concrete nuclear structures (7-2557)

C. Kerem Gulec1, Andrew S. Whittaker2, John Hooper3

1
PhD Candidate, Civil, Structural and Environmental Engineering Department,
State University of New York at Buffalo, USA, e-mail: [email protected]
2
Faculty of Civil, Structural and Environmental Engineering Department, State
University of New York at Buffalo, USA, e-mail: [email protected]
3
Principal and Director of Earthquake Engineering, Magnusson Klemencic
Associates, USA, e-mail: [email protected]

Squat (shear-critical) reinforced concrete walls are widely used in nuclear power
plants and other safety-related nuclear structures to provide resistance to extreme
earthquake loadings. Performance assessment of such structures utilize fragility
functions that relate the probability of exceeding one or more damage thresholds
to either a ground-motion or response (demand) parameter such as peak ground
acceleration, spectral acceleration at a selected period, story drift or component
plastic deformation.
Fragility functions are developed for squat reinforced concrete walls with
aspect ratio (height-to-length or (hw / lw ) of 2 or less) by review and statistical
evaluation of experimental data in the literature. The experimental data includes
tests of three cross-section types: rectangular, barbell and flanged. Per modern
practice, a demand parameter is used to construct the curves. Experimental
damage data is characterized using damage states and methods of repairs.
Documents that provide guidelines for repair of reinforced concrete walls,
observations from experimental programs, previous research on retrofit of squat
walls and expert opinion are used to identify the most appropriate damage states
and their corresponding methods of repair. Damage states are characterized
generally by direct indicators of damage such as initiation of cracking,
maximum concrete crack width, extent of concrete crushing, sliding shear
displacement, and reinforcement yielding, buckling, and fracture. Each of these
damage states is linked with one of four methods of repair, namely, cosmetic
repair, epoxy injection, partial wall replacement, and wall replacement.
Different families of fragility functions are required for each cross-section
type but the data do not support the development of fragility surfaces to
accommodate axial force, rebar ratio and aspect ratio as input variables. Story
drift is used as the demand parameter. Scopes of repair are provided elsewhere.

338
 7. Safety, Reliability, Risk and Margins

Experience from a seismic probabilistic

safety assessment of a German
PWR (7-2566)

Theodor Bloem, Ralf Obenland

Westinghouse Electric Germany, D-68167 Mannheim
e-mail: [email protected]

Introduction
Nuclear power plants in Germany have been subject to several deterministic
seismic safety assessments. After each safety assessment, components of the
plant were upgraded. Additional deterministic assessments will not essentially
improve the seismic resistance of the plant. As a new demand from authority,
seismic probabilistic safety assessments (seismic PSA) have to be conducted for
German nuclear power plants [BfS-37/05].
In this contribution experience from a seismic PSA of one of the latest erected
German PWR will be presented. Its operation started in the year 1984. A full
scale probabilistic analysis by evaluating safety products for structures, plant
systems and components was performed in order to estimate the core damage
probability caused by earthquakes. Furthermore, limitations and benefits of a
seismic PSA will be discussed.

Method
Stress calculations use the horizontal peak ground acceleration as a measure for
the strength of an earthquake. The objective is to estimate the horizontal peak
ground motion acceleration A for which the seismic response of a component
exceeds the component capacity resulting in its failure. By assumption, A is a
log-normally distributed random variable.
To perform a seismic safety analysis of a nuclear power plant, safety systems,
components and structures needed for plant shutdown and for long term heat
removal must be examined. For components which are accessible for
determining a seismic safety product, fragility curves were evaluated. A fragility
curve provides a conditional frequency of failure in dependence from horizontal
peak ground motion acceleration at the plant site. Fragility curves were
determined for pipes, core internals, buildings, switchgears, heat exchanger
supports, and pump supports. For each component analyzed, fragility curves for
two non-exceedance probability levels Q were evaluated: The fragility curve for
Q = 0.5 provides a frequency of failure determined without any conservatism.

339
7. Safety, Reliability, Risk and Margins

The fragility curve for Q = 0.95 delivers a failure frequency with 95% probability
of non-exceedance. Figure 1 shows fragility curves for a low pressure pipe with
DN 600.
0.054
Frequency of failure →

0.036
Q = 0.95

0.018
HCLPF
Q = 0.5 = 7.8 m/s²

0.000
5 m/s² 6 m/s² 7 m/s² 8 m/s²
Peak horizontal ground motion acceleration A →
Figure 1. Fragility curves for Q = 0.5 and Q = 0.95, low pressure pipe with DN 600.

At non-exceedance probability level Q = 0.5, the frequency of failure by an

earthquake with horizontal peak ground acceleration A = 7.8 m/s² is less than
0.01% for the pipe analyzed. At probability level Q = 0.95, the frequency of
non-failure by this earthquake acceleration is (1 - 0.05) = 0.95. Pipes are very
earthquake resistant, as extensive studies in the USA demonstrated too [NUREG-
4334]. The HCLPF peak ground acceleration is generally considered to be
approximately 95% confidence of less than 5% frequency of failure.
Not for all components a seismic safety product can be determined by
analytical tools. When a component is not accessible to an analytical calculation
of its strength, seismic tests or experience from earthquakes in fossil power
plants and in industrial plants can be applied. Moreover, seismic ruggedness of a
component can be deduced from its use in a vibrational environment. For
assessing electrical equipment and mechanical active components, generic
HCLPF accelerations or GERS (Generic Equipment Ruggedness Spectra) are
used. Generic HCLPF accelerations for different components are given for
instance in [NUREG-4334]. A GERS expresses a level of seismic acceleration to
have a sufficient ruggedness to perform as required [EPRI-5223].
For the seismic PSA an updated seismic hazard curve was used. In addition a
site specific ground response spectrum was used and compared to the original
ground response spectrum from the design phase of the plant.

340
 7. Safety, Reliability, Risk and Margins

Result
A calculated annual probability of core damage of 10-7 emphasizes a good
seismic plant design. This result includes conservative assumptions for limiting
the extent of the analysis. Even at low earthquake intensities a loss of offsite
power was assumed. Additionally an unavailability of non-safety related systems
was postulated. At a strong earthquake building failure is dominant.

Limitations and benefits of seismic probabilistic

safety assessment
Safety factors of buildings and structures turned out to be very difficult to derive
from the calculations at the plant design phase. The fragility curves for buildings
and structures may include some conservatism.
Small bore pipes are not periodically examined by x-ray testing. During an
earthquake there is a risk of pipe breaking starting from an undetected crack. An
earthquake can initiate a small loss of coolant.
Electrical cable connections can oxidize by time. Electrical insulation can
brittle by time. There is a small database for seismic resistivity of aged electrical
components. The probability of failure for aged electrical components was
estimated by using generic Hclpf accelerations. Aged electrical components also
contribute to the seismic risk.
A seismic PSA must consider earthquake intensities lower and higher than the
design base earthquake. A deterministic seismic risk assessment tends to review
only the code compliance of components. The seismic PSA must look beyond
code requirements.

References
[BfS-37/05] Deutsches Bundesamt für Strahlenschutz (BfS), Facharbeitskreis Probabilistische
Sicherheitsanalyse für Kernkraftwerke; Methoden zur Probabilistischen
Sicherheitsanalyse für Kernkraftwerke; Stand August 2005, BfS-SCHR-37/05.

[NUREG-4334] U.S. Nuclear Regulatory Commission (NUREG); An Approach to the Quanti-

fication of Seismic Margins in Nuclear Power Plants; NUREG-CR 4334, UCID-
2044; August 1985.

[EPRI-5223] U.S. Electric Power Research Institute (EPRI); Generic Seismic Ruggedness
of Power Plant Equipment; EPRI NP-5223-SL Revision 1; Final report, August 1991.

341
7. Safety, Reliability, Risk and Margins

342
 8. Issues Related to Operations, Inspection
and Maintenance

Life cycle management including inspection, online monitoring, maintenance,

repair and replacement. Life prediction, reliability, availability, deterministic
and probabilistic integrity risk assessment. Utility vs. regulatory perspectives
on periodic safety reviews, license renewal, safety and economic
issues. Non-destructive examination (NDE) methods and techniques for
structural integrity assessments, reliability and validation of NDE
methodologies, and NDE training. Lessons learned.

343
8. Issues Related to Operations, Inspection and Maintenance

344
 8. Issues Related to Operations, Inspection and Maintenance

Fretting wear resistance nuclear fuel design

& operating experience (8-1619)

Yong Hwan Kim

Korea Nuclear Fuel Co., ltd,
493 DuckJin-Dong YuSeong-Gu Tae-Jeon City Korea
e-mail: [email protected]

The Fretting Wear is emerging issue of PWR Nuclear Fuels in U.S., Europe and
Worldwide. Many nuclear stations still suffered from significant fuel failure
caused by the grid-to-rod fretting wear failures [1]. The grid-to-rod induced fuel
rod fretting failures occurred at various PWR fuel assembly designs. These fretting
wear may be caused by external and internal vibration sources. The extent of the
wear volume and wear depth are depend on grid-to-rod contact configurations.
In this paper it will be proposed the fretting wear resistance nuclear fuel
design with wide contact area as Fig 1. to reduced fretting wear failure even if
same amount of flow vibration sources. And this paper will be presented various
fretting wear test results for suggested wide contact area fuel design to contrast
narrow contact area fuel design it was currently used in PWR in worldwide.
Also this paper present test method and the test apparatus of grid to rod fretting
wear as shown Fig. 2. And will be presented the analysis and test results of wide
contact vs. narrow contact geometry wear depth against load as shown Fig. 3.
And will be presented the wear rate of wide vs. narrow design as shown Fig. 4.
We irradiated 4 Lead Test Assembly (LTA) in commercial PWR reactor to
verify in reactor fuel performance. So in this paper it will be presented the
results of pool side examination (PSE) and post irradiation examination (PIE).
According to irradiation results of PWR coolant activity analysis the proposed
wide contact fuel design experienced very good results of fretting wear performance.

Figure 1. Grid design concept with Wide contact vs. Narrow contact.

345
8. Issues Related to Operations, Inspection and Maintenance

Figure 2. Grid-to-rod fretting wear tester.

Wi d e A

Wi d e B

Nar r o w

Figure 3. Grid-to-rod fretting wear test results.

Wear Rat e K ( × 10 - 13 Pa - 1 ) Mat er i al A Mat er i al B

No r m al 1.721 2.344
Wi de B
Co r r o d ed 2.194 2.124
No r m al 1.868 1.896
Wi de A
Co r r o d ed 2.797 2.688

No r m al 6.437 6.264
Nar r ow A
Co r r o d ed 6.149 6.452

Figure 4. Wear rate of Wide vs. Narrow design.

References
1. M.W. Kennard, D.J. Sunderland, J.E. Harbottle. 1995. A Study of Grid-to-Rod Fretting
Wear in PWR Fuel Assembly. Stoller Report.
2. Y.H. Kim et al. Advanced Spacer Grid Design for the PLUS7 Fuel Assembly (2002
KNS-AESJ Joint Nuclear Fuel Seminar).
3. Y.H. Kim et al. Advanced Spacer Grid Design for the PLUS7 Fuel Assembly (NTHAS3
Third KOREA-JAPAN Symposium on Nuclear Thermal Hydraulics and Safety).
4. Y.H. Kim et al. Fretting wear of fuel rods due to flow-induced vibration (14th International
Conference on SMiRT 1997).
5. Y.H. Kim et al. Development of a Methodology for In-reactor Fuel Rod Supporting
Condition Prediction. J. of Korean Nuclear Society 2004, 28, pp. 17–26.

346
 8. Issues Related to Operations, Inspection and Maintenance

Aging problems and residual life time

evaluation of the WWER-1000 MW
containment shell structure (8-1622)

Dimitar Stefanov
Associate Professor, Bulgarian Academy of Sciences, Central Laboratory for Seismic
Mechanics and Earthquake Engineering, Sofia, Bulgaria
e-mail: [email protected]

Introduction
There are two units of type WWER-1000 MW which are in operation more than
20 years in NPP Kozloduy, Bulgaria. Some specific aging problems appeared
during that time and different technical solutions are applied. It is useful to share
this experience with the engineering community and to discuss the proper
measures for the future exploitation.

Aim of the work

The main goal of this paper is to generalize the specific aging problems and the
residual life time evaluation of the WWER-1000 MW containment shell
reinforced concrete structures of Unit 5 and 6 in Kozloduy NPP, Bulgaria. The
main problems of the containment shell structure are connected with the original
prestressing system. After a comprehensive analysis of the behaviour of the
system for a long time a new prestressing system is proposed and successfully
implemented.

Essential results
First of all the different factors and degradation mechanisms are investigated
following the procedures given in [1]. Several types of in situ and laboratory
tests are performed for specific elements of the civil structures. Based on these
results and on the available technical information an evaluation for the condition
of the reinforced concrete structure is done. The containment shell structure is a
prestressed structure and the condition of the prestressing system is very
important. Some problems with the original system and several new technical
solutions for the new system are discussed. The residual life time evaluation is
done on the base of the complex analysis of the all available information from
the construction time until now.

347
8. Issues Related to Operations, Inspection and Maintenance

Conclusions
Several techniques for investigation of the concrete and steel are recommended.
Prescriptions are given for the periodical inspections of the important parts and
details of the containment shell structures.
Some specific issues are considered for the instrumental monitoring and the
control of the aging mechanisms. Special attention is paid to the monitoring of
the structure – geodetic monitoring, monitoring of the stress and strain state of
the concrete and the monitoring of the prestressing system. A concept is
recommended for the future development and modernization of the monitoring
systems. A proper measures are suggested for reducing the aging effects which
are the basis of the maintenance program for these structures.

Reference
1. NUREG/CR-6424, Report on Aging of Nuclear Power Plant Reinforced Concrete
Structures.

348
 8. Issues Related to Operations, Inspection and Maintenance

Monitoring relative humidity and

temperature for life-time assessment of
sandwich-type concrete structures (8-1647)
Fahim Al-Neshawy, Esko Sistonen, Jukka Piironen, Jari Puttonen
Helsinki University of Technology, Faculty of Engineering and Architecture
Department of Structural Engineering and Building Technology
P.O. Box 2100, FIN-02015 TKK, Finland
e-mail: [email protected]

Introduction
Deterioration of concrete is one of the basic questions in the life time management
of buildings and structures in nuclear power plants. Two of the most important
factors in building deterioration subjected to outdoor conditions are moisture and
temperature. Moisture is a major factor in physical deterioration processes that are
typically caused by restrained moisture movements and freezing or they can be
connected to chemical or biological attacks. Nowadays a severe climate may be a
reason behind a chemical load on building. In addition, moisture will increase the
heat flow through a structure and thus increase the consumption of heating energy.
The continuous monitoring of temperature and relative humidity provides not only
important information for life-time management of sandwich-type concrete
structures but also introduces the possibilities of systematic condition monitoring in
developing the predictive maintenance of power plant facilities.

Methodology
A new thermal and moisture monitoring method was developed at the laboratory of
structural engineering and building physics. The new method does not require large
investment of time and facilities for collecting service life information. The method
was tested in repaired buildings. For that purpose, a network based monitoring system
was established by using modern communication techniques to gather large amounts
of data with little effort. The monitoring network system was found to be useful for
assessing the repaired building façade performance and giving knowledge about the
physical functioning of building envelopes. The system developed and tested is easily
adapted to various types of structures where the parameters traced may also vary.
The development of the thermal and moisture monitoring method included
laboratory work and field measurements. The laboratory work focused on designing
and testing the RHT-monitoring network system (Figure 1) including the calibration
of the temperature and relative humidity devices. The field measurement was carried
out to monitor the temperature and relative humidity of three facades that were
repaired with different methods. The thermal and moisture condition was monitored at
regular intervals of 15 minutes for more than two years.

349
8. Issues Related to Operations, Inspection and Maintenance

Figure 1. Schematic diagram of the RHT monitoring network system.

Essential results
The results of relative humidity and temperature monitoring provide an opportunity
to have a close look at hygrothermal performance of the wall assembly and
assessing the performance of the repaired facades.
The long-term moisture response indicators RHTT1 and RHTT2 indices were
calculated [3]. The RHTT1 index is to examine the potential for biological
growth, RHTT2 index to examine the potential for corrosion.
The freezing thawing index (FT) for examining the potential for frost damage
is defined as the number of cycles when temperatures oscillate between the
freezing and thawing point for the facade components that are almost at the
critical moisture saturation level.

Table 1. Example of the RHTT1, the RHTT2, and the freezing thawing (FT) indices for the
rendering coat and the original outer concrete panel of the repaired façade.

June 2005 to June 2006 to

June 2006 February 2007
Repaired facade
RHTT1

RHTT2

RHTT1

RHTT2
FT

FT

First floor 742 1386 116 431 1049 68

North-east
Rendering Sixth floor 448 883 78 340 856 77
coat First floor 266 515 97 203 577 63
South-west
Sixth floor 422 797 101 225 648 86
First floor 1438 1833 0 0 0 0
North-east
Original outer Sixth floor 264 332 0 0 0 0
concrete panel First floor 239 302 0 0 0 0
South-west
Sixth floor 1165 1475 0 0 0 0

350
 8. Issues Related to Operations, Inspection and Maintenance

According to Mukhopadhyaya (2005), the safe values for RHTT and FT indices
for various building materials are not available yet, but they are useful to
indicate the deterioration potential in different repaired materials and methods.

Conclusion
The RHT monitoring network system is very useful for gathering a large amount
of data about the thermal and moisture performance of repaired facades, which
provides a much better understanding of how the environment and the building
interact and can complement visual inspections. By measuring the temperature
and relative humidity of building components systematically, we can determine
the potential for deterioration, wetting and drying patterns in building
components, and changes in moisture content. Documented monitoring and
analysis are especially important when applied to maintenance and repairing in
safety classified structures such as nuclear power plants. These create a basis to
maintain also the reliability of civil engineering parts of the plants by predictive
maintenance and corrective actions that are carried out in a timely manner.

References
1. Al-Neshawy, F., Sistonen, E., Piironen, J., Huovinen, S. (2007). New Method for
Monitoring the Hygrothermal Condition of Repaired Facades. Proceedings of the
Concrete under severe conditions: Environment and loading (CONSEC’07).
Tours, France, 4.–6. June 2007. Pp. 1807–1812.

2. Al-Neshawy, F. (2007). A Network System for Monitoring the Thermal and Moisture
Performance of Repaired Concrete Facades. Licentiate Thesis, Helsinki
University of Technology, Faculty of Engineering and Architecture, Department
of Structural Engineering and Building Technology, Espoo, Finland. 127 p.

3. Mukhopadhyaya, P., Kumaran, K., Nofal, M., Tariku, F., van Reenen, D. (2005).
Assessment of building retrofit options using hygrothermal analysis tool. The
th
proceedings of the 7 symposium on building physics in the Nordic countries,
Reykjavik, Iceland, 13.–15. June, 2005. Pp. 1139–1146.

351
8. Issues Related to Operations, Inspection and Maintenance

German nuclear power plants utility ageing

management – long term fatigue evaluation
of safety relevant components (8-1652)

Reese, Sven H.1, Seichter, Johannes2

1
E.ON Kernkraft GmbH, Treskowstrasse 5,
30457 Hanover, Germany, e-mail: sven.reese-eon-energie.com
2
Siempelkamp Prüf- und Gutachter-Gesellschaft mbH, Am Lagerplatz 6a,
01099 Dresden, Germany, e-mail: [email protected]

Integral Plant Life Management comprises the technological and administrative

measurements of relevant systems and components during the lifetime of a
Nuclear Power Plant (NPP) to safeguard the safety requirements.
In this context aspects on Ageing Management (AM) cover all safety relevant
Systems Structures and Components (SSCs). In more detail, within the German
understanding a clear differentiation between Plant Life Management (PLIM)
and Ageing Management is done. PLIM refers to the entire NPP (primarily
under utility responsibility) in order to guarantee safe operation and additionally
to increase plant availability while minimizing unplanned outages and planned
downtimes. In contrast to this, AM refers to safety relevant SSCs only and is
done under the supervision of the responsible safety authority. Within AM
operational ageing mechanisms like physical and material ageing phenomena are
covered due to the fact that a sufficient knowledge of stresses and strains are
indispensable for a precise prediction of the lifetime and subsequently for safe
and reliable operation. Suitable and applicable assessment tools like comprehensive
temperature surveillance measures are indispensable.
Based on the demand of a systematic procedure, an explicit classification of
safety and availability significant SSCs into three different groups is feasible.
The differentiation of mechanical, I&C and building components is based on
diverse safety requirements.
Passive components of the reactor coolant pressure boundary account for
highest safety requirements. Especially for these components, thermal fatigue is
one significant long-term degradation mechanism, due to the fact that thermal
loadings lead to most fatigue relevant stresses and strains in piping systems.
Therefore temperature-measuring facilities were applied to various positions
on primary circuit piping where most significant thermal loads were expected.
During recent years of operation these measuring positions are being reviewed
and adapted, leading to comprehensive, representative and global information of
existing thermal loadings.
Based on this extensive amount of measured data a detailed analysis of all
relevant thermal-loading events is feasible.

352
 8. Issues Related to Operations, Inspection and Maintenance

Measured values are evaluated on a per-component-basis with regard to relevant

temperature loads (temperature alterations, thermal gradients, stratifications) and
their corresponding number of occurrences. This way bookkeeping of load
cycles is performed to determine actual fatigue usage factors. In load cycle
counting lists for each registered event a partial fatigue usage factor is stored. To
avoid too much conservatism, those load cycle counting lists are also prepared
on a per-component-basis by analyzing the temperature and stress distributions for
reference transients. Thermal boundary conditions (e.g. heat transfer coefficients) of
reference transients are determined, based on measured temperatures respecting
thermal inertia of the measuring equipment. This procedure guarantees both
adequate realism and necessary conservatism.
At NPP construction time predicted fatigue usage factors were defined based
on calculations of specified transients. So various uncertainties, resulting due to
the fact that precise operation conditions were not known in detail at plant
design time, are part of this predicted end-of-life fatigue usage factors. In
comparison to detailed analysis being based on real measured values, it is
obvious that predictions adapted on specified loads will lead to more
conservative results mostly. But nevertheless, in a small number of cases the
registered events show new effects, not accounted for at design state, so that
analyzing these data leads to substantial progress in component integrity
assessment knowledge. Additionally these technical experiences on measured
thermal loads will give a direct input in optimizing operation mode of relevant
SSCs and subsequently minimizing fatigue relevant stresses.
So the goal of all applied surveillance, measuring, and analysing measures is a
detailed, reliable and exact knowledge of the system’s status. Herein the
surveillance of thermal loads as one important root cause for ageing relevant
material degradation mechanisms is the basis for every further action. A detailed
knowledge of the loading history based on measured values will lead to the
reduction of conservative results. Additionally the assessment of new and
possibly unspecified loading events including a statement of the influence to a
SSC specific fatigue usage factor is feasible.
This publication will give a brief introduction to the basics of Ageing
Management (AM) in German NPPs while clarifying the delineation to Plant
Life Management (PLIM). Surveillance measures on primary circuit piping
components are essential for comprehensive analysis of thermal fatigue being
one significant long-term degradation mechanism. This component specific data
is the basis for further detailed analysis in order to determine a data-based actual
fatigue usage factor and finally to deduct required measures proactively.

353
8. Issues Related to Operations, Inspection and Maintenance

Considerations related to long-term

operation for CANDU 6 NPP (8-1663)

Mihail Cojan1, Corina Mocanu2, Gheorghe Florescu1

1
Institute for Nuclear Research, 115400-Pitesti, Romania
e-mails: [email protected] [email protected]
2
CNE Cernavoda NPP, 905200-Cernavoda, Romania, e-mail: [email protected]

Long term operation (LTO) of a nuclear power plant may be defined as

operation beyond an established timeframe set forth by, for example, licence
term, design, standards, licence, and/or regulations, which has been justified by
safety assessment, with consideration given to life limiting processes and features of
SSCs. Managing physical or material ageing of SSCs important to safety aims to
maintain their design safety margins above SSC specific requirements (see Fig. 1)
and thus to minimize risk to public health, environment and safety, [1].

SSC
Safety
Margin Safety Requirements
The effect of mitigation of
Safety margin with ageing
premature ageing
Safety margin without
mitigation of ageing

SSC Service Life

Figure 1. Relation between SSC safety margin and service life.

Approximately 25% of the NPPs in the world have been operated for more than
30 years, and about 70% for more than 20 years. The first commercial
CANDU® unit Douglas Point was put into commercial operation in 1968 (40
years ago) and the first CANDU 6 units operate for 25 years. There are 11
CANDU 6 units in operation. As in Canada, China and Korea the LTO Program
is already implemented in Romania where Cernavoda NPP Unit 1 was put into
commercial operation on the 2nd of December 1996 and the Cernavoda NPP
Unit 2 was put into commercial operation on the 5th of October 2007. The
following actions have been performed [2]:

354
 8. Issues Related to Operations, Inspection and Maintenance

Fuel channel inspections using Ultrasound (US) & Eddy Current (EC)
detectors were performed for Unit 1 into 1999, 2003 and 2008.
For Steam Generators the periodic inspection program was started in
1998. The tubes (EC), the nozzles (US), internal supports and the walls
(US) were inspected periodically.
Feeder inspection program was started in 2003 for wall thinning
measurement; elbows extrados cracks identification and welds monitoring.
PSR was started in 2008 to be performed for Cernavoda NPP Unit 1,
after 10 years of commercial operation.
The LTO program, proposed to be applied at Cernavoda NPP, is supported both
by the experience of CANDU 6 owners and by the results of research conducted
within INR Pitesti. Institute for Nuclear Research Pitesti (INR) is the main
responsible RTD organization for development of national technical support of
nuclear power in Romania. The activity of the Institute is oriented with priority
towards applied and engineering research within RTD programs, connected to
present and future specific issues of NPP, especially those using CANDU 6 type
reactors.
Figure 2 illustrates a number of mechanisms which have been experienced in
CANDU 6 reactor systems as they are currently perceived in terms of ageing
predictability.

Figure 2. Illustration of modelling ability and condition monitoring ability for typical ageing
mechanisms.

Such estimations indicate where R&D investment is required for predictability

to be improved. It is clear that significant R&D expenditure is needed if
modelling and condition monitoring technology is to achieve the required level
for an effective ageing management based on predictability. This should lead to

355
8. Issues Related to Operations, Inspection and Maintenance

an increase in the use of mitigation measures that, if applied early enough, can
prevent crack initiation and/ or further deterioration. Over the past 7 years, INR
Pitesti has been working on R&D Programs to support a comprehensive and
integrated Cernavoda NPP Life Management Program, [3]. A comprehensive
R&D support to LTO program applicable to CANDU 6 NPP has been proposed.

References
1. Kearney, M. IAEA activities to support Member States for Long Term Operation and
Ageing Management. Part 1 and Part 2. Lectures presented to the IAEA
Regional Workshop on Regulatory Oversight on Ageing Management and Long
Term Operation for NPPs. Mamaia, Romania 23–27 June 2008.

2. Delcea, L. Cernavoda NPP Plant Life Management Program. paper presented to the
IAEA Regional Workshop on Regulatory Oversight on Ageing Management and
Long Term Operation for NPPs. Mamaia, Romania 23–27 June 2008.

3. Cojan, M., Florescu, Gh., Roth, M., Pirvan, I., Lucan, D. R&D Support to CANDU 6
Lifetime Management. paper presented to the IAEA Regional Workshop on
Regulatory Oversight on Ageing Management and Long Term Operation for
NPPs. Mamaia, Romania 23–27 June 2008.

356
 8. Issues Related to Operations, Inspection and Maintenance

Service life management system of

concrete structures in nuclear
power plants (8-1685)

Erkki Vesikari1, Jari Puttonen2, Vesa Hiltunen3, Aki Mattila4

1
VTT Technical Research Centre of Finland, P.O. Box 1000, 02044 VTT, Finland
e-mail: [email protected]
2
Helsinki University of Technology, Department of Structural Engineering and
Building Technology, P.O. Box, 02015 TKK, Finland
e-mail: [email protected]
3
Vesa Hiltunen, Teollisuuden Voima Oy, 27160 Olkiluoto, Finland
e-mail: [email protected]
4
Aki Mattila, Fortum Power and Heat Oy, 07901 Loviisa, Finland
e-mail: [email protected]

The objective of the project is to develop a predictive service life management

system (SLMS) for concrete structures in nuclear power plants. The management
system includes prediction of service life, guarding of safety and serviceability
limits, prediction of maintenance and repair actions, calculation of life cycle
costs and environmental impacts, evaluation of risks and inspection of structures. By
the SLMS the safety, accepted structural performance and uninterrupted service
of concrete structures are ensured during the planned service life of a nuclear
power plant.
The SLMS is connected with the in-service inspection system of NPP. Thus
the observed condition of structures is brought to the process of service life
prediction and decision making on maintenance and repair actions. The system
is provided also with qualitative and quantitative risk analyses, financial and
ecological life cycle analyses and detailed structural condition analyses.
The methodological ground of the service life management system was
developed during the EC FP5 project LIFECON (2001–2003). The structures are
divided into smaller structural parts which can be treated as homogenous with
respect to materials, structural features and environmental conditions. These
structural parts are called “modules” and they serve as basic structural units in
the analysis and planning processes of the system. The structural databases
which serve as initial data sources in the calculation processes are consistent
with the modular breakdown of structures. A specific nuclear feature in
partitioning is that safety classification of structural parts and their impact on PSA
analyses can be considered, which improves the possibilities for concentrating
plant life management actions to structural parts according to their hierarchical
order of importance.

357
8. Issues Related to Operations, Inspection and Maintenance

The core of the management system consists of a combined condition, cost

and environmental impact analysis. The condition analysis is produced based on
degradation models, predefined limit states of condition and the Markov Chain
method. The condition analysis is capable of predicting the probability of the
modules to be at any of the condition states at any year during the treated design
period. The analysis contains also an automatic condition guarding system which
triggers maintenance, repair and rehabilitation (MR&R) actions whenever the
predefined limit state of condition is exceeded with a maximum allowable
probability.
The actual service life management system is supplemented by structural
analyses and risk analyses as not all degradation can be predicted by simple
degradation models. The structural analyses are used to evaluate the cracking
behavior of concrete structures in ultimate and serviceability limit state. The risk
analyses are conducted for corrosion of the steel liner and prestressing tendons
in the containment building.

358
 8. Issues Related to Operations, Inspection and Maintenance

Statistical assessment method for the

optimization of the inspection need for
nuclear steam generators based on
existing inspection data (8-1731)

Kim Wallin1, Rainer Voskamp2, Jan Schmidbauer3,

Henner Ostermeyer4, Gerhard Nagel5
1
Academy of Finland, P.O. Box 1000, FI-02044 VTT, Finland
e-mail: [email protected]
2
Kernkraftwerk Unterweser, P.O. Box 140, D-26932 Stadland
e-mail: [email protected]
3
Kernkraftwerk Grohnde, P.O. Box 1230, D-31857 Emmerthal
e-mail: [email protected]
4
Kernkraftwerk Brokdorf, D-25576 Brokdorf
e-mail: [email protected]
5
EON Kernkraft, P.O. Box 4849, e-mail: [email protected]

The cost of steam generator inspections in nuclear power plants is high.

Therefore, it is important to optimize the inspections, both with respect to timing
as well as coverage. Presently the inspection requirements vary from plant to
plant, without any clear physical or statistical basis. Because each steam
generator is an individual, it is not possible to derive common requirements for
all. However, based on a systematic statistical analysis of existing inspection
data for each steam generator it is possible to optimize the inspection need,
without reducing the reliability of the operation of the steam generator tubes.
A new quantitative assessment methodology for the accumulation of flaws
due to stochastic causes like fretting has been developed for cases where limited
inspection data is available. Additionally, a new quantitative assessment methodology
for the accumulation of environment related flaws, caused e.g. by corrosion in
steam generator tubes, has been developed. The method that combines deterministic
information regarding flaw initiation and growth with stochastic elements connected
to environmental aspects requires only knowledge of the experimental flaw
accumulation history. The method, combining both types of flaw types, provides
a complete description of the flaw accumulation and there are several possible
uses of the method. The method can be used to evaluate the total life expectancy
of the steam generator and simple statistically defined plugging criteria can be
established based on flaw behavior. This way the inspection interval and
inspection coverage can be optimized with respect to allowable flaws and the
method can recognize flaw type subsets requiring more frequent inspection

359
8. Issues Related to Operations, Inspection and Maintenance

intervals. The method can also be used to develop statistically realistic safety
factors accounting for uncertainties in inspection flaw sizing and detection.
Examples of the application of the assessment method are provided for real
steam generators, both in the case of stochastic damage as well as environment
related flaws.

360
 8. Issues Related to Operations, Inspection and Maintenance

Development of RI-ISI at STUK (8-1794)

Ari Julin, Jouko Marttila, Ilkka Niemelä, Rainer Rantala,

Olavi Valkeajärvi, Reino Virolainen
Radiation and Nuclear Safety Authority, STUK, Finland

Introduction/background
Several risk-informed applications have been introduced since the beginning of
1990’s in Finland. At the end of 1990’s, STUK completed a pilot project that
included, for example, testing of RI-ISI methods for Loviisa 1&2 and Olkiluoto
1&2. The licensees provided qualified input data for the pilot study.

Aim of the work

Together with international development of risk-informed methods, regulatory
oversight was gradually targeted towards systems, components and structures
posing significant risk. Pilot projects gave experience and insight for
development of risk-informed regulatory requirements.

Essential results
In 2003, STUK issued updates for regulatory guides YVL 2.8 and 3.8, which set
forth regulatory requirements on risk-informed in-service inspection (RI-ISI)
applications. RI-ISI scope covers all plant systems and safety classes, including
non-nuclear. The scope includes bellows, seals, hoses, small instrument pipes, etc.
STUK requires that RI-ISI applications include expert panel which turned out
valuable already in pilot project.
In 2007, Loviisa finished full-scope RI-ISI study, which included all systems
in the plant. The risk-informed in-service inspection program was introduced for
Loviisa 1, in 2008. Significant differences were implemented compared to old
inspection program. TVO will finish RI-ISI project for Olkiluoto 1, in 2009.
Olkiluoto 3 will be the first NPP unit implementing risk-informed pre-service
inspection program. No detailed international standards are yet available for pre-
service inspection of NPPs.
Methods used in Finland for RI-ISI applications follow ASME Code, Section
XI Appendix R. Also viewpoints of ENIQ Report nr. 23, “European Framework
Document for Risk-informed In-service Inspection” are included.
According to STUK experience, risk related to change of operating state may
be significant and has to be taken into account if pipe rupture causes shutdown
or if repair of the rupture requires shutdown. In these cases the plant will shut

361
8. Issues Related to Operations, Inspection and Maintenance

down with impaired systems. The full spectrum of operating states has to be taken
into account. In Loviisa unit 1, there are pipe segments where the significant risk
comes from rupture during shutdown states.

Summary/conclusions
In Finland, oversight and operation of nuclear facilities is becoming increasingly
risk-informed. As part of this, plant-wide risk-informed inspection programs are
used in Finnish nuclear power plants. Full-scope RI-ISI programs require that
the underlying PRA covers all operating states and initiating events and includes
the risks of possible changes between operating states with impaired systems.

References
1. STUK YVL Guides 2.8 and 3.8.

2. ASME Code, Section XI Appendix R.

3. ENIQ Report nr. 23. European Framework Document for Risk-informed In-service
Inspection.

4. Development of risk-informed oversight at STUK, final report of pilot projects, 2000.

(In Finnish).

5. Hietanen, O., Jänkälä, K., Bergroth, N., Paussu, R., Nikula, V. Risk-Informed Methodology
of New ISI Program for Unit 1 of Loviisa NPP, 2007.

362
 8. Issues Related to Operations, Inspection and Maintenance

Monitoring pipe thinning using two

accelerometers (8-1801)

Young-Chul Choi, Doo-Byung Yoon, Jin-Ho Park

Korea Atomic Energy Research Institute (KAERI)
1045 Daedeok-daero, Yuseong-gu, Daejeon, 305-353, Republic of Korea
e-mail: [email protected]

There are many pipes in secondary cooling system of nuclear power plants and
coal-fired power plants. In these pipes, high pressure and high temperature fluids
are moving at very high velocity, and it causes pipe thinning through flow
accelerated corrosion (FAC). In 2004, it was reported that because of pipe
thinning, there was a leakage of coolants in Mihama nuclear power plant in
Japan, and several men were killed. As we can see in this case, pipe thinning in
power plants not only does financial and time damage, but also causes damage
of people’s lives. So it is very important to monitor and supervise pipe thinning.
As of now, the most widely used monitoring method uses ultrasonic waves to
estimate the thickness of pipe wall. This method measures the thickness of
dozens of check points in pipes by ultrasonic-type sensor one by one, and
estimates the degree of pipe thinning. So, as the number of check points in pipes
increases, it requires more and more time and manpower to install the sensors.
Furthermore, if the pipes are surrounded by heat insulator, it has to be removed
before the sensor is installed, and this causes additional expenses. And in case
the number of pipes under monitoring be too high, it is impossible to judge the
degree of pipe thinning quickly, so this method’s applicability falters. Therefore
a method to inspect a large area of piping systems quickly and accurately is
needed. In this paper, we proposed the method for monitoring pipe thinning by
using two accelerometers. Our basic idea come from that a group velocity of
impact wave is different as wall thickness. If we install two vibration sensors
outside of the pipes and measures traveling velocity of flexural waves regularly,
we can estimate and monitor the degree of pipe thinning quickly. To obtain the
group velocity, time-frequency analysis is used. This is because an arrival time
difference can be measured easily in time-frequency domain rather than time
domain. In order to test the method we experimented with pipes, and get the
result that group velocity varies according to the degree of pipe thinning. It
verified this method can be used to monitor the pipe thinning.

363
8. Issues Related to Operations, Inspection and Maintenance

Activities of OECD/NEA in the fields of

integrity and ageing of components
and structures (8-1804)

Andrei Blahoianu1, Alejandro Huerta2

1
Canadian Nuclear Safety Commission, Ottawa, ON
e-mail: [email protected]
2
Organization for Economic Cooperation and Development –
Nuclear Energy Agency, Nuclear Safety Division, Paris, France
e-mail: [email protected]

The Integrity and Ageing of Components and Structures Working Group (IAGE)
of the Organisation for Economic Cooperation and Development (OECD)/ Nuclear
Energy Agency (NEA) was established, under the Committee on the Safety of
Nuclear Installations (CSNI), to advance the current understanding of those
aspects relevant to ensuring the integrity of structures, systems and components,
to provide for guidance in choosing the optimal ways of dealing with challenges
to the integrity of operating as well as new nuclear power plants, and to make
use of an integrated approach to design, safety and plant life management.
The working group operates through three subgroups dealing with a) integrity
and ageing of metal structures and components, b) integrity and ageing of
concrete structures and c) seismic behaviour of components and structures.
The group operates through annual plenary meetings and technical workshops
and by issuing state-of-the-art reports and topical opinion papers. Among other
items, the recent and planned activities of the group include the following:
– updating of the IAGE integrated plan
– conducting a meeting of specialists on seismic hazard assessment in
April, 2008 in Lyon, France, with planned publication of the proceedings
– conducting a specialist meeting on risk informed piping integrity
management in June, 2008, Madrid, Spain, with the planned publication
of the proceedings
– publishing reports on: 1) summarising the main findings and conclusions of
a series of OECD/NEA workshops and extracting the seismic information
most relevant to current nuclear practices; 2) A Decade of CSNI Activities
in the Area of Ageing of Nuclear Power Plant Concrete Structures
– discussing the worldwide implications on nuclear facilities of the July
16, 2007 Niigata-ken Chuestu-oki earthquake and its effects at the
Kashiwazaki-Kariwa Nuclear power Station,

364
 8. Issues Related to Operations, Inspection and Maintenance

– supporting a benchmark, SMART 2008, being conducted in Saclay,

France, on seismic design and assessment analysis for multi-story
reinforced concrete buildings subjected to torsion and nonlinear effects
– supporting the IAEA extra-budgetary programme on seismic safety of
existing NPP’s
– conducting a specialist meeting on ageing management of thick walled
concrete structures in October 2008, Prague, Czech Republic, with the
planned publication of the proceedings
– improving robustness assessment methodologies for structures impacted
by missiles
– thorough exchange of information on PTS Rules /Fitness for service
criteria on different member countries for LTO of RPV; fatigue; plant
ageing; LB LOCA redefinition / LBB break exclusion for operating and
new plants
– joint IAEA/NEA catalogue on nuclear facilities that have experience an
earthquake.
This paper will detail some of the recent activities and products of the IAGE
group with special emphasis on the metal and concrete activities, since another
SMIRT 20 paper will detail the activities of the seismic group.

365
8. Issues Related to Operations, Inspection and Maintenance

Effects of concrete creep and shrinkage on

the stress conditions of a post-tensioned
containment structure for steam generator
replacement project (8-1812)

Sungjin Bae1, Luis Moreschi2

1
Civil/Structural Engineer, Bechtel Power Corporation
5275 Westview Drive, Frederick, MD 21703, USA
e-mail: [email protected]
2
Senior Civil/Structural Engineer, Bechtel Power Corporation
5275 Westview Drive, Frederick, MD 21703, USA
e-mail: [email protected]

Introduction
Over the last 20 years, there has been an increasing number of Steam Generator
Replacement (SGR) projects in the United States and worldwide. In most cases,
SGR projects involve the creation of a temporary construction opening in the
containment structure to facilitate the movement of the old/new steam generators
out of/into the containment structure. Prior to the concrete removal, the post-
tensioning tendons passing through the planned opening will be detensioned and
removed. Additionally, vertical and hoop tendons in the immediate vicinity of
the opening will be detensioned to minimize the prestress level within the
opening. After completion of the steam generator replacement operations, the
construction opening will be restored by placing new concrete. Once the new
concrete achieves its target design strength, the removed/replaced and
detensioned tendons will be retensioned.
Ideally, the prestress levels in the containment wall will be restored to its design
basis condition prior to the SGR outage. However, the state of compressive
stresses will not be fully recovered due to the following conditions:
1) Redistribution of dead and remaining prestress load (after removal/
detensioning of tendons) from the opening area to the surrounding
concrete due to concrete removal.
2) Redistribution of prestress load after the concrete is restored and tendons
are retensioned due to the difference in creep and shrinkage values of the
existing and replacement concrete.
Structural analyses are required to account for both effects above and verify that
the interim and restored configurations of the containment structure will continue
to meet the plant’s licensing commitments (e.g., accident pressure/temperature,

366
 8. Issues Related to Operations, Inspection and Maintenance

safe shutdown earthquake loads, etc.). The problem of redistribution of stresses

resulting from the creation of an opening in a prestressed containment wall is
well understood, and References 3 and 4 provide details on how to properly
capture such effects. On the other hand, incorporation of the time-dependent
effects of creep and shrinkage in the new concrete and potential transfer of
prestress load from the new to the old concrete becomes more involved and the
analyst usually relies on simplified models and assumptions to include this
complex phenomenon.

Aim of the work

The purpose of this paper is to investigate the effects of concrete creep and
shrinkage on the prestress load of a post-tensioned containment structure with
two dissimilar materials (i.e., existing and new concrete for the restored area).
ANSYS general finite element program (Reference 2) is used to analyze the
containment structure during the various construction stages. The concrete creep
behavior is implemented using ANSYS defined material creep models and step-
by-step nonlinear calculations are performed to follow the evolution of the
prestress load. Shrinkage behavior is introduced in the model as temperature
increments using the ANSYS time-dependent material table feature. Based on the
results of numerical simulations, a simplified methodology is developed for
general design/analysis purposes. This approach is based on the recommendations
provided in Reference 1 combined with the principle of superposition presented
in Reference 5. The validity of the proposed simplified method is examined by
comparing stress conditions of a containment structure during/after SGR project
with the results obtained from the nonlinear finite element analysis.

Summary/conclusions
The analysis results show that the difference in creep and shrinkage values of the
existing and restored concrete has an impact on the redistribution of prestress
load and needs to be properly accounted when performing structural adequacy
evaluation of a post-tensioned containment structure for SGR project. Parametric
analyses are performed and results presented highlighting the influence of the
various variables. A simplified methodology is developed that accurately estimates
the transfer of prestress load in the vicinity of the restored opening, and provides
the practitioner a simple tool to account for this complex phenomenon.

367
8. Issues Related to Operations, Inspection and Maintenance

References
1. ACI 209R. Prediction of Creep. Shrinkage, and Temperature Effects in Concrete Structures,
1997.

2. ANSYS, Inc. ANSYS Verification Manual. Version 10.

3. Barret, P.R., Woodbury, M., Fisher, D.B. Structural Analysis of Post-Tensioned Concrete
Containment Building Repair Using 3-D Finite Elements. Proceeding of the
th
ASCE 17 Analysis and Computation Specialty Conference, Saint Louis, MI,
2006.

4. Moreschi, L.M., Farzam, F., Nixon, T. Containment Analysis of PCCV for a Temporary
Opening for Replacement of Large Components. Proceedings of the 19th
International Conference on Structural Mechanics in Reactor Technology
(SMiRT), Toronto, Canada, August 2007.

5. Concrete Society Technical Paper No. 101. The Creep of Structural Concrete. Report
of Working Party of the Material Technology Divisional Committee, London,
January 1973.

368
 8. Issues Related to Operations, Inspection and Maintenance

Microbially influenced corrosion in cooling

water systems – development of a new
protection concept for system components
conveying brackish water (8-1815)

Erika Nowak1, Simone Bartels2, Tobias Richter 2

1
EON Kernkraft GmbH
D-30048 Hannover, P.O. Box 4849, e-mail: [email protected]
2
Kernkraftwerk Brokdorf
D-25576 Brokdorf, e-mails: [email protected],
[email protected]

Background and aim of the work

In recent years, corrosive findings ascribed to Microbially Influenced Corrosion
(MIC), have been increasingly observed on cooling water systems in Northern
German Nuclear Power Plants. Despite counter measures being applied in the
form of the application of more corrosion-resistant materials, there is however,
an obvious increase in the amount of corrosion detected. The protection of
components by selected materials is thus to be considered effective only to a
limited extent so far. [1–10]

Research programme and results

In order to improve the understanding of causal connections regarding the
occurrence and progression, corrosion experiments were carried out at NPP
Brokdorf.
By means of this research program (field tests), high-alloyed materials with
different Pitting Resistance Equivalent Numbers (PREN), various surfaces, and
various coatings were evaluated with respect to their corrosion behaviour in
natural brackish water.
Within the scope of this research it has been possible to identify the actual
causes for the damage on pumps in contact with brackish water. This is
especially the case in combination with the colonization by micro-organisms,
resulting crevice conditions beneath the biofilms and an attack takes place on
especially vulnerable areas on the metal surfaces by chlorides from the brackish
water. The research program at the Brokdorf NPP showed that the use of
materials with high nominal pitting resistances as a countermeasure against such
damage is not suitable. Particularly with regard to larger components, it is not

369
8. Issues Related to Operations, Inspection and Maintenance

possible to ensure during production that the actual alloy composition meets the
requirement in each area of the component.

Conclusions
Subsequent material evaluations, in combination with other measures, provided
a new standard of knowledge for the development of a protection concept for
components conveying brackish water.
Components can be protected in a reliable way with sufficiently dimensioned,
impervious and undamaged coatings. It was also proven that a cathodic
protection will protect the components especially well against corrosive attacks
when exposed to brackish water.

References
1. Böhm, D. Werkstoffe in Schiffsrohrleitungen, VEB Verlag Technik, Berlin 1986.

2. Costerton et al. Courtesy of Center for Biofilm Engineering, Montana State University,
Bozman, 1994.

3. Feron, D., Molica, F., A. Biochemical synthetic seawater for crevice corrosion tests.
Simulation of stainless steel behavior in natural marine environment; Stainless
Steel World Oct. 2005.

4. Gümpel, P., Arlt, N., Schiller, D., Moos, O. Mikrobielle Beeinflussung des elektro-
chemischen Potentials nichtrostender Stähle; 3-Länder-Korrosionstagung Wien
2005.

5. Heitz, E., Flemming, H.-C., Sand, W. Microbially Influenced Corrosion of Materials;

Springer Verlag, Berlin 1996.

6. Linhardt, P. Microbially Influenced Corrosion of Stainless Steel by Manganese

Oxidizing Microorganisms. Materials and Corrosion 2004, 55, No. 3.

7. Mattila, K. Biofilms on Stainless Steels Exposed to Process Water. University of

Helsinki 2002

8. Mori, G., Bauerfeind, D. Pitting and Crevice Corrosion of Super austenitic Stainless
Steel. Materials and Corrosion 2004, 55, No. 3.

9. Nowak, E., Kilian, R. MIC an brackwasserberührten Komponenten von KKWs –

Tideabhängig? 4. Tagung Korrosionsschutz in der maritimen Technik 19.–20.
Januar 2005 Hamburg.

10. Sand, W. Mikrobielle Materialzerstörung; Korrosion und Korrosionsschutz, Band 1;

Einführung und wissenschaftliche Grundlagen; Berlin 2001.

370
 8. Issues Related to Operations, Inspection and Maintenance

PAMS – piping and component analysis and

monitoring system application and
visualisation (8-1835)

Paul Smeekes
Teollisuuden Voima Oy, FI-27160 Olkiluoto, Finland

Introduction/background
The PAMS system contains up to date information enabling analysis, studies and
monitoring of an operating plant. Although developed being an “as build”
system it is even used in the design phase and for parameter studies.
The system consists of separate and stand-alone programs-modules and inter-
related Microsoft® Access database tables. All modules can be independently
updated and used for their own purpose as well as used together.
When “ready” the system will consist of the following modules:
ƒ A document database containing documents associated with the
information contained in the database.
ƒ Several interconnected databases containing information like geometry,
boundary conditions, materials and material properties, loading, pipe
contents, material chemical composition, detected cracks etc.
ƒ Pre-processing, editing, visualization and animation modules for the
above mentioned items inclusive several visual and logical checks on the
soundness and validity of the data. One of the latest achievements is the
animation of transient loads.
ƒ Interface modules to make indata files for application programs and to
read data back from the result files into the result database.
ƒ Tailor-made analysis modules to perform post processing of previously
obtained results, like fatigue monitoring, crack growth monitoring,
optimization of inspection intervals (RIISI) etc.
One of the main rules is that all data is accompanied by the associated source
references. Typically these are material standards, isometric & detail drawings
etc. In the future also a validity status will be added as well as the period that the
pipe or component is installed in the plant. If the reference report of a load is
invalid then the system will “know” that the subsequent analyses and results are
not valid. This is very important as subsequent analysis, like fatigue and fracture
mechanical analysis that are also performed directly from the same database
shall use up-to-date input.

371
8. Issues Related to Operations, Inspection and Maintenance

At present the complete geometry, materials and loads of the main piping
systems in two nuclear power plants have been entered into the system.

Aim of the work

The aim of the work is to perform analyses using indata files that are generated
from the databases. Commercially available analysis programs will be used as
much as possible (FEM, CFD, fatigue, crack growth etc.). For special purposes
programs are developed separately.
A special and unique feature is that the source reference list is automatically
incorporate into the indata files and input data refers to the associated source
references.
As results are read back into the database, they can be used for documentation or
subsequent analysis. Later on visualization will be added to the result database. Then
it will be possible to visualize results obtained by application programs.

Essential results
Using the animation module the system has been used to visually check the
transient piping loading applied in a large project. Indata was made for stand-
alone piping analysis and combined piping and building analysis and analysis of
geometrical details. A large RI-ISI project was performed with a purpose made
integrated software module. Furthermore, as all information in the databases is
accompanied by its source reference the system is used as a well organized
archive where documentation associated to piping systems can be found through
the piping system model itself.

Summary/conclusions
From the development stage, the PAMS system has now been taken into productive
use with both commercial software and special purpose tailor made programs.
As the same data is used over and over for different projects the reliability of
this data increases continuously. Associated source references are automatically
integrated into indata files thus reducing the need for separate documentation.
A high reliability of the information in the system is achieved through
automatic data soundness check, reduced input possibilities through pull-down
menus and different possibilities to do alphanumerical and/or visual checks of
the information entered in the system.
The use of the system will definitely increase when other parties get acquainted
to it.

372
 8. Issues Related to Operations, Inspection and Maintenance

References
1. P. Smeekes. The TVO piping and component analysis and monitoring system (PAMS).
SMiRT 18 & 19, 2005 & 2007.

2. Raiko, H., Lipponen, A., Smeekes, P., Talja, H. Load-Case, and Combination Database.
SMiRT 16 Paper 1869, 2001.

3. Smeekes, P., Lipponen, A., Raiko, H., Talja, H. The TVO Pipeline Analysis and Monitoring
System. SMiRT 16 & 17, 2001 & 2003.

4. FPIPE, a finite element method (FEM) based piping analysis program developed at
FEMdata Oy, Finland.

5. VTT BESIT 1.0 by VTT Manufacturing Technology, Finland.

6. ASME Boiler and pressure vessel code, Section III, Nuclear Power Plant Components,
Division 1, Subsection NB, Class 1 Components.

7. Microsoft® Access 2000, Relational Database Management System for Windows.

8. Smeekes, P., Alhainen, J., Lipponen, A., Talja, H. The TVO/VTT material database,
SMiRT 17, 2003.

373
8. Issues Related to Operations, Inspection and Maintenance

Update on Canadian regulatory oversight

of ageing management for nuclear
power plants (8-1842)

Ken Kirkhope, Andrei Blahoianu, Gerry Frappier

Canadian Nuclear Safety Commission, Ottawa, Canada
www.nuclearsafety.gc.ca

This paper provides an update on Canadian Nuclear Safety Commission (CNSC)

staff perspectives on managing the safety aspects of ageing of structures,
systems, and components (SSC) of nuclear power plants (NPP). Managing the
safety aspects of NPP ageing requires a proactive, systematic, and integrated
ageing management approach for the coordination of all activities relating to the
understanding, control, monitoring, and mitigation of ageing degradation of SSC
through the lifecycle of a NPP. A CNSC regulatory document on ageing
management based on modern international guidelines is described. In addition,
the development of CNSC staff review guidelines for New Builds is included.
Finally, CNSC participation in a number of ageing-management and structural
integrity initiatives with industry and other national regulatory agencies both,
within Canada and at international level are presented.

374
 8. Issues Related to Operations, Inspection and Maintenance

Performance surveillance of Gentilly-1

reactor building GFRP repair using fiber
optic sensors and strain gauges (8-1848)

Julia Tcherner (nee Milman)1, Tarek S. Aziz1, Daman K. Panesar2,

Marc Demers3, Kenneth W. Neale3
1
Engineering and Technical Delivery, Atomic Energy of Canada Limited
Mississauga, Canada, e-mails: [email protected], [email protected].
2
University of Toronto, Civil Engineering Department, Toronto, Canada
e-mail: [email protected]
3
University of Sherbrooke, Department of Civil Engineering, Sherbrooke, Canada
e-mails: [email protected]; [email protected]

At approximately 30 years old, Gentilly-1, a permanently shutdown CANDU™

250 MWe Nuclear Power Plant that is currently in the Storage With Surveillance
(SWS) phase underwent repair of the prestressed containment concrete ring
beam. The fill concrete that protects anchorages of the prestressing cables had
deteriorated primarily due to Alkali Aggregate Reaction (AAR) and a repair
program was carried out. The repair was conducted in order to protect the
prestressing anchorages from corrosion and to ensure the continued integrity of
the structure during the time required for decay of radioactive material before
final decommissioning.
The repair strategy involved removal of all deteriorated concrete in zones
categorized as either shallow (<50 mm) or deep (>50 mm). The voids were
filled with repair mortar or repair concrete and then covered with sheets of Glass
Fiber Reinforced Polymer (GFRP). This was the first field application of GFRP
material on a CANDU1 reactor containment structure. In order to meet SWS
phase requirements of maintaining the structure, and to assess the long term
performance of the first GFRP repaired CANDU structure, a performance
surveillance program was implemented.
Instrumentation was embedded both within the concrete repair material and
also within the GFRP sheets. The program included the usage of Fiber Optic
Sensors (FOS), Vibrating Wire Strain Gauges (VWSG) and thermocouples.
Fifteen VWSGs were embedded in the repaired concrete. Measurements
captured strains induced by early age concrete hydration processes and also
hardened concrete strains and temperature effects. To compensate for temperature
effect on the gauge body and to estimate concrete strain due to temperature

1
CANDU is a trademark of Atomic Energy of Canada Limited.

375
8. Issues Related to Operations, Inspection and Maintenance

variations, temperature is measured by the thermistors provided within the body

of the VWSG.
Twelve Fabry-Perot FOS were installed. Eight of them were installed on the
GFRP sheets with two layers of protection, while the other four sensors were
installed in smart patches made of strips of Carbon Fiber Reinforced Polymer
(CFRP). To estimate concrete strain due to temperature variations, four
thermocouples were bonded to the GFRP sheets in close proximity of the FOS in
four locations (North, South, East and West).
The primary purpose of this paper is to present experience gained in order to
promote confidence in new experimental technologies adopted in the Gentilly-1
containment ring beam repair. This paper presents the 8-year performance of the
repaired concrete and the sensor technologies used. Four key aspects of the
Gentilly-1 repair surveillance program are discussed:
i) strains and temperature variations of the repaired concrete and the GFRP
ii) effectiveness and performance of the GFRP for concrete repair
iii) testing and validation of fiber optic sensor technology, and
iv) effectiveness of the remote monitoring system.

376
 8. Issues Related to Operations, Inspection and Maintenance

Recent advances in seismic non-

destructive testing, and associated finite
element based evaluation, utilized on a
pre-stressed concrete reactor containment
at a NPP in operation (8-1882)

Ola Jovall1, Nils Rydén2, Allan Kristensen3

1
Scanscot Technology AB, Lund, Sweden, e-mail: [email protected]
2
Engineering Geology Faculty of Engineering, Lund University, Sweden
e-mail: [email protected]
3
Force Technology, Brøndby, Denmark, e-mail: [email protected]

The majority of the nuclear power plants in the world have been in operation for
a long period of time. The NPP industry is facing a period of huge investments
related to large-scale modernization and power upgrade projects of the existing
plants, and the building of new ones. One major financial condition for the pay-
off of these investments is to be able to operate the plants for a longer period,
thus raising questions related to residual lifetime estimations of containment
structures.
Within the scope of the Euratom FP5 program, a consortium consisting of
Force Technology, Denmark, Scanscot Technology, Sweden, Electricité de France,
France and Barsebäck Kraft, Sweden carried out the project “Concrete containment
management using the Finite Element technique combined with in-situ Non-
Destructive Testing of conformity with respect to design and construction quality
(CONMOD)” (see [1], [2] and [3]). The main conclusion from the CONMOD
project is that a new approach combining Non Destructive Examination with Finite
Element Analysis methods is both workable and necessary in order to be able to
accurately determine and predict the condition of nuclear power containment
structures.
The CONMOD project has been a pioneering study and has shown the way to
a new approach regarding condition assessment and ageing management. It was
however, largely a feasibility study. The proposed technology needs further
research and development along with site specific procedures to be implemented
and validated in practice.
At the moment, an implementation project regarding non-destructive
examination of a pre-stressed concrete reactor containment in operation, and
associated finite element technology, is carried out at a BWR plant in operation.
The project is based on the outcome of the CONMOD project described above.
The purpose of the project is to

377
8. Issues Related to Operations, Inspection and Maintenance

− implement recent advances in non-destructive testing inspection strategies

and planning
− validate and refine results from non-destructive testing using finite
element analyses
− study and development of logistics for application to concrete containments
in service.
The thickness, dynamic Young’s modulus, depth to delaminations, voids, and
cracks in concrete structures can be evaluated non-destructively by using seismic
wave based techniques. The speed and attenuation of seismic waves reflect the
dynamic elastic properties of the material. Seismic methods are therefore well
suited to solve problems related to ageing management of thick walled concrete
structures.
The Multichannel Analysis of Surface Waves method (MASW) is a non-
destructive seismic method to evaluate the stiffness (wave speed) variation with
depth. The method makes it possible to analyse all type of seismic waves
propagating within the structure. The Impact Echo (IE) method is a technique to
estimate the thickness of concrete structures. The method relies on a good
estimate of seismic velocities in the structure and a precisely measured resonant
frequency. These inherent limitations and advantages with each method support
a combined analysis. We present recent advances on these methods where both
resonant frequencies (IE) and seismic velocities (MASW) are evaluated in a
combined manner from the same multichannel data set.
Typically each traditional seismic non-destructive testing (NDT) technique is
based on a simplified model and measurement set-up, tuned for a specific
application, e.g. Impact Echo for thickness, surface waves for stiffness. These
simplified models are 1D models based on homogeneous layers with finite
thickness extending to infinity in the other directions. Further more, these
traditional methods cannot be used to accurately predict the effect of the finite
size source and receivers, near field effects, 2D and 3D effects from the actual
geometry and variable material properties, scattering from cracks and voids etc.
For the future research and development of NDT techniques for concrete
containment walls it is important to move towards more realistic 2D and 3D
theoretical models. It can be argued that the true potential of seismic wave based
techniques has not been fully explored due to the simplified theoretical models
used for the evaluation of measured data. Recent studies have demonstrated the
benefits of a joint evaluation (inversion) of different type of waves and resonant
phenomena. This research trend emphasizes the need to move from simplified
1D models to more sophisticated 2D and 3D full waveform models for the
evaluation of seismic measurements.

378
 8. Issues Related to Operations, Inspection and Maintenance

References
1. Force Technology, Scanscot Technology, Electricité de France, Barsebäck NPP,
Euratom FP5 FIKS-CT 2001-00204, CONMOD – Main report, 2005.

2. Jovall, O. et al. Scanscot Technology Technical report SCTE/02401/TR-01, CONMOD-

Evaluation of the containment at Barsebäck 1. Step 1: Initial structural analysis,
2002.

3. Jovall, O. Scanscot Technology Technical report 02423/TR-01, Barsebäck NPP –

Reactor containment Material testing project: Summary report, 2005.

379
8. Issues Related to Operations, Inspection and Maintenance

Study on the boric acid corrosion

behavior of disk/seat materials in
SI check valves (8-1889)

Hyun Young Chang1, Won Min Kim2, Tae Eun Jin3, Jeong Ho Son4,
Young Sik Kim5, Young Ran Yoo6
1
Senior Researcher, Korea Power Engineering Company
e-mail: [email protected]
2
Supervisor, Korea Power Engineering Company
3
General Manager, Korea Power Engineering Company
4
Manager, Korea Hydro & Nuclear Power Co. Ltd.
5
Professor, Andong National University
6
Researcher, University of Maryland, CALCE

The disk/seat materials of check valves in nuclear power station’s SI system have
been repeatedly damaged. From the analysis, several causes have been deduced.
The causes were classified as three categories: 1) cavitation from the valve
leakage, 2) cavitation from the valve disk chattering, 3) corrosion from the boric
acid evaporation and concentration.
The 3rd cause have been assumed that corrosion was occurred from the
concentrated boric acid which had been formed by boric acid particles blended
again with leaked water of the valve. These boric acid particles seemed to be
formed from the evaporation of boric acid that had been remained in the seat end
part of the valve which had minute leakages. These leakages could be occurred
in the condition that 2 phase fluid existed when the pressure of foregoing part in
the check valves decreased under the saturated pressure. This corrosion
phenomenon was occurred in the boundary region between water and steam.
Therefore this condition can be satisfied only when 2 phase fluids in front of the
check valves are formed. The fact that the upper part of the disk/seat had no
corrosion implies that boric acid existed in the form of particles above 185 that
is the temperature of boric acid dissolution. The previous experience also
showed that corrosion from evaporation and concentration was inevitable when
the foregoing part of the check valves in SI system were in the condition of 2
phase fluid. This conclusion was based on the observation that the backside of
the disk also had corrosion from evaporation and concentration.
In this study, the accelerated condition was formed in the environment of high
pressure and temperature and corrosion rates were estimated varying with
concentration of boric acid as a parameter. The corrosion rates were measured from
satellite that was the weld part of the currently used and A59 that was the candidate
weld materials for replacement. These materials were also used in the calculation
modeling the geometries of check valves using BEM (Boundary Element Method).

380
 8. Issues Related to Operations, Inspection and Maintenance

The complex approach to the determination

of NPP Steam Generators Heat-Exchange
Tubes Plugging Criterion (NPP SG HET PC)
on the basis of analyses of the processes
of the tubes damaging during
NPP operation (8-1910)

G.P. Karzov, S.A. Suvorov, V.A. Fedorova

FSUE CRISM “Prometey”

As a result of eddy-current control (ECC) methods development appears the

opportunity of early revealing of defects, which incipient in a walls of heat-
exchange tubes (HET) of steam generators (SG) during operation.
Preventive plugging of a defective HET, as a result of ECC, has essentially
increased the reliability and the safety of reactor facility operation, but thus
volumes of HET plugging have simultaneously sharply increased. In some cases
HET are unreasonably muffled as a result of imperfect ECC equipment
application, or a mistakes of the control measurements, or imperfection of
criterion of HET plugging. Unreasonable HET plugging essentially reduces SG
service life. In this connection the scientifically proved determining of SG HET
Plugging Criterion gets a problem a great value. This Criterion should be to
provide a necessary level of NPP safety, but thus simultaneously to minimize the
number of HET plugging.
The approach suggested in the given work is based on the developed Phasic
Model of Damage (PMD) of NPP WWER-1000 type SG HET. In this model it
has been shown, that by key parameters determining a dynamics of HET
damages, are the thickness of deposits on a surface of SG HET and kinetics of
an oxidizer reduction in a local electrochemical cell around of growing defect.
In this paper represented the results of a HET plugging criterion substantiation
by the example of one of NPP Unit with steam generators PGV-1000. The
corrosion defects of HET have been found out during the prestarting period of
that Unit after ECC of SG HET. By nature that defects constitutes as pitting and
the corrosion cracks developing from pitting. Pitting and cracks have arisen in
HET metal under the influence of chlorides and the oxygen, the which have got
in steam generators volume during installation of NPP equipment.
The substantiation of criterion of SG HET plugging was based on:
− Defect depth in a HET wall limiting values estimated definition from the
conditions of the crosspiece breakdown in front a crack tip due to

381
8. Issues Related to Operations, Inspection and Maintenance

instability of plastic deformation of metal. Limiting values depends on the

relation of depth of crack (a) to its length on a forming line of tube (l);
− Experimental confirmation of estimated value of limiting depth of defect;
− An estimation of the admitted size of defect in view of an opportunity of
its maximal growth during the period between periodical EСС;
− An estimation of the maximal error at definition of depth of a crack on
data ECC;
− Definition of plugging criterion, as the limiting value of a defect depth
determined by ECC, at which is created a danger of a tube destruction
excess during reactor facility operation at the period between periodical
EСС.

382
 8. Issues Related to Operations, Inspection and Maintenance

Seismic qualification and upgrade of safety

important pipelines support systems
in reactor building of units 5 and 6,
Kozloduy NPP (8-1934)

Stanislav Georgiev1, Dimityr Tanev1, Marin Jordanov1,

Maya Kancheva2, GeorgiKostov2
1
Structural and Seismic Engineer, EQE Bulgaria AD, Sofia
e-mails: [email protected], [email protected], [email protected]
2
Mechanical Engineer, EQE Bulgaria AD, Sofia
e-mails: [email protected], [email protected]

The aim of the task was to qualify seismically the existing safety important
piping installed in the reactor building. The need for seismic qualification came
from reassessment of review level earthquake (RLE) established for Kozloduy
NPP site. The pipelines should be properly qualified according to the new higher
seismic level.
In the previous steps of units 5 and 6 Modernization Program execution, the
list of pipelines was determined and the corresponding pipelines were modeled
using computer codes PepS and PipePlus. The pipes, which did not satisfy the
criteria, were identified. For this group of pipelines, the supports’ modifications
in the computer models were introduced, in order to achieve acceptable demand
to capacity D/C ratio for the most critical pipe sections. For each pipeline no less
than two different modifications of pipe support were proposed and compared,
considering different aspects as, cost effectiveness, number of additional
supports, ease of implementation on site, maintenance requirements, etc. Based
on the proposed modifications the decision for appropriate pipe supports
modification was taken and the corresponding detailed design was developed.
The support system modifications include modification of existing supports as
well as addition of new supports.
The pipelines to which the upgrade of support system was performed belong
to safety important systems such as: fire extinguishing system, spent fuel cooling
system, essential service water system, core cooling system and steamgenerator
emergency feeedwater system. All pipelines are situated in the reactor building.
The capacity of the supports was calculated according to ASME BPVC
Section III Subsection NF “Supports” and according to design data from the
support maker. For the check of the linear supports, the criteria from ASME NF-
3320 and ASME NF-3623 was used, with coefficient from Table NF-3623 (b)-1
according to the level of the analysis and the type of the load case.

383
8. Issues Related to Operations, Inspection and Maintenance

For the calculation of the additional and the modified supports, design during
the detailed design phase, the same code was used. For the design of the
necessary steel structure and the connection with the existing ones, the Bulgarian
standards for design of steel structures was used. For the design of the anchoring
in the concrete structures, Bulgarian codes for design of concrete and reinforced
concrete were used, along with technical data from the anchors producers.
The seismic upgrade of safety systems pipelines, vulnerable to seismic event
were successfully implemented on both units 5 and 6 during units outage in
2006.

384
 8. Issues Related to Operations, Inspection and Maintenance

Proactive Management of Materials

Degradation (PMMD) and enhanced
structural reliability (8-1954)

Doctor, S.R., Bond, L.J., Cumblidge, S.E., Hull, A, B., Malik, S.N.

The Nuclear Regulatory Commission (NRC) has undertaken a program to lay

the technical foundation for defining proactive actions so that future degradation
of materials in light water reactors (LWRs) is limited and, thereby, does not
diminish either the integrity of important LWR components or the safety of
operating plants. The United States is currently implementing license extensions
of 20 years on many plants, and consideration is now being given to the concept
of “life-beyond-60”, license extension from 60 to 80 years and potentially
longer. In support of NPP license renewal over the past decade, various national
and international programs have been initiated. This paper discusses efforts by the
U.S. Nuclear Regulatory Commissions Proactive Management of Materials
Degradation (PMMD) to determine the effectiveness of emerging NDE techniques.
All parts of a nuclear power plant (NPP) are subject to the continuous time-
dependent degradation of materials due to normal service conditions, which
include normal operation and transient conditions. The PMMD program is
investigating the many materials and components in NPPs and the materials
degradation phenomena that affect them in an attempt to predict and prevent
future problems. As some forms of degradation, such as stress corrosion
cracking, are characterized by a long initiation time followed by a rapid growth
phase, new inspection or monitoring technologies may be required. New NDE
techniques that may be needed include techniques to find SCC precursors, on-
line monitoring techniques to detect cracks as they initiate and grow, and
improved current NDE technologies. As the reactors operate well beyond their
originally-planned lifetimes many reactor components may have insufficient
NDE programs in place to prevent failures, as the NDE programs were not
designed for 60 or more years of operation.
The need for new NDE technologies and techniques necessitates a
methodology for the evaluation of new NDE techniques to determine if the new
NDE technologies have a strong technical basis in a timely fashion. Previous
efforts in evaluating NDE technologies and techniques have lacked a structured
methodology and have been very time consuming. The methodology is based on
previous evaluations of NDE technologies. The methodology consists of four
possible distinct phases. The first phase of the evaluation is an expert review of
the new technique that reviews the physics of how the technique works and the
applicability of the technique to the affected components. The second phase of
the methodology is a series of laboratory tests to determine the effectiveness of

385
8. Issues Related to Operations, Inspection and Maintenance

the technique under laboratory conditions and to explore the essential variables
for the technique. In the third phase, after the laboratory testing, a round-robin
test would be conducted using commercial vendors to determine if the field-
deployed systems can operate under realistic conditions. The final phase would
be the implementation of performance-based testing to assure that the inspectors,
technology, and technique are all able to provide an adequate probability of
detection for degradation in NPP components.
Additionally, a comprehensive review of reactor components will be needed
to determine if new inspection regimes may be required to deal with new
degradation mechanisms that may emerge over time. As reactors lifetimes are
expanded, degradation mechanisms previously considered too long-term to be of
consequence (such as concrete and wiring insulation degradation) may become
more important.
The paper describes the methodology developed to evaluate new NDE
technologies and techniques. A model BWR component is scrutinized to
determine the efficacy of the current NDE requirements for the component when
viewed at extending the lifetime of the component beyond 60 years is also
presented.

386
 8. Issues Related to Operations, Inspection and Maintenance

Ageing management of steam generator

internals (8-1959)

Koenig, Guenter1, Schoeckle, Friedrich2

1
EnBW Kernkraft GmbH, Kernkraftwerk Neckarwestheim
2
AMTEC Services, Lauffen, Germany
e-mail: [email protected]

Keywords: nuclear, power plant, steam generators, ageing, ageing management,

integrity, monitoring, preventive maintenance, failure orientated maintenance

Safe operation as well as economical aspects of nuclear power plants depend on

integrity and function of the components and systems in use. More general,
integrity and function can be expressed in terms of quality. Component quality is
established during the state of design and the manufacturing process. In
operation this quality (safety, remaining life) can be affected by ageing
phenomena.
Therefore, the EnBW nuclear power plants in Neckarwestheim, Germany,
have introduced a comprehensive ageing management process. Within ageing
management the quality of components and systems in operation has to be
assessed, regularly. Depending on the demands on the quality, three groups of
systems, structures and components have been defined:
− group 1: Integrity Concept (guarantee of integrity – control ageing phenomena)
− group 2: Preventive maintenance (maintain initial quality – prevent systematic
failures)
− group 3: Failure orientated maintenance (re-establish initial quality).
Components and systems that are relevant to the safety of a plant are classified into
groups 1 and 2. These groups usually are target of ageing management programs.
In the paper, the entire ageing management process is discussed using the
example of the steam generator internals.
The steam generator tubes (heat exchanger) are categorized into group 2
within the ageing management procedure, i.e. the quality in operation is
maintained by use of inspection and testing. Results of damage mechanisms in
operation like reduction of wall thickness, cracks etc. are controlled by non-
destructive testing. On the basis of the surveillance results the quality status is
reviewed, regularly (design aspects, actual structure, operation experience and
actual knowledge). There were some problems with wastage in the first years of
operation. Since a change in water chemistry these problems have been reduced,
drastically. Loose parts fretting still occurs, from time to time. This effect is

387
8. Issues Related to Operations, Inspection and Maintenance

controlled by non-destructive testing and by a very sensitive leakage (activity)

monitoring on the secondary side of the steam generators. In total, the number of
closed pipes due to wastage and fretting signs is low.
Based on the results of review discussions, other steam generator internals
have been introduced into the ageing management process. If the feedwater
distribution and support structure shows failures, for example, falling parts can
affect the integrity of the steam generator tubes. Consequently, these structures
have been included in the ageing management process. As with other systems,
structures and components of group 2, extensive measures of preventive
maintenance concentrate on the prevention of systematic failures.

388
 8. Issues Related to Operations, Inspection and Maintenance

Assessment of gaseous pollution from

hot cutting processes in NPP
disassembling (8-1960)
Franco G. Cesari1, Massimo Rogante2, Angelo Giostri3
1
Nuclear Engineering Lab., DIENCA, Univ. of Bologna,
Via dei Colli 16, 40136 Bologna, Italy
2
Rogante Engineering Office
Contrada San Michele 61, 62012 Civitanova Marche, Italy
3
Sogin S.p.A., 29012 Caorso, Piacenza, Italy

The gaseous pollution in NPP decommissioning is a main concern, and it should

be appropriately estimated. It is important, consequently, to generate a
methodology for the preventive evaluation of gaseous emissions occurring from
cutting process. Said assessment can be also referred to various other cases.
Cutting processes have been carried out in an experimental campaign at
Caorso NPP – Boiling Water Reactor (BWR), 870 MWe – built in the 70s,
completely operating in the years 1981–1986, shut down on 1987 and at present
under dismantlement.
In the present work, the calculated values of the pollution arisen from plasma
torch and oxyacetylene arc cutting processes are shown, related to the volatilized
material, the volume of the powders per hour, and the concentration in the
environment and at the inhalation. The cut components – carbon steel with
different surface treatments – are no activated plates and pipes exposed to the
reactor coolant steam.
The evaluation of the dilution between emissions and air at the inhalation is discussed,
introducing also suggestions to perform the said cutting procedures in a safe way.

References
1. F. Cesari, M. Rogante, A. Giostri. Results of the experimental campaign on contaminated
metal components parameters and suggestions for safely NPP component
dismantling. Nuclear Engineering and Design, Volume 238, Issue 10, (2008),
pp. 2801–2810.
2. F.G. Cesari, M. Rogante, A. Giostri, G. Conforti. Contaminated metal components in
dismantling by hot cutting processes. Proc. ICONE14 Int. Conf. on Nuclear
Engineering, July 2006, Miami, Florida, USA, ASME ICONE14-89451 (2006).
3. F. Cesari, A. Giostri, M. Rogante, E. Sirito, M. Sirito. Hot cutting processes and emissions
characterization in metal components dismantling. Proc. of 11th Int. Scientific Conf.
on the Contemporary Achievements in Mechanics, Manufacturing and Materials
Science CAM3S`2005, Gliwice-Zakopane, Poland, 6th–9th December 2005, L.A.
Dobrzański, Ed., International Organising Committee of the Scientific Conferences
AMME World Press, Gliwice, Poland (2005). Pp. 1111–1116.

389
8. Issues Related to Operations, Inspection and Maintenance

Repair and strengthening of damaged

reinforced concrete slabs
with CFRP (8-2047)

D. Dauffer1, A. Limam2, D.T. Nguyen2, J.M. Reynouard2

1
EDF-Septen, France
2
Université de Lyon, Insa-Lyon, LGCIE, F-69621, Villeurbanne, France

In the past few years, several major international research programs were
launched to investigate the feasibility of using technologies of polymer
composite for repair and retrofitting structural systems. One of these successful
applications is the external strengthening for repair and upgrade the structural
capacity and rigidity of axially loaded concrete members in particular beams or
columns. In these cases, several laminates of composite are bonded to the
finished concrete surface in the hoop or longitudinal direction for enhancing the
flexural capacity of these members.
The object of the present study supported by EDF (Electricité de France)
focuses on the investigation of the behaviors of large-scale RC slabs retrofitted
by CFRP for the purpose of the application of this reinforcement process to the
RC shell of hyperbolic cooling towers. The objectives were to observe the
phenomenological behavior of the reinforced and unreinforced slab subjected to
combined membrane and bending loads and to provide useful data in checking
the accurateness of non-linear finite element simulation.
A specific experimental device built at INSA Lyon allows the application to
concrete slabs (2000 × 2000 × 100 mm), of varied combinations of membrane
and flexural loads in two directions. Several tests are conducted to gauge the
pertinence and the limit of the CFRP reinforcement on the non damaged or pre-
damaged concrete slab. The initial damage is obtained thanks to an initial
loading where membrane and/or flexural loads are combined; the structure is
then unloaded and repaired with carbon fibre reinforced polymer strips. Test
results indicate that the composite system restores not only the original capacity
of the damaged slab but also an appreciable enhancement of the bearing
capacity.
The finite element software CAST3M is used to analyze the unreinforced and
reinforced concrete slabs. For this purpose, the multilayered DKT element
(Discrete Kirchhoff Triangle) is used to simulate the concrete, the steel
reinforcement and the CFRP layer. Comparison between experimental and
numerical results shows good agreement for both the control slab, and the non-
damaged reinforced or the damaged reinforced concrete slabs.

390
 8. Issues Related to Operations, Inspection and Maintenance

References
1. Täljsten, B., Enochsson, O. Lundqvist, J., Rusinowski, P., Olofsson, T. CFRP strengthened
openings in two-way concrete slabs – An experimental and numerical study.
Construction and Building Materials, Vol. 21, Issue 4, April 2007. Pp. 810–826.

2. Rochdi, E.H., Bigaud, D., Ferrier, E., Hamelin, P. Ultimate behaviour of CFRP strengthened
RC flat slabs under a centrally applied load. Composite Structures, 72(2006),
pp. 69–78.

391
8. Issues Related to Operations, Inspection and Maintenance

Development of integrity evaluation

program for pipe wall thinning (8-2071)

Sung Ho Lee, Tae Ryong Kim, Chi Yong Park

Korea Electric Power Research Institute
65 Munji-ro Yuseong-gu, Daejeon, Korea
e-mail: [email protected]

Evaluation of pipe integrity against local wall thinning due to erosion and corrosion
is increasingly important in maintenance of carbon steel pipes in nuclear power
plants. Though a few programs for assessment of the thinned pipe integrity have
been developed in domestic nuclear engineering, they are limited to straight
pipes using methodology proposed in ASME Section XI Code Case N-597.
In this study, an engineering program for evaluation of the integrity of all
kinds of pipes such as straight, elbow, reducer and branch pipes was successfully
developed, which was designated as PiTEP® (Pipe Thinning Evaluation
Program). The developed program includes four sequential steps to evaluate the
integrity such as, first evaluation step on design code (ASME Section XI Code
Case N-597), next its engineering method and then a couple of evaluation
methods of our own. As PiTEP® is constructed in user friendly GUI (graphic
user interface) environment, we expect plant engineer can easily operate it only
with measured thickness data, basic operation conditions and pipe data. Some
experimental tests were also conducted with elbows to verify that the results of
the program for the assessment of thinned pipe integrity are appropriate.
From the development of evaluation program for pipe integrity against local
wall thinning, we had some conclusions as follows; first, it was possible to have
optimum criteria for thickness of thinned pipe for its integrity based on the recent
technological methods in the program. Secondly, it was found that the results of
the program have sufficient conservative margin comparing to the verification test
results. Thirdly, PiTEP® can be easily applied to nuclear plants because the
program is constructed in user friendly GUI (graphic user interface) environment.

References
1. Chexal, B., Horowitz, J., Dooley, B., Millett, P., Wood, C., Jone, R. 1998. Flow
accelerated Corrosion in Power Plant. EPRI/TR-106611-R2.

2. ASME B&PV Sec. XI Div.1, 1998ed., Code Case N-597. Requirement for Analytical
Evaluation of Pipe Wall Thinning.

3. Korea Hydro & Nuclear Power Co., 2008. Optimization of Thinned Pipe Management
Program and Application.

392
 8. Issues Related to Operations, Inspection and Maintenance

Fabrication flaw density and distribution in

piping weldments1 (8-2476)

S.R. Doctor
Pacific Northwest National Laboratory
P.O. Box 999, MSIN K5-26, Richland, WA 99354
[email protected]

The U. S. Nuclear Regulatory Commission supported the Pacific Northwest

National Laboratory (PNNL) to develop empirical data on the density and
distribution of fabrication flaws in nuclear reactor components. These data are
needed to support probabilistic fracture mechanics calculations and studies on
component structural integrity. PNNL performed NDE inspections and
destructive testing on archived piping welds to determine the fabrication flaw
size and distribution characteristic of the flaws in nuclear power plant piping
weldments. Eight different processes and product forms in piping weldments
were studied including wrought stainless steel and dissimilar weldments.
Parametric analysis using an exponential fit was performed on the data. Results
were created as a function of the through wall size of the fabrication flaws as
well as the length distribution. The results are compared and contrasted with
those developed for reactor pressure vessel processes and product forms. The
most significant trend was that the density of fabrication flaws versus through
wall size was higher in piping weldments than that for the reactor pressure vessel
weldments but smaller than that found for reactor pressure vessel weld repairs.
Curves showing these distributions are presented.

_________________
1
The work was sponsored by the U.S. Nuclear Regulatory Commission under U. S. Department of
Energy Contract DE-AC05-76RL01830; NRC JCN Y6398; Mr. Wallace Norris, Program Monitor.

393
8. Issues Related to Operations, Inspection and Maintenance

Fabrication flaw density and distribution

in weld repairs1 (8-2477)

S.R. Doctor
Pacific Northwest National Laboratory
P.O. Box 999, MSIN K5-26, Richland, WA 99352, USA
e-mail: [email protected]

The Pacific Northwest National Laboratory (PNNL) is developing a generalized

flaw distribution for the population of nuclear reactor pressure vessels and for
piping welds in the U. S. operating reactors. The purpose of the generalized flaw
distribution is to predict component-specific flaw densities. The estimates of
fabrication flaws are intended for use in fracture mechanics structural integrity
assessments. Structural integrity assessments, such as estimating the frequency
of loss-of-coolant accidents, are performed by computer codes that require, as
input, accurate estimates of flaw densities. Welds from four different cancelled
reactor pressure vessels and a collection of archived pipes have been studied to
develop empirical estimates of fabrication flaw densities.
This paper describes the fabrication flaw distribution and characterization in
the repairs weld metal of vessels and piping. This work indicates that large flaws
occur in these repairs which are complex in composition and sometimes include
cracks on the ends of the repair cavities. Parametric analysis using an
exponential fit is performed on the data.
Construction records where available were reviewed. It is difficult to make
conclusions due to the limited number of construction records reviewed.
However, the records reviewed to date show a significant change in repair
frequency over the years when the components in this study were fabricated. A
description of repair flaw morphology is provided with a discussion of fracture
mechanics significance.

_________________
1
The work was sponsored by the U.S. Nuclear Regulatory Commission under U. S. Department of
Energy Contract DE-AC05-76RL01830; NRC JCN Y6398; Mr. Wallace Norris, Program Monitor.

394
 8. Issues Related to Operations, Inspection and Maintenance

Graphite blocks reloading consideration

in HTR-PM (8-2486)

Sheng Xuanyu, Wu Xinxin, He Shuyuan

Institute of Nuclear and New Energy Technology
Tsinghua University Beijing, 100084 P.R. China

After the experimental reactor HTR-10 was built and succeeded in operation,
INET is designing another reactor, HTR-PM. The service life was 40 years and
the graphite blocks reloading was not considered. But we design the graphite
reloading scheme for the future use in HTR.
There are some common requirements for graphite block reloading in high
temperature reactor:
1. Strict control leakage of radiation.
2. Shield requirement.
3. High reliability, simple structure of reloading equipment.
4. Remote control of the reloading equipment.
Compare to the HTGR and HTTR, there are some differences for HTR-PM:
1. Only the graphite structure was reloaded in HTR-PM, other metal
equipments were not reloaded.
2. The surrounding condition was air, not the Helium. For HTGR and
HTTR, the gas in the reactor is helium when reloading graphite.
3. The reloaded material has less radiation for HTR-PM. The estimated
radiation was 1Ci.
4. The graphite block reloaded in HTR-PM was more complicated in
shape. There were 3 kinds of graphite: top reflector, central graphite
cylinder and side reflector.
Special process was designed for the graphite block reloading. The preparation
process includes 6 steps. For the reloading process, exactly 90 steps were
contrived to reloading 3 kinds of reflectors. We don’t give detail here in the
abstract, but will give in our paper.

395
8. Issues Related to Operations, Inspection and Maintenance

8 kinds of assembly products were designed for the graphite reloading:

1. Top reflector reloading machine, shown in Fig. 1.
2. Central graphite cylinder reloading machine, shown in Fig. 2.
3. Side reflector reloading machine, shown in Fig. 3.
4. Hoisting barrel.
5. Sealing gate
6. Transfer channel
7. Transfer trolley
8. Crane.

Figure 1. Top reflector reloading machine.

396
 8. Issues Related to Operations, Inspection and Maintenance

Figure 2. Central graphite cylinder reloading machine.

397
8. Issues Related to Operations, Inspection and Maintenance

Figure 3. Side reflector reloading machine.

398
 8. Issues Related to Operations, Inspection and Maintenance

Korean experience in aging management

for long term operation of NPP (8-2533)

Tae Ryong Kim

Korea Electric Power Research Institute
65 Munji-ro Yuseong-gu, Daejeon, Korea
e-mail: [email protected]

Recently long-term operation of nuclear power plant beyond its licensed term
has become a worldwide trend in so called nuclear renaissance era. To obtain
license approval of the long-term operation, aging management strategies and
programs of systems, structures, and components should be established and
implemented [1].
Aging management programs (AMP) were established for two leading nuclear
power plants in Korea, that is, Kori Unit 1 for PWR and Wolsong Unit 1 for
PHWR. A total of 39 AMPs were prepared for Kori Unit 1 according to
NUREG-1801[2]. These included an in-service inspection of safety class 1, 2, 3
components, nickel-alloy nozzles and penetrations, reactor vessel surveillance,
loose part monitoring, electrical cables and connections not subject to
environmental qualification requirements, etc. Alloy 600 AMP was prepared to
ensure that integrity of nickel-alloy nozzles and penetrations is maintained
against the primary water stress corrosion cracking (PWSCC), which recently
has become an important issue worldwide. This includes a survey of all Alloy
600 dissimilar metal welds in Kori Unit 1 and an assessment of the integrity of
components highly susceptible to PWSCC. Residual stress analysis and
measurement in dissimilar metal weldment were also studied.
Time-limited aging analysis (TLAA) of the critical SSC related to safety
should be also done and the report should be submitted for the long term
operation of a NPP according to MEST Notice 2008-17 [3]. A total of 10 TLAA
items for Kori Unit 1 were identified and reviewed. These included reactor
vessel neutron embrittlement analysis, metal fatigue analysis, environmental
qualification of equipment, wear of the neutron flux detector tube, the reactor
coolant pump flywheel, the spent fuel pool liner, thermal aging embrittlement of
the cast austenitic stainless steel, etc.
It was found that most of the existing AMPs can properly and effectively
manage the aging phenomena shown in the critical SSC. However, for some of
them, e.g. the nickel-alloy nozzles and penetrations, the AMP revision was
necessary. 13 AMPs in total including boric acid corrosion program and Alloy
600 program were revised or were newly prepared. In the TLAA of the metal
fatigue of the main components (reactor vessel, control rod drive mechanism,
reactor internals, reactor coolant pump, steam generator, reactor coolant system
piping, safety injection tank), the cumulative usage factor was found to be less than

399
8. Issues Related to Operations, Inspection and Maintenance

1.0 (acceptance criteria), even with the occurrence of transients conservatively

predicted at the end of the extended operation of 40 years. A fatigue monitoring
system was additionally installed in Kori Unit 1 to ensure operational safety for
another 10 years of operation. It was subsequently confirmed that the integrity of
the main components would be maintained until the end of continued operation
of Kori Unit 1[4].

References
1. IAEA, Implementation and Review of NPP Aging Management Program. Safety Series
No. 15, 1999.

2. NRC, NUREG-1801, Generic Aging Lessons Learned (GALL) Report. Vol. 1, Rev. 1,
2005.9.

3. MEST, Notice 2008-17. Guidance for Technical Standards applied to Continued Operation
of Nuclear Facilities beyond their Design Life. 2008.4.18.

4. Tae-Ryong Kim, Tae-Eun Jin. Technical Evaluation of the Continued Operation of

NPP. Nuclear Engineering and Technology, Vol. 40, No. 4, 2008.6.

400
 8. Issues Related to Operations, Inspection and Maintenance

Numerical and analytical framework

for analysis of crack initiation and
propagation under thermal
fatigue loading (8-2581)

Igor Varfolomeyev
Fraunhofer IWM, Freiburg, Germany

Material degradation at thermal fatigue loading is an important problem arising

in the life-time management of piping components of nuclear power plants
(NPP). In particular, cracking phenomenon in tee connections due to mixing of
the hot and cold fluids has been extensively investigated in a past decade within
national and European research projects, e.g. THERFAT, NULIFE.
This paper considers several aspects related to the numerical and analytical
modelling of crack initiation and propagation under thermal fatigue loading.
These include the assessment of thermal-hydraulic boundary conditions, stress
analysis, fatigue and fracture mechanics calculations. Appropriate analytical
models and tools are presented demonstrating possibilities for a simulation of
surface crack growth under random thermal-mechanical loading. The significance
of the data scatter is additionally discussed and examples of a probabilistic
analysis are presented.

401
8. Issues Related to Operations, Inspection and Maintenance

A model to monitoring real-time fracture of concrete

subjected to the load from tendons by
AE technique (8-2603)

Yu-Cheng Kan1, Kuang-Chih Pei2, Dong-Wei Lin2

1
Associate Professor, Department of Construction Engineering, Chaoyang
University of Technology, Taiwan
2
Engineers, Institute of Nuclear Energy Research, Taiwan

Using acoustic emission technique (AE) to monitor the load test on RC structures
can reveal more details for the material integrality. Many of relevant studies [1, 2]
have verified that Kaiser Effect exists in reinforced concrete structures, which
could be applied in an aged pre-stressed concrete structure that the tendons need to
be replaced. This application may be useful on the safety-related or high-strength
demanded structures, such as the post-tension (tendon) system of the pre-stressed
reinforced concrete containment (PRC). The tendon system results in a distributive
pressured state within the concrete of the entire containment wall to prevent
fracture [3] due to the inside peak pressure; however, some parts of this system
like the anchor trumplate or the bearing plate may cause noticeable concentration
load on the concrete. It is noted that most non-destructive testing (NDT) techniques
are not possible to inspect the defects within these areas. This research is addressed
on the development of AE technique to monitor the routine in-service inspection
(ISI) program with loosing and reloading procedures. Through the monitoring
results, the integrality situation at these critical areas may be evaluated. The issue
is significant for those containments which need to be extended in their service lives.
This research provides a preliminary study for a mock-up test later to model
the concrete around and under the bearing plate. The AE technique using a self-
developed instrument is applied to record/monitor the emitted ultrasonic wave in
a small-scale specimen during load test being performed in a MTS system. The
obtained hit-count history profiles (vs. time or load), the Kaiser Effect (or the
crack incubation) and the crack-control mechanism of re-bars in concrete will be
observed and discussed in this paper.

References
1. Kan, Y.-C., K.-C. Pei, D.-W. Lin. 2008. Monitoring the Real-time Fracture within
Reinforced Concrete Under Load Test Using Acoustic Emission Recording
Technique, Journal of ACI Structural. (In review)
2. Kan, Y.-C., K.-C. Pei, D.-W. Lin. 2008. Using Acoustic Emission Technique for
Monitoring Real-time Fracture Behavior within Reinforced Concrete Structure
Element under Load Test. Proceeding, 17th WCNDT, October 25-28, Shanghai,
China. 28 Oct 2008, Shanghai, China.
3. Shah, S.P., Swartz, E.S., Ouyang, C. 1995. Fracture Mechanics of Concrete. 1st ed.,
John Wiley & sons, Inc.

402
 8. Issues Related to Operations, Inspection and Maintenance

Methodology research on prediction for

operating lifetime of PWR RPV (8-2619)

Yinbiao He, Wangping Zhang, Ming Cao, Qiuping Shen

Shanghai Nuclear Engineering Research and Design Institute
Shanghai, China
e-mail: [email protected]

The residual lifetime of the operating PWR NPP is always concerned by the
owners and the RPV plays a critical role in the lifetime prediction of PWR NPP
due to irradiation induced embrittlement of the active core beltline and its non-
replaceablity. The PWR NPP in question have been operating over half of the
initial design lifetime in China and overall survey and investigation for the RPV
have been carried out focusing on the Time Limited Aging Analysis (TLAA).
Based on the assessment of the pressure - temperature limits on maintaining the
reactor coolant pressure boundary, integrity of RPV under the potential pressurized
thermal shock (PTS) and the upper shelf energy (USE) of irradiated beltline, the
residual lifetime of RPV is predicted in this paper.

403
8. Issues Related to Operations, Inspection and Maintenance

The effectiveness of chemical cleaning

in reducing the risk of leakage in
steam generator tubing:
a Bayesian approach (8-2621)

Mahesh D. Pandey, S.V. Datla, Mikko I. Jyrkama,

Department of Civil and Environmental Engineering, University of Waterloo
200 University Avenue West, Waterloo, Ontario, N2L 3G1, Canada
e-mail: [email protected]

Steam generators (SG) experience a variety of corrosion related degradation

mechanisms, such as pitting, wall thinning and stress corrosion cracking.
Chemical cleaning and water lancing are two effective maintenance methods
that remove the sludge from the steam generator and minimize the potential for
under deposit degradation of SG tubing. In a risk-based life cycle management
of SG fleet, it is important to quantify the benefit of the maintenance (chemical
cleaning or water lancing) in terms of reduction in the risk of tube leakage that
leads to the forced outage of the reactor.
The aim of the paper is to present a probabilistic model to evaluate the risk of
tube leakage in steam generators and calibrate the parameters using realistic data
from a nuclear station in Canada. A Bayesian methodology is presented to
evaluate the effectiveness of chemical cleaning based on the data regarding the
occurrence of leakage before and after chemical cleaning on SGs in nuclear
plant in Canada.
The practical examples presented in the paper show that chemical cleaning is
partially effective in reducing the risk of tube leakage.

404
 8. Issues Related to Operations, Inspection and Maintenance

A proposal for a unfied model on nuclear

power plant life management including
maintenance optimisation (8-3173)

P. Contri
European Commission, JRC-Institute for Energy, Safety of Current Nuclear
Reactors Unit, PO Box 2, 1755 ZG Petten, The Netherlands1
e-mail address of main author: [email protected]

In recent years many electric utilities and nuclear power plants adopted policies
for improved coordination of both safety and non-safety programs, called plant
life management (PLIM), also in view on plant life extension programs, but
mainly for an optimisation of operating costs. The implementation of PLIM
programs has followed many different approaches, being intrinsically dependent
on the national regulatory framework and technical traditions.
In Countries with some experience, the PLIM program proved very convenient,
especially when coupled with Maintenance, Surveillance an Inspection (MS&I)
optimization: average savings are reported in the range of 20–30% of total
(maintenance) costs.
Moreover, in terms of safety, the control of equipment reliability, significantly
improved with PLIM models for example through Ageing Management Program
(AMP) and Reliability Centred Maintenance (RCM), made a long term asset
management of the overall plant possible and the overall safety indicators
significantly improved in many cases.
This is why R&D tasks are needed in this phase, not only in the long term
extrapolation of the component integrity and behaviour, but a also in new
management strategies at the plant (PLIM), able to address organisational issues,
spare part management, staff ageing, component obsolescence, etc, which are
typical components of the PLIM.
The Framework Programme 7 of the EU in the area dedicated to the reactor
systems calls for a research effort “to underpin the continued safe operation of all
relevant types of existing reactor systems (including fuel cycle facilities), taking into
account new challenges such as life-time extension and development of new
advanced safety assessment methodologies (both the technical and human element)”.
A unified European model for PLIM was developed at the JRC with the support
of a network of stakeholders (SENUF), and validated at some EU nuclear plants.

________________
1
The views expressed in this paper are those of the author and do not necessarily reflect those of
the EC.

405
8. Issues Related to Operations, Inspection and Maintenance

This paper provides a summary of the model features, the result of its
validation at some plants and summarises the perceived scientific/technological
challenges for the FP7 on which JRC proposes to focus, based upon its
competencies and skills, having in mind both the European and world-wide
context and its potential evolution.

406
9. Waste Management, Fuel Cycle Facilities
and Decommissioning

Evaluation of existing facilities. Risk-informed and performance-based

considerations. Probabilistic safety assessment. Siting, design and
construction of fuel cycle facilities. Issues related to long-term and short-
term waste storage.

407
9. Waste Management, Fuel Cycle Facilities and Decommissioning

408
 9. Waste Management, Fuel Cycle Facilities and Decommissioning

Development of neutron shielding materials

for nuclear fuel storage facilities (9-1707)

Herve Issard
TN International (AREVA Group)
1, rue des hérons 78180 Montigny le bretonneux, France
e-mail of corresponding author: [email protected]

In the context of optimisation of nuclear power plants, spent fuel discharged

from reactors have higher burn up and consequently have more residual heat.
Installations and storage casks have to withstand the severe temperatures resulting
from increasing residual heat of the spent nuclear fuel.
A family of new neutron shielding materials was developed by AREVA TN
International for spent nuclear fuel in storage facilities and transport/storage
casks showing high shielding capabilities for a range of temperature corresponding
to the needs.
These materials are composed of a thermosetting resin (vinylester or polyester
resins) and mineral fillers (alumine hydrate and zinc borate). The cross linking
of the polymer leads to a rigid three-dimensional lattice, solid and resisting to
transport conditions, especially the temperatures.
Tests of the neutron shielding materials have been performed to establish:
− Shielding/attenuation efficiency in compliance with nuclear site specification,
in case of transport compliance with IAEA safety standards TS-R-1.
− Good behavior in accident conditions.
− Radiation and heat resistance during the period considered: storage (40
years) or transport (depending of transport frequency).
− Quality and homogeneity of the shielding material.
− Fire resistance, with self extinction property.
Its shielding ability for neutron radiation is related to a high hydrogen content
(for slowing down neutrons) and a high boron content (for absorbing neutrons).
Source of hydrogen is organic matrix (resin) and alumine hydrate; source of
boron is zinc borate. Atomic concentrations are equal to 5.1022 at/cm3 for hydrogen
and 9.1020 at/cm3 for boron.
The manufacturing process of these materials is easy: it consists in mixing the
fillers and pouring at ambient temperature. It allows to obtain any geometry.
Temperature resistance of these materials was evaluated by performing
accelerated tests of samples at different temperatures (120°C to 170°C). According
to tests results, the range of maximal temperature of use can reach 160°C.
Several shielding materials from TN International are now qualified for use in
Germany, France, Belgium and other countries.

409
9. Waste Management, Fuel Cycle Facilities and Decommissioning

Modelling the aging of concrete as a

technical barrier in nuclear waste
disposal facilities (9-1834)

Olli-Pekka Kari, Jari Puttonen

Helsinki University of Technology
Faculty of Engineering and Architecture
Department of Structural Engineering and Building Technology
P.O. Box 2100, FIN-02015 TKK, Finland
e-mail: [email protected]

Introduction
The low- and intermediate level wastes accumulating during the operation of
nuclear power plants may be disposed of in an underground repository in the
bedrock. The safety of the repository is ensured by multiple engineered barriers
wherein the concrete-made containers usually play an important role. It is
required that the barriers must be serviceable at least 500 years after sealing the
repository. However, there exists neither experience nor any historical evidence
of reinforced concrete structures, whose service life is even close to that
demanded. The reinforced concrete has been used as a construction material
about 150 years. Therefore, the designing and justification of structures of that
kind have to be based on knowledge of fundamental degradation mechanisms of
reinforced concrete under such conditions. This also emphasized the importance
to develop a mathematical model by which the influence and interaction of
recognized degradation mechanisms can be estimated.

Scope of the work

The primary object of the study is to develop a generalized numerical model for
estimating concrete degradation in final disposal conditions in Finnish bedrock.
Specific requirements are that the numerical implementation of the model has to
be made by using the finite element method and a computer program that is
commonly available. The mechanisms recognized and studied were: aerial
carbonation of concrete, moisture ingress, chloride penetration, concrete
corrosion caused by both sulfate and magnesium intrusion and leaching of
cement paste compounds into ground water. In addition, the effects of concrete
admixtures (herein silica fume and blast furnace slag) were included into the
FEM-model. The main objectives of the FEM-based modeling are to make it
possible to consider different geometrical shapes and possible variation in
material properties of concrete structures with a reasonable accuracy.

410
 9. Waste Management, Fuel Cycle Facilities and Decommissioning

Essential results
In this phase a conceptual structure of the model has been created and the model
has been tested by two-dimensional numerical analyses made for some details of
nuclear waste disposal facilities under consideration in Finland. In the analyses,
the coupling between all the relevant deterioration mechanisms was considered.
The results have been compared with values obtained from experiments and
conventional calculation methods.
The results clearly demonstrate the effects of different types of concrete mixes
on deterioration of reinforced concrete structures in studied environment. A low
water-to-binder ratio retards the ingress of harmful compounds and delay
corrosion initiation. Using of silica fume or blast furnace slag as a replacement
of cement has beneficial effects on the durability of reinforced concrete as well.
The results received have already confirmed the initiating hypothesis of the
study that the long-term deterioration of reinforced concrete cannot be estimated
with a sufficient accuracy by conventional single phenomenon models, but the
interaction between different deterioration phenomena shall be considered. The
ingress of harmful compounds into concrete depends on compound specific
diffusivities. The results received indicate that the interaction have a significant
and time-depended effect on the diffusivities that also affect the estimated
penetration depths of compounds in long term. Therefore, the benefits of the
method developed and its numerical implementation are obvious for safety
analyses of nuclear waste disposal facilities.

411
9. Waste Management, Fuel Cycle Facilities and Decommissioning

Numerical and experimental structural

assessment of a half scale model of a
nuclear spent fuel elements transportation
package under 9 m drop tests (9-1927)

Miguel Mattar Neto, Carlos Alexandre de Jesus Miranda, Gerson Fainer

Nuclear and Energy Research Institute, IPEN – CNEN/SP, Av. Prof. Lineu
Prestes 2242, 05508-000, São Paulo, SP, Brazil
e-mails: [email protected], [email protected], [email protected]

The applied qualification requirements for the packages used in the transportation of
nuclear spent fuel elements are very severe due the nature of the radioactive
content. They include the so-called normal conditions of transport and the
hypothetical accident conditions.
The 9 m drop tests are the most critical hypothetical accident conditions. The
package qualification under these conditions shall be conducted using full scale
models (prototypes), small scale models, numerical simulations and a combination
of physical tests and numerical simulations. The choice of the qualification
approach depends on economical and safety aspects.
To comply with the nuclear safety functions, as the containment of the internal
products and biological shielding, the package itself has several components
connected to each other in different ways (welded parts, flanged connections,
surface contacts, etc.).
This paper presents a discussion on the combination approach with tests and
numerical simulations for the structural assessment of a half scale model of a package
for transportation of nuclear research reactor spent fuel elements under 9 m drop tests.
The numerical simulations of the 9 m free drops over a rigid surface of half scale
model of the transportation package under different orientations were conducted using
a finite element explicit code considering several nonlinear aspects as the nonlinear
materials models and properties, the different package materials stiffness, and the
different types of the contacts between the package components and between the
package and the rigid surface, including the friction in the contacts.
The tests were also conducted for different drop orientations. The half scale
model under test was instrumented to assess the deceleration levels and all tests
were recorded with pictures, high speed camera movies, size measurements and
localized damage characterization.
The numerical and experimental results are compared and comments and
conclusions are addressed based on the comparison. Also, some recommendations
are issued on the use of the numerical simulations for the full scale tests of the
package.

412
 9. Waste Management, Fuel Cycle Facilities and Decommissioning

Technical challenges related to the spent

nuclear fuel dry cask storage/transportation
analysis and design (9-1936)

Tripathi Bhasker (Bob) P.

United States Nuclear Regulatory Commission
Mail Stop: EBB-3D-02M, Washington, DC 20555-0001, USA
e-mail: [email protected]

Keywords: spent nuclear fuel, dry cask storage, storage and transportation, technical
challenges

With the introduction of new, and highly sophisticated spent fuel dry cask
storage and transportation designs involving three-dimensional non-linear dynamic
finite element analyses, using codes such as: ANSYS, LS-DYNA, ABAQUS and
others, industry as well as the U. S. Nuclear Regulatory Commission (NRC) is faced
with unique technical challenges for which there are few precedents.
Licensees, spent nuclear fuel dry cask storage and transportation vendors, and,
in turn, the NRC are managing the potential application of: 1) exotic and un-
conventional non-code approved materials; 2) increased mass and fissile
material loads; 3) increased number of fuel assemblies for both, boiling water
reactor (BWR) and pressurized water reactor (PWR) fuels; 4) burn-up credit; 5)
fracture mechanics analytical approach; 6) loading high burn-up fuels; and 7)
analysis vs. testing, etc. Addressing these design considerations has created new
technical challenges to demonstrate that the integrity of the fuel assemblies is
maintained during the storage and/or transportation phase, and to demonstrate
compliance with the applicable requirements of Title 10 of the Code of Federal
Regulations (10 CFR) Part 72, and 10 CFR Part 71 respectively.
This paper will discuss, in general, these and other issues from an overall
perspective. This paper will also elaborate, in detail, some of the issues related to
analysis vs. testing, and the merits and pitfalls of full-scale vs. scaled model
testing used to demonstrate the structural integrity of the cask and its contents. I
will also briefly discuss pre-test and post-test calculations, validation and
benchmarking of analyses based on drop tests, test positions, test sequences and
other state-of-the-art measurement techniques to measure the decelerations, and
other relevant issues.

413
9. Waste Management, Fuel Cycle Facilities and Decommissioning

Optical strain measurement of plastic strain

localization in nuclear waste copper
canisters (9-2012)

Kati Savolainen, Tapio Saukkonen, Hannu Hänninen

Department of Engineering Design and Production, Laboratory of Engineering
Materials, Helsinki University of Technology, Espoo, Finland
e-mail: e-mail: [email protected]

Keywords: copper, weld, plastic strain, localization

In Finland and Sweden the spent nuclear fuel will be stored in a deep repository
in copper corrosion barrier canisters surrounding cast iron inserts. The 50 mm
thick copper canisters will we sealed using either electron beam welding (EB) or
friction stir welding (FSW) to join the tubes and the lids/bottoms. The canisters
will deform in the repository conditions e.g. due to hydrostatic pressure. The
deformation will localize to different discontinuities, such as defects and
microstructural heterogeneity.
This study compared the localization of plastic deformation in EB and FSW
welds as well as in the base materials (both forged and extruded) using optical
strain measurement methods. The results show that in the base materials the
deformation occurs very uniformly over the entire gauge length. In FSWelds the
deformation localizes in the middle of the weld, however, the tensile strength is
similar to that of the base materials. In EB welds the deformation localizes to the
large grains in the middle of the weld and to the steep grain size gradient
between the weld and the base material. Tensile strength is lowest in the EB
welds (175 MPa as compared to 200 MPa or higher for the other samples).

414
 9. Waste Management, Fuel Cycle Facilities and Decommissioning

Planning of one-piece removal of BWR

reactor pressure vessels at Barsebäck
Nuclear Power Plant,
Unit 1 & 2 (9-2571)

Ola Jovall1, Leif Johansson2

1
Scanscot Technology AB, Lund, Sweden, e-mail: [email protected]
2
Barsebäck Kraft AB – Decommissioning, Löddeköpinge, Sweden
e-mail: [email protected]

The decommissioned Barsebäck Nuclear Power Plant (NPP) with its two units
Barsebäck 1 and 2 is situated on the Swedish west coast in the rural district of
Kävlinge. The building process of Barsebäck 1 was started in 1970 and the plant
was in commercial service in 1975. Barsebäck 2 was in service two years later.
The two reactors are in principal identical and of BWR-type (Boiling Water
Reactor) delivered by ASEA-Atom with the capacity of 615 MW each. The
Barsebäck NPP units 1 and 2 where decommissioned through governmental
decisions November 30 1999 and May 31 2005 respectively.
Even if the shutdown dates for the two units differ, it is probable that the
dismantling of Barsebäck 1 and 2 will be conducted as a joint project. This,
according to the present time-table, will mean that a pilot project will be initiated
around 2015 and dismantling of Barsebäck would start sometime around 2020.
At the moment, extensive planning activities are carried out. One major part of
the demolition of the plant is the dismantling of the reactor pressure vessels
(RPVs) and its internals. There exist two major optional strategies for
dismantling of a RPV and its internals;
− Segmentation and cutup of the RPV and the RPV internals at plant.
− One-piece removal of the whole RPV, with or without the RPV internals.
The study presented here covers the examination and planning of a one-piece
removal of the RPVs at Barsebäck Unit 1 & 2, including the RPV internals.
In Sweden, it is the responsibility of the plant owner to plan and carry out the
dismantling of the decommissioned NPP, and to restore the site. Management,
transportation, intermediate storage and disposal of spent fuel and radioactive
waste from the Swedish nuclear power plants is, on the other hand, the
responsibility of the Swedish Nuclear Fuel and Waste Management Co. There is
at the moment two ongoing investigations in Sweden regarding the one-piece
removal of RPVs. The first one is initiated by Barsebäck Kraft AB, and carried
out by Scanscot Technology AB, regarding the detailed planning of the one-
piece removal of the two RPVs at the deommissioned Barsebäck NPP, as a basis

415
9. Waste Management, Fuel Cycle Facilities and Decommissioning

for a future management decision regarding what strategy to apply for dismantling
of the RPVs and its internals. The scope of this project covers the removal of the
RPV from the reactor containment, and its transportation to the quay at site, for
further transportation by boat. The second project is managed by the Swedish
Nuclear Fuel and Waste Management Co to from a more general perspective
investigate the possibility and cost for one-piece removal of RPVs in Sweden,
including the transportation and storage of the RPVs.
In a pilot study [1], six different alternatives for one-piece removal were studied,
and the pros and cons listed for each of them. The outcome of the pilot study was an
identification of three methods as the most promising ones (see Figure 1);
a) Lifting with a crane.
b) Removal in horizontal position using tower gantry.
c) Lowering inside the reactor containment.
These techniques has now been studied in detail, including technical lifting
aspects, necessary building modifications and demolition, and radiological
consequences, all ending up in cost estimations for each of the three alternatives
([2], [3]). Based on the outcome of this study, the preferred one-piece removal
technique will be identified.

a) b) c)

Figure 1. Decommissioned Barsebäck NPP Unit 1 and 2. Investigated reactor pressure

vessel one-piece removal techniques (taken from [2]); a) Lifting with crane, b) Removal in
horizontal position using tower gantry, c) Lowering inside the reactor containment.

416
 9. Waste Management, Fuel Cycle Facilities and Decommissioning

References
1. Scanscot Technology Report 07202/R-03, Barsebäck Unit 1 and 2 One-piece removal
of reactor pressure vessel – Pilot study (in Swedish), 2007.

2. Scanscot Technology Report 07202/R-04, Barsebäck Unit 1 and 2 One-piece removal

of reactor pressure vessel – Detailed study (in Swedish), Draft 2008 (to be
finalized March 2009).

3. ALARA Engineering Report 08-0039R, B1/2 – Protection against radiation, and radiology,
for one-piece removal of reactor pressure vessel (in Swedish), 2008.

417
9. Waste Management, Fuel Cycle Facilities and Decommissioning

Stability analysis of storage of spent fuel

in stack of trays in pool (9-2604)

O.P. Singh1, P.P. Karkhanis1, K. Agarwal1, K.N.S. Nair1, Dr. G.R. Reddy2
Bhabha Atomic Research Centre
Trombay, Mumbai, India
1
Nuclear Recycle Group
2
Reactor Safety Division

The spent fuel bundles from PHWRs are stored under water in spent fuel storage
facilities in stacks of 20 trays inside the water depth of 6–8 meters.
All the safety systems, structures and components of storage facility are required
to meet OBE level of earthquake (0.1 g). A stability analysis of stack of trays of
fuel bundles in seismic condition has been carried out for 20, 25 & 30 trays
using transient FEM analysis based on time history of pool and discussed in this
report. These results are being verified with shake table tests using dummy fuel
bundles.
A finite element model of stack of spent fuel trays has been developed to
represent the behavior under seismic condition. The models are created for each
individual tray and spent fuel bundles are modeled as a lumped mass in tray FE
model. The interlock between trays has been simulated using combination (i.e.
spring + Gap) element. The contact element has been used in FE model to
capture the lifting and sliding motion between trays and the same has been used
for contact between resting surface and bottom most tray of stack. A time history
analysis was performed for stack of 20, 25, and 30 trays with different friction of
coefficient (0.1, 0.2 & 0.3) to verify its overall stability against turning and
sliding under seismic event. These stacks of trays are found stable under the
designed level of earthquake.

418
 10. Challenges of New Reactors

10. Challenges of New Reactors

Design and construction programs for advanced, generation IV and

fusion reactors. Performance and behavior of metallic and non-metallic
materials and components in high temperature environment. Thermal
hydromechanics. Regulatory framework, safety research and development
of codes and standards. Passive safety features. Proliferation-resistant
design.

419
10. Challenges of New Reactors

420
 10. Challenges of New Reactors

The challenge of nuclear reactor structural

materials for Generation IV Nuclear
Energy Systems (10-1586)

Li Chengliang
China Nuclear Power Design Company Ltd. (Shengzhen)
Shanghai Branch, Shanghai 200030, China

Generation IV Nuclear Energy Systems have been studied in the framework of

the “Generation IV” International Forum, there can be little doubt that nuclear
reactor structural materials technology, including: Selection, development and
qualification, is one of the key issues to success of Gen IV Nuclear Energy
Systems. By striving to meet this challenge, which is beyond the experience of
the current nuclear power plants, the belt zone structural materials are required
good resistance to irradiation damage, high thermal stress capacity, excellent
resistance encompassing stress-corrosion cracking, and highly predictable
responses to extreme levels, compatibility with Heat-Transfer media and other
materials, very long-term stability enhanced in the system, adequate resources and
easy fabrication as well as weld-ability, etc, to guarantee the security and integrity of
the pressure boundary. This paper analyses and compares several materials under
active consideration performance requirements, and also discusses candidate
materials for use in different reactor components, which include various ferritic-
martensitic steels, advanced austenitic stainless steels, nickel-base super-alloys,
oxide-dispersion strengthened alloys, refractory alloys and etc. It is demonstrated
that new materials, such as metals, carbides, nitrides, oxide, novel alloy, solid
solutions or composites focuses on higher stability and better mechanical
performances. In addition, the emergence of new international structural materials
initiatives and their potential roles will be described.

421
10. Challenges of New Reactors

Design considerations for developing a

steam generator for integral modular
reactor SMART (10-1589)

Suhn Choi, Dong-ok Kim, Han-ok Kang, Keung-koo Kim

Korea Atomic Energy Research Institute
1045 Daedukdaero, Yuseong, Daejeon, Korea
e-mail: [email protected]

SMART (System-integrated Modular Advanced ReacTor) is a small sized

integral type PWR with a sensible mixture of new innovative design features and
proven technologies aimed at highly enhanced safety and improved economics.
SMART is applicable to an energy source for electricity generation and seawater
desalination, promising a new era of nuclear energy utilization only with small-
sized reactors.
The steam generator is one of the key and unique components to attain the
arrangement of integral feature. SMART has eight identical SG cassettes located
in the annular space between the reactor pressure vessel and the core support
barrel. Each steam generator cassette is of once-through design with a number of
helically coiled tubes where the feed-water flows upward producing superheated
steam, and the primary reactor coolant flows downward in the shell side. The
steam generator is also used as the heat exchanger for the passive residual heat
removal system.
The design of the steam generator cassette shall meet the requirements for
thermal hydraulic performance, arrangement and size, water chemistry and
materials, structural integrity including in-service inspection, operation and
maintenance, and flow induced vibration.
In this paper, the design requirements developed for SMART steam generator
cassette are then introduced, and the various designs of the SMART steam
generator along the development stage are summarized. Also the design details,
rationale, and some technically unresolved issues with their remedial ideas are
given and reviewed. The unresolved issues given in this paper seem to be
common to the integral reactor, thus their remedial ideas can be useful for the
development of a modular type steam generator cassette for the integral reactor.

422
 10. Challenges of New Reactors

Dynamic analysis methodology for stacked

graphite fuel blocks of a VHTR using
a commercial structural analysis
code (10-1678)
Dong-Ok Kim, Woo-Seok Choi, Jae-Man Noh
Nuclear System Development, Korea Atomic Energy Research Institute
1045 Daedeok Street, Yuseong-gu, Daejeon 305-353, Korea, e-mail:
[email protected]

The gas-cooled reactor systems which are currently considered as new candidates
for a very high temperature heat source for industries and hydrogen production
plants have many columns of stacked graphite blocks such as reflector blocks, fuel
blocks, and core support blocks in its reactor internal structures. An earthquake
loading on the stacked block columns causes rocking responses of the blocks and
solid impacts between them, and may lead to structural integrity problems, because
the blocks are not fully constrained and have gaps between neighboring blocks.
The dynamic analysis of block structures has a long history. In the historically
early stages of the structural and dynamic analysis of the stacked graphite fuel
blocks, the special computers of high computing power with the dedicated computer
programs were needed for the analyses to make short the computational time and
reduce the cost. In 1975, T.H. Lee presented a methodology for analyzing the
nonlinear response of a column of stacked prismatic fuel blocks for GT-MHR
[1]. In 1979 T. Ikushima and T. Nakazawa presented their work results on a seismic
analysis of a column of stacked prismatic fuel blocks for HTTR [2]. They made
and used the dedicated computer programs for the analyses. Figure 1 shows the
schematics of the models they developed.

Figure 1. Schematics of the classical models for the stacked graphite fuel blocks.

At the present days, the computing power of personal computers has been
remarkably increased and the commercial codes for structural analyses provide

423
10. Challenges of New Reactors

many useful modeling procedures and analysis options. The purpose of this paper
is to introduce the finite element models and the dynamic analysis examples of the
stacked graphite fuel blocks of a prismatic type gas-cooled reactor performed on a
personal computer using the commercial structural analysis code, ABAQUS.
Figure 2 and Figure 2 show examples for the proposed finite element models and
the seismic response results of the single and multi stacked graphite blocks.

Figure 2. A FE model and the dynamic response of a single simple block on a moving floor.

Figure 3. A FE model and the dynamic response of two columns of multi stacked blocks
on a moving floor with vertical walls.

Few attentions have been paid to the dynamic analysis methodology for the stacked
block structures using commercial codes and useful example studies reported are
rare. This study shows that the commercial code can be very useful for the dynamic
analysis of a reactor core internal structure consisting of the stacked graphite blocks.

References
1. Lee, T.H. 1975. Nonlinear Dynamic Analysis of a Stacked Fuel Column Subjected to
Boundary Motion. Nuclear Engineering and Design, Vol. 32, pp. 337–350.
2. Ikushima, T., Nakazawa, T. 1979. A Seismic Analysis Method for a Block Column
Gas-Cooled Reactor core. Nuclear Engineering and Design, Vol. 55, pp. 331–342.

424
 10. Challenges of New Reactors

RCC-MR 07 code: specificities and recent

developments (10-1686)

D. Bonne1, O. Gelineau1, M. Sperandio2, B. Drubay3

1
Areva NP, Lyon, France
2
Secretary of Afcen RCC-MR Committee, Lyon, France
3
CEA Saclay, Gif-sur-Yvette, France

The fourth edition of RCC-MR code has been issued on October 2007 by
AFCEN (Association Française pour les règles de Conception et de Construction
des Matériels des Chaudières Electro-nucléaires) and results of an important
work by AREVA NP and CEA to develop and improve design and construction
rules of the previous edition.
The improvements and new developments added in the new version of the
RCC-MR differ from the last 2002 edition by an enlargement of the scope of the
code not only applicable to mechanical equipments in fast breeder reactors
working at high temperatures but also to the ITER vacuum vessel, and other
nuclear components.
The last evolutions of the code are summed up as follow:
1. In the field of design rules, improvement of defect assessment rules and
the creep-fatigue rules for shells and pipes, and extension of the scope of
the code concerning bolts,
2. In relation with the development program of the ITER vacuum vessel,
introduction of a new quality class for the box type structures and a
specific appendix dealing with fabrication requirements of the ITER
vacuum vessel,
3. Modification or addition of requirements in accordance with the European
Pressure Equipment Directive with in particular the replacement of
French standards by European ones and as far as possible by harmonized
European standards.

425
10. Challenges of New Reactors

FE analysis of ITER 40º vacuum vessel

sector and stress assessment according to
French nuclear code RCC-MR (10-1705)

Julio A. Guirao Guijarro, Angel Bayón Villajos,

Joaquín Polo Ruiz, Lawrence Jones
Numerical Analysis TEChnologies (NATEC INGENIEROS)
C/ Marqués de San Esteban 9, 4th floor D, 33206 Gijón, Spain
e-mail: [email protected]
IBERDROLA INGENIERÍA, División de Generación Nuclear, Av/ Manoteras
20 Edf. C, 2nd floor, 28050 Madrid, Spain, e-mail: [email protected]
IBERDROLA INGENIERÍA, División de Generación Nuclear, Av/ Manoteras
20 Edf. C, 2nd floor, 28050 Madrid, Spain, e-mail: [email protected]
Fusion For Energy (F4E), VV Group, C/ Josep Pla 2, Torres Diagonal Litoral
B3, 9th floor, 08019 Barcelona, Spain, e-mail: [email protected]

The progressive increase in computational resources and the recent extensive

Finite Element Method (FEM) development has enabled the use of these codes
in the mechanical calculation of pressure vessel and equipment (DBA, Design
By Analysis) and in nuclear calculation codes due to the high level of accuracy
and reliability. The analysis procedures used for the checking and treatment of
stresses are very similar for the most widely used Nuclear codes such as are
described in ASME Sec. III and RCC-MR, version 2007, which is the code used
in the ITER Vacuum Vessel. A linear analysis of the model is used in the FE
stress assessment technique and a further stress check is carried out based on
stress categorization, which is required for a further assessment by comparison
with allowable limits. This is a complex and non-intuitive process, especially so
in the current analysis since this procedure is more complex compared to
standard pressure equipment due to the unusual geometry, which includes many
gross structural discontinuities and zones with potentially high stress concentration
effects. To assist in the process of separation of the membrane, bending and peak
stresses, “ad-hoc” tools are programmed in the FEM software (ANSYS). In
order to overcome the difficulties caused by the stress categorization of
secondary stresses, which in RCC-MR, RB-3224.35 are defined as “the fraction
of the total stress which can disappear as a result of small scale permanent
deformation minus the peak stress”, sub-modelling techniques with elastoplastic
material behaviour laws are used to accurately quantify when the zones
overstressed are small enough to disappear as a consequence of localized plastic
permanent deformations so that the stresses may be considered as secondary.
This paper presents an approach based on the construction of a detailed 3D

426
 10. Challenges of New Reactors

model of the ITER Vacuum Vessel and its subsequent FEM analysis, utilising
the RCC-MR Ed. 2007 code.
Considering that new reactors have a complex geometry which is in some
cases very different to conventional reactors (pressurized double shell, housings,
vacuum inside…), the new analysis strategies allow facing the mechanical
design problems derived from these complex geometries.

427
10. Challenges of New Reactors

A high temperature gas loop to simulate

VHTR and nuclear hydrogen production
system (10-1870)

Yong Wan Kim1, Chan Soo Kim2, Sung Deok Hong2,

Won Jae Lee2, Jonghwa Chang2
1
Nuclear Hydrogen Technology Development Division, Korea Atomic Energy
Research Institute, Daejeon, Korea, e-mail: [email protected]
2
Korea Atomic Energy Research Institute, Daejeon, Korea

Introduction
Very high temperature gas cooled technology and nuclear hydrogen production
technology are being developed in KAERI for the nuclear hydrogen production
system. The outlet temperature of the high temperature gas cooled reactor developed
so far ranges 750 to 900oC. However, to produce hydrogen with economical
efficiency, the coolant outlet temperature outlet temperature of VHTR should
exceed 950oC. In this paper, a development of small scale gas loop to investigate
coupling of the VHTR and hydrogen production system is introduced. A process
heat exchanger developed which connects VHTR and hydrogen production
system is tested in the gas loop. The whole process of gas loop development
from the design to the construction is introduced. Also, some of the test result of
heat exchanger structural integrity analysis and test is discussed.

Small nitrogen gas loop

Primary loop

The primary loop simulates VHTR heating system. The design pressure and the
design temperature of the primary loop are 6 MPa and 1000oC respectively. The
process gas of the primary loop is nitrogen gas. Primary loop consists of pre-
heater, main heater, hot gas duct, gas circulator, and isolation valves. Most of the
pipe is thermally insulated to prevent over heating at the outer pipe structure.

Secondary loop

The secondary loop simulates the sulfuric acid hydrogen production process. A
sulfuric acid (H2SO4) loop is an open loop and consists of a H2SO4 storage tank,
a H2SO4 feed pump, a pre-heater, a heat exchanger (evaporator), a PHE, a
separator, a SO2 trap, and a H2SO4 collector (Figure 1). Cold 98% H2SO4 is

428
 10. Challenges of New Reactors

superheated to 500°C. In a superheating process, H2SO4 decomposes into H2SO4

and SO3. In the PHE, some fraction of the SO3 is dissolved into SO3 and O2. The
toxic SO3 is separated in the separator and the sulfur dioxide in the mixture gas
is removed in the NaOH trapping system.

Process heat exchanger

Process heat exchanger is a coupling components between VHTR system and
nuclear hydrogen system. A concept has been developed and structural analysis
utilizing finite element analysis has been done before the experiment. The finite
element analysis results have shown that the process heat exchanger can
withstand a considerable amount of pressure difference between loops. This heat
exchanger was tested by circulating sulfur-acid in the secondary flow channel.

H 2 SO4
De c om p o ser

Sp a ce
Pr e- h ea t er F or
T est
Sec t ion
Se pa r at or
H ig h - T
C oo ler

C o o ler

Sc r ub b er
Pum p s
Pu m p s

Figure 1. 3D Gas Loop Model. Figure 2. Gas Loop of secondary side.

Summary
Small scale nitrogen gas loop was designed and constructed for the test of
process heat exchanger at the elevated temperature. The development process
including the design and the fabrication of the test loop is presented. Process
heat exchanger was tested in sulfuric acid environment. As a next step, middle
size He loop is being constructed in KAERI from 2009.

429
10. Challenges of New Reactors

References
1. J. Chang et al. A Study of the Nuclear Hydrogen Production Demonstration Plant.
Nuclear Engineering and Technology, Vol. 39, No. 2, pp. 111–122, 2007.

2. Y.W. Kim et al. A design of sulfuric-Acid Decomposer for Nuclear Hydrogen Production
System Utilizing Very High Temperature Gas Cooled Reactor. Proceedings of
SMiRT 19, Toronto, August 2007.

3. S.D. Hong et al. A High pressure and High Temperature Sulfuric Acid Experimental
System. KNS ’08 Autumn proceedings, 2008.

430
 10. Challenges of New Reactors

Impact of engineered safety features on

AHWR containment (10-1881)

I. Thangamani, Vishnu Verma, R.K. Singh, A.K. Ghosh

Reactor Safety Division,
Bhabha Atomic Research Centre, Mumbai, India
e-mail: [email protected]

The proposed Advanced Heavy Water Reactor (AHWR) employs a double

containment envelope viz, a primary and secondary containment, which together
act as last physical barrier to the release of radioactivity following an accident.
The primary containment is completely surrounded by the secondary containment.
As a part of the Probabilistic Safety Assessment (PSA), it was required to
perform studies pertaining to effect of Engineered Safety Features (ESFs) on
containment pressure, temperature and flows and leakages from the AHWR
containment.
The containment building of AHWR is proposed to provide with Engineered
Safety Features (ESFs) such as Containment coolers, Primary containment
controlled discharge (PCCD), Primary containment filtration & pump back
system (PCFPB) and Secondary containment filtration, recirculation and purge
system (SCFRP). ESFs are provided for post-accident management such as
containment clean up, containment depressurization etc. to minimize the amount
of activity release to the environment during an accident. Activity release
consists of ground and stack release and it depends upon long term containment
pressure, temperature transients and activity concentration in various volumes of
the containment. The containment pressure and temperature depends upon the
discharge rate of high enthalpy fluid into the containment. Containment pressure
is also influenced by leakage rate from the containment, steam condensation on
containment walls and status of ESFs as well. To study the effect of ESFs on
flows and leakages from the AHWR containment, an analysis was carried out
based on availability and non availability of various ESFs.
Thermal hydraulic transient analysis was carried out using in-house
containment thermal hydraulics code ‘CONTRAN’. Modules for simulating the
engineered safety features were incorporated with this code. Deposition of
aerosols in containment volumes were calculated using NAUA code. For the
purpose of analysis, a three volume configuration of entire AHWR containment
with Gravity Driven Water Pool (GDWP) is considered. The primary
containment is divided into two accident based volumes viz. V1 volume and V2
volume housing the high enthalpy and low enthalpy systems respectively. The
region between primary and secondary containment is considered as third volume.
This paper presents the results of analysis of a postulated LOCA case, 200%
RIH break with failure of shutdown systems (1 & 2). Blowdown mass and

431
10. Challenges of New Reactors

energy discharged into containment were given as input to CONTRAN code for
estimating the pressure, temperature, junction flow rates etc. in all volumes for
72 hrs, which were then supplied, along with activity release rates, as inputs to
NAUA code for calculating deposition rate in the containment volumes. Finally,
the blowdown mass, energy data and activity deposition rate (calculated using
NAUA code) were then given as inputs to CONTRAN code and recalculated the
pressure, temperature, activity concentration in all volumes and leakages from
containment volumes.
Analysis was carried out for a number of cases, postulated based on
availability/unavailability of ESFs. Activity released out of containment were
obtained for all the cases in terms of ground level, stack level and total activity
release from containment for 72 hours from the initiation of the accident.
This paper highlights the importance of operation of ESF in reducing the
activity release to the environment.

432
 10. Challenges of New Reactors

Preliminary analysis of the structural

effects due to dynamic loads of the
isolated next generation lead
cooled reactor (10-1887)

Rosa Lo Frano, Giuseppe Forasassi

Department of Mechanical, Nuclear and Production Engineering, University of PISA
via Diotisalvi, n°2-56126 Pisa, Italy
Tel: +39-050- 836689; Fax: +39-050-836665
e-mail: [email protected]

The main purpose of this preliminary study deals with the evaluation of the
structural effects due to the dynamic loads exerted and propagated through the
lead coolant during a safety shut down earthquake with reference, as an example, to
the isolated ELSY system configuration (CEE-7 Framework Project).
Seismic base isolation is increasingly used to protect structures and their
contents against dangerous ground motions as well as mitigate the structural
effects, on the internals walls and reactor components of the induced dynamic
load and of the coupling between coolant and vessel.
An adequate predictive numerical modelling, by means a 3-D finite element
model, was set up and a non-linear approach was used for the foreseen structural
preliminary analyses and simulations of the plant and internals behaviours, in
order to describe the interactions among the different subsystems.
Moreover the fluid-structure interaction problem, due to the high density of
the retained primary coolant, has received a particular attention in relation to the
possible hydrodynamic interaction, between lead and the surrounding internals,
as well as the sloshing wave motion (the lead may be accelerated and can impact
on the structures walls) that may significantly influence the stress level in the
reactor pressure vessel (RPV). As for the seismic analysis, the isolation systems
may influence the seismic capacity of as-built structure to reduce the intensity of
the propagated seismic loads.
Numerical results are presented and discussed highlighting the importance of
the fluid-structure interaction effects as well as the isolation technique
effectiveness, which is expected to be effective in raising the reliability of
internals and vessel structures, during an earthquake event.

433
10. Challenges of New Reactors

Radiotoxicity perspectives for different

ELSY working hypotheses: towards a
sustainable fuel cycle (10-1905)

Barbara Vezzoni1, Giacomo Grasso2, Carlo Petrovich3, Massimo Sarotto3, Carlo

Artioli3, Eleonora Bomboni1, Sara Bortot4, Rasha Ghazy4, Giuseppe Forasassi1,
Marco Sumini2, Marco Ricotti4
1
Faculty of Engineering, University of Pisa
via Diotisavi 2, 56126 Pisa, Italy e-mails: [email protected],
[email protected], [email protected]
2
Nuclear Engineering Laboratory (LIN) of Montecuccolino
University of Bologna, v. dei Colli 16, 40136 Bologna, Italy
e-mails: [email protected], [email protected]
3
Italian National Agency for New Technologies, Energy and the Environment
(ENEA), v. Martiri di Monte Sole 4, 40126 Bologna, Italy
e-mails: [email protected], [email protected],
[email protected]
4
Department of Energy Nuclear Division (CeSNEF)
Politecnico of Milan, v. La Masa 34, 20156 Milan, Italy
e-mails: [email protected], [email protected], [email protected]

The acceptability of nuclear energy in part arises from the capability of handling
its own wastes until a final safe disposal. In particular, the spent fuel represents
the most problematic waste to be managed: the presence of Long Lived
Radioisotopes (i.e. Minor Actinides) imposes binding restrictions regarding the
waste management and its disposal in geological repositories.
Therefore, the radiologic load of the spent fuel is one of the main drawbacks
of nuclear energy production.
In the present climate of nuclear renaissance new solutions have to be pursued
in order to guarantee the full sustainability of the nuclear fuel cycle. Within such
context, the formalization of the Generation-IV prescriptions is intended to guide
the design of new concept reactors capable to overcome the drawbacks of the
present generation systems, mainly by implementing the fast spectrum features.
According to this, the present study aims at demonstrating the possibility to
reduce the spent fuel radiotoxicity, by analyzing different fuel cycle hypotheses
for the European Lead-cooled System (ELSY), the 600 MWe Generation-IV
Lead cooled Fast Reactor under investigation in Europe within the 6th
EURATOM Framework Programme.
The up to date ELSY reference configuration (characterized by square,
wrapper-less Fuel Assemblies) has been modeled and some realistic operating
conditions have been then simulated using the MCNPX transport code.

434
 10. Challenges of New Reactors

A standard operation for a core designed to achieve unitary Breeding Ratio

was studied at first. Two different desirable operative conditions are then
analyzed, concerning the extension of the burn up to 100 GWd/t and the
adoption of the “adiabatic” core concept. The latter configuration refers to a
system where the burn up of Minor Actinides equals their build up by loading
the respective equilibrium concentrations in the fresh fuel.
For each of the conditions indicated the wastes radiologic load has been calculated
by assuming a typical reprocessing upon a 2 years aged spent fuel with 0.1%
losses. The evolution of the resulting waste radiotoxicity (evaluated using
ICRP72 coefficients) has been then analyzed and compared with the one of
present Generation-II reactors spent nuclear fuel (i.e. typical 1000 MWe PWR).
A preliminary sensitivity analysis of the results has been also performed by
adopting different burn up codes (CINDER90 in MCNPX v.2.6.0, ORIGEN
v.2.2 with MCNP5 and ERANOS) and cross section libraries (JEFF3.1 and
ENDF/B-VII).
A comparative analysis upon the collected data showed a significant reduction
in the long term activity and the corresponding radiotoxicity, suggesting the
viability of the adiabatic core configuration as the most interesting solution for
an actual reduction of both the High Level Waste volumes and the associated
radiotoxicity.

435
10. Challenges of New Reactors

Material challenges of the new advanced

gas cooled systems (10-2038)

Derek Buckthorpe
AMEC, Booths Park, Knutsford, Cheshire, UK
e-mail: [email protected]

Introduction/background
For new Generation IV gas cooled systems such as the very high temperature
reactor (VHTR) and gas cooled fast reactor (GCFR) industrially established
materials are at or near their limits in certain applications and new materials
and/or an understanding of the behaviour of existing materials under more severe
environmental reactor conditions will be required. For metals, ODS alloys,
ferritic-martensitic steels, and additional super-alloys offer some solutions but
require, significant R & D in terms of their properties and behaviour under
component conditions. Gas Cooled Generation IV systems may also require the
deployment of non-metallic materials (e.g. high-temperature fibrous insulation
in the reactor or power conversion systems, composite materials as alternatives
to metals for in-core components, ceramic materials for use at high temperatures
(950–1000°C) for components such as the Heat exchanger. Such materials are
difficult to bond, pressure form, and machine. The codification and understanding
of the behaviour of these materials with respect to design and operation is very
much in its infancy.

Purpose and results

Previous programme details and intermediate results have been presented for the
HTR system [1]. This paper reviews the present understanding and the progress
made in interpreting and establishing property and behavioural information on
metallic and carbon based materials for the VHTR and GCFR systems in the
RAPHAEL and EXTREMAT EU 6th Framework projects. The work will
summarize the main findings of the experimental programmes undertaken,
including tests to develop Mod 9Cr 1Mo steel as a vessel steel, irradiation tests
on graphite for the VHTR core, progress in understanding composites and
ceramics for key component applications. Results will be presented for joints
and welds, irradiation and corrosion and microstructural modeling. Cross cut
information from similar situations in the Fusion reactor will also be included.

436
 10. Challenges of New Reactors

Summary/conclusions
The work will conclude on the main achievement so far and on the needs in terms
of future R & D activities in Europe for these systems. The recommendations will
take into account the expected input of information from the Generation IV
International programmes and from ongoing national programmes.

Reference
1. D. Buckthorpe, R. Couturier, B. van der Schaaf, J. van der Laan, B. Riou, H. Rantala,
P. Ennis, G. Haag, K. Kuehn, A. Buenaventura, B.-C. Friedrich. Progress of
HTR-M projects for the HTR, paper S443, SMiRT17, Prague, Czech Republic,
August 2003.

437
10. Challenges of New Reactors

The stress assessment of reflector graphite

bricks in high-temperature gas-cooled
reactor (10-2044)

Libin Sun, Zhengsheng Zhang

Faculty of The Institute of Nuclear and New Energy Technology (INET)
Tsinghua University, Room 313, Energy Science Building B, Beijing 100084, China
e-mail: [email protected]

This article briefly describes the structural arrangement and features of the
graphite internals of the 10 MW high-temperature gas-cooled reactor-test
module (HTR-10) before the stress and deformation of the side reflector graphite
bricks are analyzed under normal and accidental conditions for the whole
lifetime. The maximal stress and deformation of the graphite bricks are caused
by the fast neutron irradiation and the high temperature. The safety assessment
of the stresses in the graphite bricks has carried out and the deformation can not
affect the graphite internals function, it can be drawn that the stress and
deformation of the graphite bricks will not affect the reactor safety during the 20
years lifetime.

438
 10. Challenges of New Reactors

Study on steam generator helical tube

integrity assessment of HTR (10-2051)

Dong Jianling
INET, Tsinghua University
Beijing 100084, P.R. of China
e-mail: [email protected]

In a nuclear power plant, the steam generator (SG) heat transfer tubes account
for almost 80% of the primary loop pressure retaining boundary. However, they
are also the weakest link in the loop.
In pressurized water reactor (PWR), heat transfer tube rupture can result in the
bypass of the containment, which will provide radioactive fission materials in
the primary loop coolant a direct pathway to communicate with atmosphere,
resulting in a LOCA (loss of coolant accident). Multiple tube ruptures could lead
to a meltdown of the nuclear reactor.
In high temperature gas-cooled reactor (HTR) with steam cycle, steam
generator connects and isolates the primary radioactive helium loop from the
secondary non-radioactive water and steam loop. The pressure of the secondary
side is higher than the helium pressure of the primary side. Heat transfer tube
rupture will cause water and steam in the SG to flow into the primary loop,
mixing with the helium flow in the reactor core. The increase of steam
concentration in the pebble bed enhances the reactor core moderation, which
introduces positive reactivity and increases the reactor power and temperature.
When a large amount of steam leaks into the primary loop, the iodine deposit on
the steam generator is washed into helium coolant and the incoming steam reacts
with the fuel elements and the graphite components at high temperature, which
produces water gas. This directly affects the integrity of the graphite
components, which release radioactivity to the environment. The expanding gas
in the superheated reactor core will increase the primary loop pressure. When the
pressure in the reactor exceeds the set pressure of the safety valve in the pressure
relief system of the primary loop, the safety valve will open and release
radioactive products to the reactor building and the atmosphere. In reference [1],
the analysis of a double-ended guillotine break (DEGB) of a HTR-10 SG tube
showed that the safety valves in the pressure relief system will open and
discharge a gas volume equal to 1/4 of the total helium inventory in the primary
loop. Therefore, the integrity of SG tube is also very important to the safety of HTR.
In nuclear industry, mainly two kinds of steam generator are used, that is
U-tube steam generator and helical coil tube steam generator. The former is used
in PWR with loop-type layout reactor coolant system, such as AP1000 and EPR,
while the later finds its application in HTR, such as HTR-10, and PWR with
integral reactor coolant system, such as SMART and IRIS. The HTR-10 SG heat

439
10. Challenges of New Reactors

transfer tube is a typical helical coil tube. This paper takes the helical tube of
HTR-10 steam generator as an example to present some research results on the
tube integrity assessment of the helical coil tube steam generator.

References
1. Gao Zuying, Jiang Zhiqiang, Li Baoyan, Wang Chunyun., 1993. Accident Analysis of
10MW High Temperature Gas-Cooled Test Reactor. Nuclear Science and
Engineering, 13(4), pp. 77–89. (In Chinese)

440
 10. Challenges of New Reactors

Researches for development of regional

energy reactor, REX-10 (10-2070)

Jong-Won Kim1, Yeon-Gun Lee, Hyeong-Min Joo, Byeong-Ill Jang,

Sung-Won Lim, Yong-Hee Choi, Moo-Hwan Kim, Goon-Cherl Park2
1
Seoul National University, Seoul, Korea Republic, e-mail: [email protected]
2
Seoul National University, Seoul, Korea Republic, e-mail: [email protected]

Recently, today's global pattern of energy supply is not sustainable. The

provision of affordable energy services is a fundamental prerequisite for
economic growth and development. In addition, the environmental problems
such as energy crisis, global warming and acid rain are issued and more demand
for reliable electricity supply increases. As one of the realistic solutions, the
extension of the peaceful uses of nuclear energy has been suggested. Small and
medium nuclear reactors with non-electric applications arise as an alternative
energy source. The main non-electric applications are defined as district heating,
desalination (of sea, brackish and waste water), industrial heat supply, ship
propulsion and the energy supply for spacecraft. The small and medium reactors
are under development in several advanced countries with goals of low capital
costs, short construction periods, high performance and enhanced safety. RERI
(Regional Energy Research Institute for Next Generation) introduces a new
paradigm for energy supply system. REX-10 which is stable small nuclear
reactor with thorium fuel for power generation and nuclear district heating has
been developed. The design objectives of REX-10 are inherent safety, non-
proliferation and economical efficiency. The newly designed REX-10 has been
developed to maintain system safety in order to be placed in a densely populated
region, island, etc. For high safety, natural circulation, pool-type vessel and low
operation pressure are introduced. In addition, the thorium fuel cycle with 20
year lifetime without exchanging fuel is considered for the sake of the non-
proliferation. Moreover, the economical efficiency is ensured by the unmanned
automatic control. The system pressure and capacity are determined properly for
regional energy reactor. The operation pressure is 2.0 MPa and the thermal
power is 10 MWth. The major research activities for REX-10 design are natural
circulation, steam-gas pressurizer and thorium fuel cycle.

441
10. Challenges of New Reactors

Summary of SMIRT20 preconference topical

workshop – Identifying structural issues in
advanced reactors (10-2504)

William Richins, Stephen Novascone, Cheryl O’Brien

Idaho National Laboratory, US Dept. of Energy, Idaho Falls, Idaho, USA
e-mail: [email protected]

The Idaho National Laboratory (INL, USA) and IASMiRT sponsored an

international forum Nov 5–6, 2008 in Porvoo, Finland for nuclear industry,
academic, and regulatory representatives to identify structural issues in current
and future advanced reactor design, especially for extreme conditions and
external threats. The purpose of this Topical Workshop was to articulate
research, engineering, and regulatory Code development needs. The topics addressed
by the Workshop were selected to address critical industry needs specific to
advanced reactor structures that have long lead times and can be the subject of future
SMiRT technical sessions. The topics were; 1) structural/materials needs for
extreme conditions and external threats in contemporary (Gen. III) and future
(Gen. IV and NGNP) advanced reactors and 2) calibrating simulation software
and methods that address topic 1 The workshop discussions and research needs
identified are presented.
The Workshop successfully produced interactive discussion on the two topics
resulting in a list of research and technology needs. It is recommended that
IASMiRT communicate the results of the discussion to industry and researchers
to encourage new ideas and projects. In addition, opportunities exist to retrieve
research reports and information that currently exists, and encourage more
international cooperation and collaboration. It is recommended that IASMiRT
continue with an off-year workshop series on select topics.

442
 10. Challenges of New Reactors

Development of high efficiency and high

capacity gas/gas heat exchanger for
gas-cooled reactors (10-2535)

Masanori Tanihira, Yasuyuki Miyoshi, Keiichi Nakashima & Ryoji Kishikawa

Mitsubishi Heavy Industries, Ltd.
16-5, Konan 2-Chome, Minato-ku, Tokyo 108-8215, Japan
e-mail: [email protected]

Helium gas cooled reactors are candidates for future main power supply from
viewpoints of economical efficiency, safety, operability and multi-purpose heat
utilization. For the practical application of gas cooled reactor, high efficiency
and high capacity (600 ~ 1000 MW) gas/gas heat exchanger is one of the most
important equipments. The compactness and the performance upgrading of the
equipment determine the feasibility (success or failure) of the whole plant. The
high surface area to unit volume ratios of a plate-fin type heat exchanger mean
that it is about twenty times smaller than the equivalent tubular heat exchanger.
Problems of severe service condition such as high temperature (650 ~ 950°C)
and high pressure (~ 7 MPa), however, have not been solved. In order to overcome
these problems, a development program of the high capacity compact gas/gas
heat exchanger has been carried out.
High heat exchange volume density is necessary for economical low-cost of
the heat exchanger. Narrow pitch fin is expected to give a plate-fin type heat
exchanger high heat exchange volume density and high structural strength.
The result of our study is summarized as follows.
1) The trial manufactures of the plate-fin type heat exchanger of the large
scale structure body with narrow 1.2 mm pitch fin from Hastelloy-X,
which is Ni-based alloy and can be used at 850°C and above, and from
stainless steel 316 were carried out for the first time in the world.
2) The strength characteristics of the ultra plate-fins became clear through
the various element examinations and structure examinations. Moreover
a practical stress analysis method was investigated on the basis of the
homogenization approach. A data-base for the strength design including
the fundamental deformation of the ultra offset plate-fins structure was
developed.

443
10. Challenges of New Reactors

3) Heat transfer characteristics and pressure loss were evaluated experimentally,

and heat transfer and flow analysis was practiced by simulating the
structure of actual equipment. It was confirmed that the plate-fin type
heat exchangers had the volume density more than 25 MW/m3 by these
results.
4) By studying and organizing design requirements of recuperators and
intermediate heat exchangers of the planned HTGR plant, numerical
evaluation of both the thermal performance and the structural strength
was carried out and recuperators and intermediate heat exchangers for
HTGR plant were designed.

Acknowledgments
This work shows the R&D results in 2004 to 2007 by Mitsubishi Heavy
Industries in the program of “The Technology Development of High Efficiency
and High Capacity Gas/Gas Heat Exchanger Necessary to the Practical
Application of Gas-cooled Reactor”, which is entrusted from Ministry of
Education, Culture, Sports, Science and Technology of Japan.

444
 10. Challenges of New Reactors

Computational assessment of the reactor

vessel cooling options in a prismatic
core VHTR (10-3139)

Min-Hwan Kim, Nam-il Tak, Wonjae Lee

Korea Atomic Energy Research Institute
1045 Daedeok-daero, Yuseong-gu, Daejeon, Korea
e-mail: [email protected]

The Very High Temperature Reactor (VHTR) has been selected for the Nuclear
Hydrogen Development and Demonstration (NHDD) project. For the NHDD
plant, the primary coolant inlet and outlet temperatures are considered to be 490
and 950oC, respectively. Due to its high operating temperature, the design of the
reactor pressure vessel (RPV) is one of the important issues in the NHDD
design. Both the SA508/533 steel and high-Cr steels (e.g. 9Cr-1Mo-V steels) are
expected to be candidate materials for the VHTR pressure vessel. Because of its
extensive experience base as an ASME Section III code-approved material for
Light Water Reactor, the SA508/533 steel has emerged as a strong candidate for
the RPV. In order to use this material, however, a design is needed to maintain
the RPV temperature below the ASME code limit, which is 371oC during normal
operation and 538oC for up to 1000 h during accident conditions.
In this study, three types of vessel cooling options for a prismatic core VHTR
to keep the RPV temperature below the normal operating limit are suggested. In
the first option, the coolant inlet flow is routed through riser channels in the
permanent side reflector (PSR), which is a base configuration of all three
designs. A vessel cooling system (VCS) supplying cold helium flow between the
RPV and the core barrel is added to cool down the RPV in case that the RPV
temperature is still higher than its limit. In the second option, external vessel
cooling is introduced with the modified inlet flow configuration. The cooling
fluid is air in the reactor cavity, outside of the RPV. Air blowers should be
installed around the bottom side of the RPV. The last option is to use insulation
material instead of the direct cooling of the RPV by internal cold helium flow or
external air flow. The location of insulator can be either on the inner surface of
the RPV or at the interface surface between the PSR and the core barrel.
The performances of the vessel cooling options are evaluated by using a
system thermo-fluid analysis code, GAMMA+, and a commercial computational
fluid dynamics code, CFX, during normal operation and accidents. The
GAMMA+ model includes the reactor coolant system, the reactor cavity, the
passive Reactor Cavity Cooling System (RCCS), and the VCS. The CFX code
was used to model in more detail a 1/54 sector corresponding to the region
associated with a single PSR riser channel, extending in the radial direction from

445
10. Challenges of New Reactors

the PSR to the RCCS downcomer wall. According to the results, the modified
inlet flow configuration with the VCS flow provides the most viable results. The
external cooling option does not ensure an effective cooling of the RPV. The
insulation option provides an effective temperature reduction of the RPV but
needs careful consideration in a view of the fuel safety margin during accidents.

446
 10. Challenges of New Reactors

Generation IV material issues –

case SCWR (10-3143)

Sami Penttilä, Aki Toivonen

VTT Technical Research Centre of Finland, Materials and Building
P.O. Box 1000, FIN-02044, VTT, Espoo, Finland
Tel: +358 40 5950338, Fax: +358 20 7227002
e-mail: [email protected]

Generation IV nuclear power concepts have become an active research topic all
over the world during the last 5 - 10 years. There are six concepts accepted by
the GenIV international forum (GIF) with the common aims to promote both
efficiency and safety of the technology. New concepts will offer attractive
features but at the same time they also bring new and demanding challenges e.g.
for the materials technology due to increased operating temperatures and
irradiation doses as well as more aggressive coolants and/or longer life time
expectations than of GenII and GenIII plants. In this paper an overview of the
material issues is given with special emphasis on supercritical light water reactor
concept (SCWR).
This paper reviews the performance of commercial candidate materials for
SCWR in-core applications focusing on the corrosion and stress corrosion
cracking issues (SCC). Within the FP6 program “HPLWR Phase 2” -project (High
Performance Light Water Reactor) general corrosion tests (i.e., oxidation rate
tests) and SCC susceptibility have been performed on selected iron and nickel
based alloys at 500°C and 650°C in supercritical water under the pressure of
25 MPa. The oxygen concentration of the inlet water was 0–150 ppb in all tests.
The oxidation behavior was studied using weight gain measurements, scanning
electron microscopy in connection with energy dispersive spectroscopy (SEM
and EDS, respectively) and X-ray diffractometry (XRD). The SCC tests were
slow strain rate tests (SSRT) performed using a step motor controlled loading
device. The samples were strained with a nominal rate of 3 × 10-7 s-1.
Ferritic–martensitic (F/M) steels containing chromium have generally good
resistance to stress corrosion cracking. However, they suffer from fast oxidation
in the SCW conditions. Austenitic stainless steels and Ni-based alloys have
better oxidation resistance but, on the other hand, are more susceptible to stress
corrosion cracking. SSRT test showed that 316 NG, 1.4970, 347 H and PM2000
are not susceptible to SCC at 500°C based on fracture surface examination, but the
experimental steel BGA4 showed a considerable susceptibility to intergranular
SCC. The austenitic stainless steels were generally observed to be SCC
susceptible at 650°C, which corresponds well with the data reported in literature.
The high chromium ODS (Oxide Dispersion Strengthened) steel PM2000 was
SCC resistant at both test temperatures.

447
10. Challenges of New Reactors

Alloys with high nickel content were not considered for the SCC studies
because Ni has a strong negative effect on neutronics of the reactor core.
Therefore, the present candidate materials for the core internals are austenitic
stainless steels and high chromium ODS alloys.

448
 10. Challenges of New Reactors

The Finnish Sustainable Energy (SusEn)

project on New Type Nuclear Reactors
(NETNUC) (10-3163)

Riitta Kyrki-Rajamäki, Lappeenranta University of Technology (LUT)

Rainer Salomaa, Helsinki University of Technology (TKK)
Timo Vanttola, Sami Penttilä, VTT Technical Research Centre of Finland
Liisa Heikinheimo (TVO)

Increased international attention has recently been devoted to reactor concepts

that differ essentially from the existing light water reactors. Basic processes of
these new concepts, known as Gen IV reactors, are fundamentally different from
those used today. Some new features may create new type safety challenges.
They also aim to push nuclear reactor technology to completely new regimes of
performance parameters, thus raising engineering challenges.
In order to allow Finland to benefit from these new technologies and influence
their development, it is necessary to join relevant international projects, develop
domestic expertise on critical technologies involved, and participate in ongoing
international efforts to develop safety requirements for them. The NETNUC
project is a multidisciplinary consortium to carry out basic research to generate
scientific knowledge needed for Gen IV reactors and to educate a new
generation of research scientists in the field. Five targeted tasks are carried out in
LUT, TKK and VTT.
Main research areas in LUT are the Super Critical Water Reactor (SCWR) and
gas-cooled reactors, the latter being a new field in LUT. Linking of the strong
thermal hydraulic background of LUT to the safety research of reactors
operating in supercritical pressure conditions has been studied. CFD calculations
have been carried out to simulate the BWR pool tests of LUT, which are relevant
also for SCWR. The research of gas-cooled reactors in LUT has focused on the
numerical modelling of coolant flow and heat transfer in the reactor core of a
pebble-bed type high-temperature gas-cooled reactor. There has been co-
operation with a research group in LUT that has experience in the modelling of
combustion processes and gas-phase.
Main research areas for TKK are thermal-hydraulics of SCWR and Monte
Carlo calculations on reactors utilizing thorium fuel. In addition to this the
different aspects of the fuel cycle have been analyzed. SCWR thermal hydraulics
has been investigated by simulations with APROS process simulation software.
Also co-operation with Kurchatov Institute, Russia has been initiated. Thorium
breeding in nuclear reactors has been investigated with MCNP simulations.
Several auxiliary programs for thorium fuel performance calculations with
FEMAXI-6 have been developed.

449
10. Challenges of New Reactors

Main research areas in VTT are fast reactors and SCWR concept. In 2008, the
first calculation system capable of analysing fast reactor concepts was taken into
use at VTT. The validation of this system was started as well. The work included
development of calculation methods by creating a new burnup calculation
method based on calculating matrix exponentials. In the field of supercritical
water cooled reactors, the participation in the EC HPLWR2 project was
continued with development of thermal hydraulics models and applying them to
the safety studies of the High Performance Light Water Reactor.
Understanding of corrosion phenomena of candidate materials under SCWR
(Supercritical Water Reactor) conditions necessitates a reliable experimental
testing of materials and therefore also a development of monitoring techniques
for the relevant conditions. The long term objective of this project is to perform
a state-of-art study that would serve as a guide for the selection of in-core
materials for the SCWR. One of the key performance indicators for the material
selection is the corrosion and oxidation behaviour of materials in SCWR
conditions. Therefore, in situ studies of the oxide films forming on the internal
component candidate materials in contact with supercritical water are needed. At
VTT, the Contact Electric Impedance (CEI) technique has been successfully
used for in situ characterization of the electrical and transport properties of the
corrosion layers both in LWR and SCWR conditions. In 2008, first step towards
employing a pneumatic servo-controlled bellows system for the oxidation
studies under SCWR conditions were performed. Calibrations and installations
of the double-bellows system into the supercritical autoclave have been
performed. Usability of the double-bellow system was verified up to 650°C and
25 MPa. The materials to be studied by CEI technique have been chosen based
on their general oxidation rates at SCW conditions.
In the future, it is likely, that process heat and steam from new type high-
temperature nuclear power plants is used in the integrated industry, such as bio-
refineries. It is important that also simulation tools support processes with multi-
phase chemistry. The feasibility of combining rigorous multi-phase chemistry
with dynamic process simulator, APROS, was evaluated at VTT. Three different
case processes were studied: fibre suspension and bleaching chemistry in pulp
and paper mills and boiler water chemistry. The multi-phase chemistry and
process simulation were successfully combined.
The national Finnish network GEN4FIN has continued its operation; the
results of NETNUC are disseminated through GEN4FIN to the Finnish industry.
VTT has also been active in the preparation of Strategic Research Agenda
(SRA) of SNETP. Also Nordic co-operation on GEN IV field has been started –
the NOMAGE 4 network funded by NKS is participated by industrial companies
and other partners and co-ordinated by Studsvik company from Sweden. The
work will be linked with NETNUC.

450
 10. Challenges of New Reactors

Thermal hydraulic transient analysis of the

high performance light water reactor
using APROS and SMABRE (10-3164)

Joona Kurki1, Malla Seppälä2

1
VTT Technical Research Centre of Finland, P.O. Box 1000, FI-02044 VTT, Finland
e-mail: [email protected]
2
VTT Technical Research Centre of Finland, P.O. Box 1000, FI-02044 VTT, Finland
e-mail: [email protected]

Introduction
High Performance Light Water Reactor (HPLWR) is a European supercritical
light water-cooled reactor concept that is currently studied in the “HPLWR2”
project. VTT Technical Research Centre of Finland has participated in performing
safety analyses for the HPLWR using two thermal hydraulics system codes:
APROS [1] and SMABRE [2, 3]. The functionality of these codes was extended
to supercritical pressures by means of introducing a pseudo two-phase region,
which is located along the pseudo-critical line, in the supercritical pressure
region [4, 5]. This approach ensures that second order phase transitions between
liquid and gaseous states, through the supercritical pressure region, are
calculated in a sound manner, effectively treating the phase transition as a first-
order transition. The material properties of water and steam have also been
extended to cover the supercritical region. In addition, various heat transfer and
friction correlations recommended in literature for the supercritical region [6–8]
were implemented in APROS.
Simulation models of the HPLWR were created for APROS and SMABRE.
The aim of these models is to be as mutually similar as possible, in order to
enable comparison between the codes, and to represent the current work-in-
progress design of HPLWR with high accuracy. Both models include a detailed
description of the HPLWR reactor pressure vessel (RPV) with the current three
pass core design [9]. The SMABRE model includes also a rough model of the
steam cycle. Core neutronic solution is not included in either model.

Aim of the work

The capability of APROS and SMABRE for safety analyses of HPLWR is
illustrated by comparison of various transient scenarios including power
decrease and the transition from supercritical to subcritical pressures. These
calculations are a test to the codes' functionality as well as the HPLWR design

451
10. Challenges of New Reactors

and models. The comparisons give promising results about the capability of the
codes for supercritical region calculations.
APROS is also used for comparing the differences induced by the use of
different heat transfer and friction correlations for sub- and supercritical
pressures in supercritical region calculations. The comparisons are performed in
steady-state calculation and in one transient scenario. The comparison results are
presented and their relevance for transient analyses discussed.

Summary
The technical feasibility of the European design for supercritical water-cooled
reactor is studied in the “HPLWR2” project. VTT participates in the project by
performing safety analysis using two system codes, APROS and SMABRE,
which were modified for functioning at supercritical pressures. Thermal
hydraulic simulation models of the HPLWR have been created for the codes
according to the current work-in-progress design.
In this paper, a few transient analyses simulations for the HPLWR are
presented. These results prove that APROS and SMABRE can be applied to
simulations at supercritical pressures as well as to transitions from supercritical
to subcritical pressures,. These results are also used for assessing the
functionality of the current design of the HPLWR. Finally, the effect of the heat
transfer and friction correlations on the simulation results is examined with
APROS, and their relevance for safety analyses calculations discussed.

References
1. Hänninen, M., Ylijoki, J. The one-dimensional separate two-phase flow model of APROS.
VTT Tiedotteita – Research Notes 2443. Espoo, 2008.

2. Miettinen, J. A Thermohydraulic Model SMABRE for Light Water Reactor Simulations.

Licensiate Thesis, Helsinki University of Technology, Espoo, September 1999.

3. Miettinen, J., Hämälaäinen, A. Development and validation of the fast running

thermohydraulic model SMABRE for simulator purposes. Proceedings of
ICONE8, Baltimore, MD USA, April 2–6, 2000. 12 p. CD-ROM Proceedings,
ASME, New York, NY, 2000.

4. Kurki, J. Simulation of thermal hydraulics at supercritical pressures with APROS.

Master’s thesis, Helsinki University of Technology, Espoo, 2008.

5. Seppälä, M. Neutronic and thermal-hydraulic modelling of High Performance Light

Water Reactor. Master’s thesis, Helsinki University of Technology, Espoo, 2008.

6. Pioro, I., Duffey, R. Experimental heat transfer in supercritical water flowing inside
channels (survey). Nuclear engineering and design, 235, 2005, pp. 2407–2430

452
 10. Challenges of New Reactors

7. Pioro, I., Duffey, R. Heat transfer to supercritical fluids flowing in channels – empirical
correlations (survey). Nuclear engineering and design, 230, 2004, pp. 69-91.

8. Pioro, I., Duffey, R. Hydraulic resistance of fluids flowing in channels at supercritical

pressures (survey). Nuclear engineering and design, 231, 2004, pp. 187–197.

9. Schulenberg, T., Starflinger, J. Three Pass Core Design Proposal for a High Perfor-
mance Light Water Reactor. Proceedings of INES-2, Yokohama, Japan,
November 26–30, 2006.

453
 Series title, number and report
code of publication

VTT Symposium 257

VTT-SYMP-257

Author(s)
Seppo Vuori & Rauno Rintamaa (eds.)
Title

20th International Conference on Structural

Mechanics in Reactor Technology
SMiRT 20. Book of abstracts. Vol. 2
Abstract

The international conferences on Structural Mechanics in Reactor Technology

(SMiRT) have traditionally provided innovative and practical mechanics-based
solutions to the planning, design, construction, operation, and regulation of NPPs
and related facilities. SMiRT 20 will continue this tradition, bringing together
experts and practitioners from around the world to share their knowledge of
technology that is most relevant at this time in the nuclear energy industry for
both current operations and future development like Generation IV design.

ISBN
978-951-38-6337-1 (soft back ed.)
978-951-38-6338-8 (URL: http://www.vtt.fi/publications/index.jsp)
Series title and ISSN Project number
VTT Symposium 27180
0357-9387 (soft back ed.)
1455-0873 (URL: http://www.vtt.fi/publications/index.jsp)
Date Language Pages
July 2009 English 453 p.
Name of project Commissioned by

Keywords Publisher
Nuclear power plants, Nuclear facilities, Nuclear
VTT Technical Research Centre of Finland
safety, Structural safety, Advanced reactors,
Mechanics of materials, Aging, Plant life management, P.O. Box 1000, FI-02044 VTT, Finland
Inspection, Maintenance, Design and qualification, Phone internat. +358 20 722 4520
Fracture Mechanics, Structural evaluation, Structural Fax +358 20 722 4374
reliability, Probabilistic safety assessment, Extreme
loads, Earthquakes, Fuel and core structures, Severe
accident management, Computational mechanics,
Metal materials, Concrete materials, Containment
structures, Seismic loads, Seismic analysis, Design
methods
VTT CREATES BUSINESS FROM TECHNOLOGY
Technology and market foresight • Strategic research • Product and service development • IPR and licensing
• Assessments, testing, inspection, certification • Technology and innovation management • Technology partnership
VTT SYMPOSIUM 257

• • • VTT SYMPOSIUM 257

IASMiRT

20th International Conference on

SMIRT 20 – Book of abstracts. Vol. 2

STRUCTURAL MECHANICS IN REACTOR TECHNOLOGY
Dipoli Congress Centre, Espoo (Helsinki), Finland
August 9–14, 2009

The international conferences on Structural Mechanics in Reactor Technology (SMiRT)

have traditionally provided innovative and practical mechanics-based solutions to
the planning, design, construction, operation, and regulation of NPPs and related
facilities. SMiRT 20 will continue this tradition, bringing together experts and prac-
titioners from around the world to share their knowledge of technology that is most
relevant at this time in the nuclear energy industry for both current operations and
future development like Generation IV design.

SMiRT 20 – Book of abstracts

Vol. 2
20th International Conference on
ISBN 978-951-38-6337-1 (soft back ed.) ISBN 978-951-38-6338-80 (URL: http://www.vtt.fi/publications/index.jsp)
STRUCTURAL MECHANICS
ISSN 0357–9387 (soft back ed.) ISSN 1455-0873 (URL: http://www.vtt.fi/publications/index.jsp)
IN REACTOR TECHNOLOGY
TIETORATKAISUT
JULKAISUJEN TOIMITUS