ESTIMATION OF THE SEISMIC RESPONSE OF

BUILDINGS AND THE EFFECT OF DIFFERENT SCALING

METHODS FOR GROUND MOTION

Sundar Ram Krishna Murthy Mohan Ram

A Thesis
In
The Department
Of
Building, Civil and Environmental Engineering

Presented in Partial Fulfillment of the Requirements

For the degree of Master of Applied Science (Civil Engineering) at
Concordia University
Montreal, Quebec, Canada

December 2012

© Sundar Ram Krishna Murthy Mohan Ram 2012

 CONCORDIA UNIVERSITY
School of Graduate Studies
This is to certify that the thesis prepared
By: Sundar Ram Krishna Murthy Mohan Ram
Entitled: ESTIMATION OF THE SEISMIC RESPONSE OF
BUILDINGS AND THE EFFECT OF DIFFERENT
SCALING METHODS FOR GROUND MOTION

And submitted in partial fulfillment of the requirements for the degree of

Master of Applied Science (Civil Engineering)
Complies with the regulations of the University and meets the accepted standards with
respect to originality and quality.

Signed by the final examining committee:

________________________ Chair (Dr. L. Tirca, BCEE)
________________________Examiner External to the Program (Dr. Al Dlatabadi, MIE)
________________________Examiner (Dr. L. Tirca, BCEE)
________________________Examiner (Dr. A. Bhowmick, BCEE)
________________________Supervisor (Dr. A. Bagchi, BCEE)

Approved by _________________________________________
Chair of Department or Graduate Program Director

________20______ ____________________
Dean of Faculty

ii
 ESTIMATION OF THE SEISMIC RESPONSE OF BUILDINGS AND THE

EFFECT OF DIFFERENT SCALING METHODS FOR GROUND MOTION

Sundar Ram Krishna Murthy Mohan

ABSTRACT

The Seismic design code of Canada is changing rapidly to accommodate the needs of the

future generation of buildings for management of earthquake hazard mitigation. In this

context the recent advancement in Earthquake Technology and Structural Engineering

has emphasized on the need for a better methodology and in-depth investigation into the

area of structural performance evaluation in order to ensure that structures designed for

the areas of high and moderate seismic hazard to the expected standards and meet the

objective of life safety and collapse prevention in a real life scenario. In order to ensure

the above performance objectives for a building structure, it is necessary to estimate its

capacity with respect to the demand, and the dynamic response corresponding to the

design levels of earthquakes. The research carried out here aims to investigate: (i) the

earthquake demand and capacity profiles of a set of set of moment frame buildings

designed according to the latest version of the National Building Code of Canada, and (ii)

the effect of scaling and spectral matching techniques commonly applied to ground

motions on the seismic demand parameters determined using the dynamic time history

analysis. A set of buildings with steel moment resisting frames of 5, 10, 15, 20 stories in

height and located in Vancouver area of Canada have been considered in this study. An

extensive review has been conducted to determine the existing methods for performance-

based design and the techniques available to selecting and scaling suite of earthquake

iii
records to perform a fully non-linear dynamic analysis in time domain. Based on that, a

range of scaling techniques including linear scaling techniques, and spectral matching

technique have been considered for an ensemble of recorded ground motion time

histories. In addition a set of artificially generated spectrum-compatible earthquake

records are also considered. The static pushover analysis has been carried out and the

corresponding capacity curves have been obtained and interpreted with commonly used

performance-based design methods. It is observed that all the methods considered here

confirm that the existing design based on the code procedure is adequate and

conservative. The pushover curves are also compared to the results obtained from the

Time history analysis to determine the performance achievements of the buildings. The

interstory drift obtained from the time history analysis using different scaling methods

show a uniform and consistent pattern of deformation in low rise to medium rise frames

whereas dispersion greater dispersion of the results has been observed in tall buildings.

Other response quantities such as the lateral drift, base shear and bending moment show

similar patterns. Based on the results from the research it is suggested to use the artificial

records if site specific real ground motion records are unavailable. The scope for further

research lies in exploring ways to the possibility of new scaling techniques that can

control the dispersion in the response more effectively.

iv
 Acknowledgements
It’s my deepest pleasure to express my sincere gratitude and heartfelt thanks to

my program supervisor Dr.Ashutosh Bagchi, Associate Professor at Concordia University

for having been the foremost source of inspiration, guidance and support through my

M.A.Sc. program. “Sir, please accept my sincere regards and heartfelt thanks for all that

you have done for me through the course of my M.A.Sc., program”

I fully acknowledge and convey my sincere gratitude and heartfelt appreciation to my

mother Smt.Sugandha for her timeless encouragement and support in my student life.

Acknowledgement of thanks goes to my course work supervisors Dr. Oscar Pekau, Dr.

Lucia Tirca, Dr. Kinh H. Ha, Dr. Mohammed Ali and Mr.Adel Zaki. I also thank Jenny,

Olga and the staff at BCEE department, Concordia University for their support during my

graduate study.

The research was partially funded by Dr. Bagchi’s research grant from the Natural

Sciences and Engineering Research Council of Canada (NSERC) and Concordia

University which is highly appreciated and acknowledged with thanks. Support from

Canara Bank, India is also acknowledged. My sincere thanks and appreciation goes to my

family and my colleagues in the Building, Civil and Environmental engineering

department and to for their encouragement through my M.A.Sc. Program. Last but not

the least, I whole heartedly pray and thank to pay my due respects to the Supreme God in

guiding me to complete my graduate program with success.

v
 TABLE OF CONTENTS

List of figures................................................................................................................... vi
List of tables ...................................................................................................................... vi

Chapter 1 ............................................................................................................................. 1
Introduction ......................................................................................................................... 1
1.1 Preface....................................................................................................................... 1
1.2 NBCC 2010 - Seismic Design Provisions. ............................................................... 4
1.3 Performance–based Seismic Design. ........................................................................ 5
1.4 Ground Motion Scaling Techniques. ........................................................................ 8
1.5 Thesis Objectives and Scope. .......................................................................... …….9
1.6 Thesis Outline and Structure. ......................................................................... …….10
Chapter 2 ........................................................................................................................... 11
2.1 Introduction ............................................................................................................. 11
2.2 Seismic Performance Evaluation ............................................................................ 13
2.3 Selection of Ground Motion Records (GMR) ........................................................ 15
2.3.1 Real Accelerograms .......................................................................................... 15
2.3.2 Artificial Records .............................................................................................. 16
2.3.3 Synthetic Accelerograms .................................................................................. 17
2.4 Spectral Matching and GMR Scaling .................................................................... 17
2.4.1 Target Spectral Matching ................................................................................. 18
2.4.2 Ground Motion Scaling in Time Domain ........................................................ 18
2.4.3 Spectral Matching in Frequency Domain ........................................................ 19
2.4.4 Spectral Matching in Time Domain .................................................................. 19
2.5 Modal Pushover Scaling(MPS) .............................................................................. 22
2.6 Steel Moment Resisting Frame .............................................................................. 23
2.7 Review of NBCC 2010 Code Provisions ............................................................... 24
2.8 Summary ............................................................................................................... 27
Chapter 3 ........................................................................................................................... 27
Design of Steel Moment Resisting Frames ....................................................................... 28
3.1 Introduction ............................................................................................................. 28
3.2 Automation of Structural Analysis and Design ..................................................... 31
3.3 Design of Steel Moment Resisting Frames ............................................................ 32
3.4 Modal Analysis ...................................................................................................... 35
3.5 Modal Analysis using ETABS Software …………...…………………………….43
3.5.1 Modeling …………………………………………………………………….44
3.5.2 Loading …………………………………………………………………...…45
3.5.3 ETABS Analysis …………………………………………………………….45
3.6 Summary ………………………………………………………………………….46
Chapter 4 ........................................................................................................................... 48
Selection and Scaling of Ground Motions ....................................................................... 48
4.1 Introduction ............................................................................................................. 48
4.2 Selection of Ground Motion Records( GMR) ........................................................ 49
4.2.1 Selection based on Magnitude (M) & Distance (R)......................................... 52
4.2.2 Selection based on Site Soil Conditions .......................................................... 53

vi
 4.2.3 Selection based on Spectral Matching of strong motion parameters……...…...54
4.2.3.1 Evaluation of a/v ratio ................................................................................... 56
4.2.3.2 Effect of duration of ground motion shaking ................................................ 56
4.3 Scaling of the selected GMRs ................................................................................ 57
4.3.1 PGA Scaling..................................................................................................... 57
4.3.2 Ordinate Scaling Method ................................................................................ 59
4.3.3 Least Square Method ...................................................................................... 60
4.3.4 Partial Area Method ........................................................................................ 62
4.3.5 PSa Scaling Method ........................................................................................ 63
4.3.6 ASCE - 2007 Scaling Method.......................................................................... 65
4.3.7 Spectrum matching Technique ........................................................................ 66
4.3.8 Spectrum- Compatible artificial earthquake record ......................................... 67
4.4 GMRs used in the present study ............................................................................. 67
4.5 Summary. ............................................................................................................ 82
Chapter 5 ........................................................................................................................... 83
Static Pushover Analysis................................................................................................... 83
5.1 Introduction ............................................................................................................. 83
5.2 Performance estimation based on Pushover analysis.............................................. 88
5.2.1 Capacity demand diagram method................................................................... 88
5.2.2 N2 Method ...................................................................................................... 98
5.2.3 DBSD Method .............................................................................................. 100
5.2.4 Yield Point Spectra Method ( Aschheim M., 2004): ..................................... 103
5.3 Summary .............................................................................................................. 108
Chapter 6 ......................................................................................................................... 110
Non-linear Time-history Analysis .................................................................................. 110
6.1 History................................................................................................................... 110
6.2 Non-linear Time-history Analysis ........................................................................ 111
6.3 Discussion of Interstorey drift results……..………………………………….…112
6.3.1 PGA Scaling................................................................................................... 112
6.3.2 PSa Scaling method ........................................................................................ 117
6.3.3 Ordinate Scaling Method .............................................................................. 121
6.3.4 Partial Area Scaling Method ......................................................................... 125
6.3.5 ASCE 2007 Scaling Method ......................................................................... 129
6.3.6 Least Square Scaling Method ....................................................................... 133
6.4 Discussion of Interstory drift results from spectrum compatible records ........... 137
6.4.1 Spectral matching using Seismo Match software ........................................... 137
6.5 Discussion of Interstory drift results from Atkinson’s artificial records ............ 141
6.5.1 Spectral matching using Atkinson’s artificial records .............................. 141
6.6 Summary ............................................................................................................... 144
Chapter 7 ......................................................................................................................... 148
Observations and Conclusion ........................................................................................ 148
7.1 Observation ........................................................................................................... 148
7.2 Conclusion ........................................................................................................... 150
7.3 Scope for future work .......................................................................................... 152
Reference: ................................................................................................................... 153

vii
 LIST OF FIGURES

Fig.1.1 Growth in Worldwide Number of Seismic Codes (International Atomic

Energy Agency , 2004) ………. ………………………………….……....3
Fig.2.1 The use of real earthquake accelerograms (Bommer and Acevedo,
2004)...…...................................................................................................21
Fig.2.2 Design Spectra for Vancouver (NBCC 2010)…………………...………27
Fig.3.1 Layout Plan of the Building ……………………………………………..29
Fig.3.2 Elevation of 5 & 10 Story frames………………………………………..29
Fig.3.3 Elevation of 15 & 20 Story Frames ……………………………………..30
Fig.3.4 Model of DRAIN-2DX Element Type-2 (Prakash et al., 1995)…….…...32
Fig.3.5 Mode Shapes of 10 and 5 Story Building Frames; (a) Mode Shapes of 10
Story Building Frame, (b) Mode Shapes of 5 Story Building Frame……36
Fig.3.6 Mode Shapes of 20 and 15 Story Building Frames; (a) Mode Shapes of 20
Story Building Frame, (b) Mode Shapes of 15 Story Building Frame ….37
Fig.3.7 Flow-chart for the design and evaluation of Steel Moment Resisting
Frames ………………………………………………………………...…42
Fig.3.8 Plan view of the building in ETABS.……………….……………...…....44
Fig.3.9 3-Dimensional view of the structure in ETABS ………..……...…...…...45
Fig.4.1 Available Selection procedures for strong motion records Bommer and
Acevedo (2004) ………………………………………………………….50
Fig.4.2 Peak Ground Acceleration in the selected GMR ……………...…....……58
Fig.4.3 Peak Ground Acceleration from the NBCC Code Spectrum ……...….…58
Fig.4.4 Ordinate at T1 on the input GMR …………...……………….……...…...59
Fig.4.5 Ordinate at T1 on the NBCC Code Spectrum …………………………...60
Fig.4.6 Least Square Scaling Ordinates on the input Ground Motion Record…...61
Fig.4.7 Least Square Scaling Ordinates on the NBCC Spectrum ……………….61
Fig.4.8 Partial Area Scaling Ordinates on the input Ground Motion Record ...…62
Fig.4.9 Partial Area Scaling Ordinates on the NBCC Code Spectrum ………….63
Fig.4.10 PSa Scaling Ordinates on the input Ground Motion Record ………....….64
Fig.4.11 PSa Scaling Ordinates on the NBCC Spectrum …………………………64
Fig.4.12 ASCE-7 Scaling Ordinates on the input Ground Motion Record …….…65

viii
Fig.4.13 ASCE Scaling Ordinates on the NBCC Spectrum …………………..…..66
Fig.4.14 Graph of records matched with code spectrum using Seismomatch
software……………………………………………………………....…..69
Fig.4.15 Time History of Atkinson’s synthesized ground motions.
(a) Short period ground motions (b) Long period ground
motions.................................................................................................….69
Fig.4.16 Spectra of selected Synthesized Ground Motion Records along with the
NBCC 2005/2010 design spectrum……………………………...………69
Fig.4.17 Spectra of selected Ground Motion Records along with the NBCC
2005/2010 design spectrum………………………………………….…..82
Fig.5.1 Pushover Curves for (a) 5 Story SMRF (b) 10 Story SMRF (c) 15
Story SMRF (d) 20 Story SMRF ……...………………………..……..84
Fig.5.2 Seismic demand curves using different R-µ-T relations (KN Krawinkler
and Naser (1992);VFF Vidic et al. (1994); and NH Newmark and Hall
(1982)) ………………………………………………………………...…94
Fig.5.3 Application of the CDD, N2 and DBSD methods (20 story
building)………………………………………………………………….95
Fig.5.4 Application of the CDD, N2 and DBSD methods (15 story
building)……………………………………………………………….....96
Fig.5.5 Application of the CDD, N2 and DBSD methods (10 story
building)…………………………………………………………………..97
Fig.5.6 Application of the CDD, N2 and DBSD methods 5 story building ….….97
Fig.5.7 Yield point spectrum of NBCC 2010 response spectrum for Vancouver
(20 story building) ……………………………………………………...104
Fig.5.8 Yield point spectrum of NBCC 2010 response spectrum for Vancouver
(15 story building) …………………………………………..……….…106
Fig.5.9 Yield point spectrum of NBCC 2010 response spectrum for Vancouver
(10 story building) …………………………………………..………….107
Fig.5.10 Yield point spectrum of NBCC 2005 response spectrum for Vancouver
(5 story building)…………………………………………..………….. .108
Fig.6.1 Interstory drift graphs from RHA of GMRs scaled using PGA method

ix
 (a) 5 story SMRF (b) 10 story SMRF………………………………….114
Fig.6.2 Interstory drift graphs from RHA of GMRs scaled using PGA method
(a) 15 story SMRF (b) 20 storySMRF………………………..……….115

Fig.6.3 Interstory drift graphs from RHA of GMRs scaled using PSa method
(a) 5 story SMRF (b) 10 story SMRF……………………………….….118
Fig.6.4 Interstory drift graphs from RHA of GMRs scaled using PSa method
(a) 15 story SMRF (b) 20 story SMRF……………………..…………119
Fig.6.5 Interstory drift graphs from RHA of GMRs scaled using Ordinate method
(a) 5 story SMRF (b) 10 story SMRF……………………….…………122
Fig.6.6 Interstory drift graphs from RHA of GMRs scaled using Ordinate method
(a) 15 story SMRF (b) 20 story SMRF…………..……………………123
Fig.6.7 Interstory drift graphs from RHA of GMRs scaled using Partial area
scaling (a) 5 story SMRF (b) 10 story SMRF…………….………..….126
Fig.6.8 Interstory drift graphs from RHA of GMRs scaled using Partial area
scaling (a) 15 story SMRF (b) 20 story SMRF ……………………...127
Fig.6.9 Interstory drift graphs from RHA of GMRs scaled using ASCE 2007
scaling (a) 5 story SMRF (b) 10 story SMRF…………………...…130
Fig.6.10 Interstory drift graphs from RHA of GMRs scaled using ASCE 2007
scaling (a) 15 story SMRF (b) 20 story SMRF………………..…..131
Fig.6.11 Interstory drift graphs from RHA of GMRs scaled using Least square
scaling (a) 5 story SMRF (b) 10 story SMRF .…………………….134
Fig.6.12 Interstory drift graphs from RHA of GMRs scaled using Least sqaure
scaling (a) 15 story SMRF (b) 20 story SMRF ……………………….135
Fig.6.13 Interstory drift graphs from RHA of Spectral matching using Seismo
match (a) 5 story SMRF (b) 10 story SMRF…………….…………….138
Fig.6.14 Interstory drift graphs from RHA Spectral matching using Seismo match
(a) 15 story SMRF (b) 20 story SMRF ……………………….………139
Fig.6.15 Interstory drift graphs from RHA of Spectral matching using Atkinson’s
artificial earthquake records (a) 5 story SMRF (b) 10 story SMRF…...142
Fig.6.16 Interstory drift graphs from RHA Spectral matching using Atkinson’s
artificial earthquake records (a) 15 story SMRF (b) 20 story SMRF..…143

x
 LIST OF TABLES

Table 1.1 Design Earthquakes (SEAOC Vision 2000, 1995)…………….………….7

Table 1.2 Structural Performance Level (Vision 2000)…………………….….…….7
Table 1.3 Performance level (FEMA-273) ...............................................................13
Table 1.4 Values of Mv for various structural systems (NBCC 2010)......................25
Table 1.5 Design Spectra of NBCC 2005 (Adams and Atkinson, 2003)...................26
Table 3.1 Design loads…………………...................................................................33
Table 3.2 Fundamental periods of buildings .............................................................34
Table 3.3 Base shear of SMRF……….……………….............................................35
Table 3.4 Sample calculation of base shear after modal analysis ………...………..38
Table 3.5 Sections of Columns ………………………………….. ...….....………..40
Table 3.6 Sections of beams......................................................................................41
Table 3.7 Modal periods from DRAIN 2D and ETABS............................................46
Table 4.1 Characteristics of Atkinson’s Synthesized Ground motion ......................69
Table 4.2 Summary of Real Ground Motion ………………………………………70
Table 4.3 Modal periods of the building frames........................................................71
Table 4.4 Scale Factors using PGA Method ………………………………………72
Table 4.5 Scale Factors using Ordinate Method (5 and 10 storey)…………...……73
Table 4.6 Scale Factors using Ordinate Method (15 and 20 storey) ………...…….74
Table 4.7 Scale Factors using Least Square Method (5 and 10 storey)………….…75
Table 4.8 Scale Factors using Least Square Method (15 and 20 storey)………..….76
Table 4.9 Scale Factor using Partial Area Method (5 and 10 storey)…….….……..77
Table 4.10 Scale Factor using Partial Area Method (15 and 20 storey) …………….78
Table 4.11 Scale Factors using PSa Scaling …………………………………………79
Table 4.12 Scale Factors using ASCE-7 method (5 and 10 storey) …………………80
Table 4.13 Scale Factors using ASCE-7 method (15 and 20 storey) ………………..81
Table 5.1 Base Shear Coefficient at the first beam and first column yielding ….…86
Table 5.2 Design base shear and yield points of the SMRFs.……………………....87
Table 5.3 Displacement at failure (i.e., point of instability or 2.5% drift) of the
frames ………………………………………………..…………………..87
Table 5.4 Roof displacement (% H) at Maximum M+SD of interstorey drift ……..88

xi
Table 5.5 Base shear distribution in 20 story SMRF …………………….….……..89
Table 5.6 Base shear distribution in 15 story SMRF ……………………………....90
Table 5.7 Base shear distribution in 10 story SMRF …………………………...….91
Table 5.8 Base shear distribution in 5 story SMRF…………………………….…..91
Table 5.9 Values of Γ and m* of SMRF………………………………….………..92
Table 5.10 Yield and ultimate base shear and displacement values of the equivalent
SDOF systems in A-D format using the CDD method……….……....… 93

Table 5.11 The SDOF and MDOF demand using the CDD method ……………..…97

Table 5.12 Yield and ultimate base shear in acceleration format and the elastic period
of the equivalent SDOF using the N2 method……………………...……98

Table 5.13 The SDOF and MDOF demand using the N2 method …………….....…100

Table 5.14 Estimation of the design base shear using the DBSD method………..…102

Table 6.1 PGA Scaling method - Summary of Interstory drift for Real ground
motion ……………………………………………………………….....113
Table 6.2 Base Shear (KN) from PGA scaling method …………………...……..116
Table 6.3 PSa Scaling Method - Summary of Interstory Drift for Real Ground
Motion ………………………………………………………………….117
Table 6.4 Base Shear (KN) from PSa scaling method …………………...…...….120
Table 6.5 Ordinate Scaling Method - Summary of Interstorey Drift for Real Ground
Motion………………………………………………………………….121

Table 6.6 Base Shear (KN) from Ordinate scaling method ……………...……....124
Table 6.7 Partial Area Scaling Method -Summary of Interstory Drift for Real
Ground Motion……………………………………………….…………125
Table 6.8 Base Shear (KN) from Partial area scaling method ……………….......128
Table 6.9 ASCE 2007 Scaling Method- Summary of Interstorey Drift for Real
Ground Motion………………………………………………………….129
Table 6.10 Base Shear (KN) from ASCE 2007 Scaling method …………...……..132
Table 6.11 Summary of Interstory Drift for Real Ground Motion from Least Square
Scaling Method ………………………………………………….......…133
Table 6.12 Base shear from Least square scaling method ……………………...….136

xii
Table 6.13 Summary of Interstory Drift for Real Ground Motion from Seismo Match
Scaling method ………………………………………………..………..137
Table 6.14 Base Shear (KN) from Seismo Match Spectral matching …...….…….140
Table 6.15 Summary of Interstory Drift for Real Ground Motion from Atkinson’s
Artificial earthquake records ……………………………….......………141
Table 6.16 Base Shear (KN) from Atkinson’s artificial records …............….…….144
Table 6.17 Summary of interstorey drift and base shear for the 5-storey frame ..…145
Table 6.18 Summary of interstorey drift and base shear for the 10-storey frame….145
Table 6.19 Summary of interstorey drift and base shear for the 15-storey frame.…145
Table 6.20 Summary of interstorey drift and base shear for the 20-storey frame …146

xiii
My Loving Mother

xiv
 Chapter 1

Introduction
1.1 Preface

Earthquake is described a seismic event involving sudden release of energy in the earth

crust giving rise to seismic waves which causes ground motion or shaking of the ground,

In the history of the mankind several strong earthquake incidents and their aftermaths

have led to massive damage of property, destructive fires, tsunami and huge loss of

human life. The recent earthquake in Japan (2011) of magnitude 9.0 is one such

catastrophic scenario where in the aftermath effects provoked nuclear hazard in the

region. Although the occurrence of an earthquake cannot be predicted to take

precautionary measures and avoid loss of life and damage to property, the need of the

hour is to aim for a possible response preparedness to deal with such a scenario in future

in case of occurrence.

Earthquake engineering is an inter-disciplinary area which developed in the late 20th

century; it is a branch of civil engineering dealing in mitigating earthquake hazards by

applying mainly the principles and knowledge of engineering sciences and seismology.

The scope of the earthquake engineering include (a) Investigation of regional earthquake

hazard to select a suitable location for the proposed structure (b) Estimation of the hazard

at the selected location considering an adequate time interval and the local site conditions

(c) Estimation of the structural response under the imposed hazard forces beforehand to

facilitate adequate design of structure in whole and its members to resist such forces in

order to prevent failure or collapse.

1
Looking back into the history of the built environment, many buildings, bridges and other

facilities were designed by engineers to improve the performance during earthquakes

However, the technology and the construction methods for earthquake resistant design of

structures are still evolving. There is still a need for the development of effective tools for

engineering analysis to compute the design seismic demands of the structural components

and carry out performance-based seismic design.

It is important to note that earthquake engineering has been widely acknowledged and

received well by researchers and engineers all over the world in the late 20th century, and

the knowledge database has been growing in a significant rate supported by advancing

technologies, building code regulations, education and training professionals, and by

public support. Fig.1 shows the worldwide growth trend of seismic design codes.

2
 60

54
50
46

40

35

30
28 27

20

13
10

0
1960 1970 1980 1990 2000 2010
YEAR

Figure 1.1: Growth in Worldwide Number of Seismic Codes ( Source: International

Association for Earthquake Engineering (2004, 2000, 1996, 1992, 1980a, 1980b,
1976, 1973, 1966, 1963, 1960))

Fig. 1.1 shows the development of seismic codes worldwide this led to subsequent

growth in knowledge and research for technology to make buildings and structures more

earthquake resistant, which were later investigated and adopted quickly by many

developed and developing countries. Under the Canadian scenario the NBCC 2010 being

the latest version was revised to a large extent in its previous edition NBCC 2005 to

streamline seismic deign provisions for practicing engineers., The latest Canadian code

recommends the use of dynamic analysis for seismic performance evaluation of existent

3
and new structures. While NBCC 2010 is not a performance-based design code, it is said

to be an objective-based code that allows the use of new materials or design processes

based on acceptable solutions to achieve the stated objectives in the design. In the context

of performance-based design procedure, it is necessary to determine if a building

designed according to the current provisions of the code actually is capable of achieving

the given performance objectives assumed in the design and determine possible

modifications to incorporate multiple levels of performance corresponding to various

levels of seismic hazard. The research presented here looks at a number of buildings

designed according to the current seismic provisions in Canada in the context of their

performance achievements under the design level of seismic hazard utilizing various

methods of response prediction and performance evaluation.

1.2 NBCC 2010 - Seismic Design Provisions

The seismic design of buildings in Canada is required to be performed according to the

provisions of the National Building Code of Canada (NBCC). NBCC 2010 is the latest

version of the building code which is based on the revision of NBCC 2005. It allows for

the use of Equivalent Static Load Method (ESLM) for estimating the lateral forces due to

seismic hazard for buildings with simple and regular shape and geometric configurations,

and of a limited height. While dynamic analysis is recommended for all buildings, it is

mandatory for structures of irregular, complex geometry and buildings of height above 60

m.

NBCC 2010 addresses the overall building performance in a broader perspective, by

considering the parameters of ground motions, site soil effects, analysis and design

4
methodologies (De Vall, 2003). The important features and noteworthy points for seismic

design under NBCC 2010 are as follows

 It provides the Uniform hazard spectra for the specific site to be used for Seismic

design purpose, the hazard spectrum has 2% probability of exceedance in 50 years

with a recurrence interval of 2500 years (Humar and Mahgoub, 2003), further the

probability of exceedance of the Uniform hazard spectra is said to be a function of

time, with similar characteristics to that of the hazard spectra.

 NBCC 2010 has a broader objective to achieve the required performance and

safety of the structure, hence it allows for use for alternate methods of analysis

and design to meet the acceptable levels of performance, which may not be

specified in the code.

 NBCC 2010 also provides a description and guidelines for structural irregularity

1.3 Performance–based Seismic Design

Recent innovation and advancement in the area of earthquake engineering have led to

development of new state-of–art approaches towards performance evaluation and design

of structures called Performance-Based Seismic Design. Here, the emphasis is given to a

priori evaluation of performance of a structure related to the site specific seismic hazard.

Seismic hazard includes ground fault, rupture, ground-shaking liquefaction, lateral

spreading and land sliding. The new approach helps in pre-emption of the structural

performance through qualitative and quantitative means based on controlling the response

and damage parameters at the time of the design.

5
Performance-based seismic design is a two-step process which involves performance

evaluation and structural design. The main purpose of performance evaluation is to check

the performance of structure up to a desired level under dynamic forces induced by

ground motion. The aim of performance-based design is to design the structure based on

the desired or assumed performance level to be achieved under seismic excitation. The

capacity and the seismic response need to be determined accurately to estimate the level

of damage and corresponding performance of a structure. Damage parameters such as the

interstorey drift, roof-drift, joint rotation etc. which are displacement based quantities are

among the most widely used parameters (Bagchi, 2001) to determine the level of seismic

performance. These damage parameters can be determined using static and dynamic

analyses of a structure. Usually a nonlinear time history analysis of a structure subjected

to seismic ground acceleration is considered a more appropriate method to determine the

response parameters accurately. However, the selection of seismic ground motion and

scaling them appropriately for the use in the nonlinear time history analysis are important

issues which still require further research.

In order to achieve the required performance level by design, a performance objective is

predefined and consists of specification of performance level of the structure and a

corresponding probability that this performance level may exceed (Yun et al 2002).

The Structural Engineers Association of California have laid down guidelines for
Performance objectives under different seismic hazard levels Table (1.1) and a
description of different types of structural performances in Table (1.2).

6
 Table 1.1: Design Earthquakes (SEAOC Vision 2000, 1995)

Earthquake Design Level Recurrence Interval Probability of Exceedance

Frequent 43 years 50% in 30 years
Occasional 72 years 50% in 50 years
Rare 475 years 10% in 50 years
Very Rare 970 years 10% in 100 years
Extremely Rare 2500 years 2% in 50 years

Table 1.2: Structural Performance Level (Vision 2000)

Performance Description Transient Permanent

Level drift
_
drift
Fully No significant damage has occurred to <0.2% Negligible
Functional structural and non-structural components.
Building is suitable for normal intended
occupancy and use.

Operational No significant damage has occurred to <0.5% Negligible

structure, which retains nearly all of its pre-
earthquake strength and stiffness. Non-
structural components are secure and most
would function, if utilities available. Building
may be used for intended purpose, albeit in an
impaired mode.
Life Safe Significant damage to structural elements, with < 1.5% <0.5%
substantial reduction in stiffness, however,
margin remains against collapse. Non-
structural elements are secured but may not
function. Occupancy may be prevented until
repairs can be instituted.
Near Substantial structural and non-structural <2.5% < 2.5
Collapse damage. Structural strength and stiffness
substantially degraded. Little margin against
collapse. Some falling debris hazards may
occur.

It is noteworthy to recognize that NBCC 2010 does not provide guidelines for seismic

performance evaluation but specifies the maximum allowable interstorey drift as 2.5%

beyond which the structure is assumed to have failed.

7
1.4 Ground Motion Scaling Techniques

The next generation of design codes, especially those adopting the framework of

performance based seismic design, shall include the option of design based on

displacement parameters rather than forces. Non-linear dynamic time history analysis

conducted as part of a performance-based seismic design approach typically involves the

following steps

 Obtain site specific input accelerograms for dynamic analysis

 Perform nonlinear dynamic analysis to compute internal forces and displacements

 Check for the strength of structural elements by computing capacity ratios

 Take adequate steps to ensure structural integrity, safety and performance

It is however noted that suitable Ground Motion Records (GMR) which are site specific

are usually unavailable and uncertain. Nonlinear dynamic analysis require the ground

motion acceleration time histories which cover the spectral ordinates of the site specific

target spectrum prescribed in the codes. In this scenario the ground motion records are

obtained using one of the three alternative procedures.

 Selection of a real accelerogram from a GMR database with site specific

conditions and characteristics (e.g., magnitude [M], distance [R], duration [D],

soil condition [SSI]);

 Simulate GMR from seismological model of fault rupture mechanisms; or

 Artificial or synthetic ground motions generated from filtered noise.

8
Nonlinear dynamic analysis requires scaling of the real accelerograms for a GMR to that

of the target spectrum, which can be done by scaling spectral ordinates without altering

the spectral shape or scaling the spectral ordinates and modifying the spectral shape to

match the target spectrum. Ideally, the analysis requires scaled real accelerograms

without altering the spectral shape. This is because nonlinear displacement and ductility

demands are sensitive to the details of the ground motions containing sequences of peaks

and valleys as well as long duration pulses. The scaling of spectral ordinates and

modification of the spectral shape could however be done in frequency domain or in time

domain. From the structural damage assessment point of view, the effect of spectral

matching and scaling techniques used to obtain the site specific ground motion

characteristics and the related damage potential needs to be studied as there is lack of

knowledge in this area. The present research attempts to address the above need.

1.5 Thesis Objectives and Scope

The objectives and scope of the research carried out are outlined below

 To determine the capacity and the seismic demand characteristics of steel

moment-frame buildings designed according to the Canadian code provisions

 To determine the effect of ground motion scaling techniques on the seismic

performance parameters of the above steel moment resisting frames

 To develop a methodology or guideline to reduce the effect of scaling techniques

on the seismic demand of a structure.

In order to achieve the above goals, the following tasks have been undertaken:

9
  Design a set of building with steel moment resisting frames according NBCC

2010, where the buildings are assumed to be located in Vancouver;

 Select a set of GMRs to represent the seismicity of Vancouver;

 Implement a number of established GMR Scaling Techniques to scale the selected

GMRs which are used for carrying out Nonlinear Time History analysis of the

Steel moment resisting frame buildings mentioned above;

 Perform a statistical analysis of the seismic demand parameters corresponding to

all the GMRs and compare them for different scaling methods; and

 Based on the results, provide a guideline for the selection and scaling of GMRs

for time history analysis and evaluation of the seismic response of buildings.

1.6 Thesis Outline and Structure

The thesis has been organized into seven chapters. Objective of the thesis with some

introductory materials are presented in the current chapter i.e Chapter 1. A review of

literature on this topic is provided in the Chapter 2. Design of the Steel moment resisting

frame buildings considered in the research has been presented in Chapter 3. Chapter 4

details the Selection and Scaling of Seismic Ground Motion Records. Chapter 5

discusses the response of the building frames to Static Pushover Analysis under which, it

is interpreted using various Performance-based Seismic Design (PBSD) procedures.

Chapter 6 presents and discusses the seismic response of the building frames obtained

using Non-linear dynamic analysis and the summary of the present thesis and conclusions

are presented in the Chapter 7. A list of references used in the thesis has been provided at

the end.

10
 Chapter 2

Literature Review

2.1 Introduction
Earthquake engineering in the context of civil engineering deals with seismic hazards

assessment and design of structures to cope with the expected levels of hazard. It lays

down guidelines for planning, analysis and design of structures in such a way that are

capable of achieving an expected level of performance to a given level of seismic hazard.

These principles have been the basis of most seismic codes over the decades which are

broadly classified towards the following three goals according to the Structural

Engineering Association of California (SEAOC 1959-1999):

a.) A minor level of earthquake ground motion without damage;

b.) A moderate level of ground motion without structural damage but possibly

experience some non-structural damage;

c.) A major level of ground motion having an intensity equal to the strongest , either

experienced or forecast for the building site without collapse , but possibly with

some structural as well as non-structural damage.

The recent research advancement in earthquake engineering has often advocated for an

innovative performance-based design code instead of the current code which prescribes

design guidelines in a simplistic methodology. The aim of performance-based design is to

design the structure for a no-collapse condition after evaluating the seismic performance

of the structure under a suitable ground motion record which is site specific, and also

11
ensure different levels of performance corresponding to different levels of seismic

hazard. The seismic performance of a structure is assessed and evaluated using inter-

storey drifts, inelastic deformations, strains and many other damage indices. Several

simplified methods are developed for performance-based design such as displacement

based design method, damage spectrum ( Bozorgnia and Bertero, 2004) and yield point

spectra, to name a few. Although several simplified methods are developed for such

design, these methods are still very approximate and differ significantly from each other,

since no reliable and robust performance-based design method is available, NBCC 2010

still allows for the traditional force-based design for simple and regular buildings

However, NBCC 2010 requires the use of dynamic analysis for complex structures or

structures exceeding 60 m in height. While dynamic analysis is required to deal with

seismic design, it is still not very practical for everyday office use as it requires

significant time in solving for representative ground motion records and in data

processing. Given that there is a wide range of uncertainties related to selection of

appropriate ground motion records, material modeling and analysis algorithms,

significant training and experience is required for a design engineer to carry out such

assessment and interpret the results carefully. On the other hand, design codes are

required to be simple and robust, and their procedures are expected to be directly based

on sound understanding of the physical nature of the problem. Achieving this is difficult,

especially when large, nonlinear, and uncertain dynamic response is involved.

In these circumstances it is noted that to reach the goal of performance-based design the

performance levels need to be defined, which can be done through rigorous performance

evaluation. Hence, it is evaluation of the seismic performance of a structure which is

12
considered as an important step in realizing a reliable and robust performance-based

design. The literature reviewed here mainly focuses on the available techniques on

seismic performance evaluation and ground motion scaling techniques which is an

important aspect of dynamic response evaluation of a structure subjected to seismic

forces.

2.2 Seismic Performance Evaluation

FEMA-350 (2000) provides a reliability-based probabilistic approach to performance

evaluation, considering the uncertainties involved in the judgment and prediction of the

characteristics of the earthquake parameters. In FEMA-273 (1997) four levels of

structural performance are mentioned. In FEMA-273 (1997), both the peak and residual

interstorey drifts are utilized in defining the performance levels as an indicator of

damage. But only two, Immediate Occupancy (IO) and Collapse Prevention (CP) levels

are mostly used in the evaluation of performance. The characteristic parameters of these

two performance levels for Steel moment resisting frames are given in the Table 1.3.

Table 1.3: FEMA-273,1997 - Performance Level for SMRF

Performance Level/Limit State Limit Drift (%) Limit Residual drift (%)
Immediate Occupancy (IO) 0.7 -
Life Safety (LS) 2.5 1.0
Collapse Prevention (CP) 5.0 5.0

Structures designed according to the current design codes are found to undergo

significant inelastic deformation under a strong earthquake which is generally defined in

the form of a response spectrum of the ground acceleration history. The elastic analysis

of structures subjected to seismic actions typically in the form of response spectrum

13
analysis, do not always predict the hierarchy of failure mechanisms. It is also not possible

to predict the amount of energy absorption and the force redistribution pattern that result

from the plastic hinge formation in a structure. This information can only be obtained by

studying the inelastic structural response in the time domain. Inelastic analytical

procedures help to understand the actual behavior of structures by identifying failure

modes and the potential for progressive collapse (Priestley, 2000). Inelastic analysis

procedures basically include inelastic time history analysis and inelastic static analysis

which is also known as pushover analysis. Pushover analysis is very useful in

determining the capacity of a structure, the failure mechanism and the sequence of

yielding. It also forms a basis for many performance-based seismic design procedures

(e.g., Chopra and Goel 1999; Fajfar 2000; Aschheim, 2004; Humar and Ghorbanie-Asl

2005). During the last decade, elastic and inelastic dynamic analyses in the time domain

have been made feasible for complex structures because of the rapidly increasing

computational power and the evolution of engineering software. Linear elastic dynamic

time history analysis is very useful when the dominant modes of vibration are closely

spaced or for multiply supported structures (i.e., bridges) where higher modes are

excepted to be excited due to the random nature of the incoming seismic waves

(Katsanos et al. 2010). The information on behavior of the structure obtained specially

from inelastic structural response in the time domain is critical for the assessment of

existing or new structures of high importance (i.e., tall and high-rise buildings, storage

tanks and nuclear power plants), with complexity (coupled soil–structure systems,

massive and irregular buildings), of high degree of inelasticity (i.e., structures designed to

14
exhibit large deformations), and having geometrical nonlinearity (i.e., base-isolated

structures).

Nonlinear dynamic analysis in time-domain is necessary to capture the response of the

structure to severe ground motion and obtain reasonable estimates of the demands on the

structure. This analysis method is considered to be the most accurate method provided the

structure (and constituent elements) and the seismic input to the structure can be modeled

to be representative of the reality. The ground motion records can be obtained from

natural earthquake records, or can be generated synthetically and artificially.

2.3 Selection of Ground Motion Records (GMR)

2.3.1 Real Accelerograms

The advantage of using real accelerograms is that they are genuine records of seismic

shaking produced by earthquakes. Hence, they carry all the ground motion characteristics

(amplitude, frequency and energy content, duration and phase characteristics) and reflect

all the factors that influence accelerograms. However, the real accelerograms are often

not smooth as compared to that of the target or design response spectrum of seismic

hazard for a given site. In the design codes, the seismic scenario, which is based on a pair

of magnitude, distance and soil conditions, is generally represented by means of a

spectral target shape. Guidance given in seismic design codes on how to select

appropriate real records is usually focused on compatibility with this response spectrum

rather than seismological parameters. Therefore, real earthquake records, which have

similar characteristics (magnitude, distance, site condition, and fault type) with the site

under consideration, have to be selected to match elastic response spectrum given in the

15
code. When selecting the earthquake records, it is desirable to use earthquake magnitudes

within 0.25 magnitude units of the target magnitude (Stewart et al, 2001). Selection of

records having appropriate fault-site distances is important especially for near-fault sites.

Site conditions have a major effect on the characteristics and frequency content of the

strong ground motion records. Even though the ground motions are amplified in soft

soils, the high frequency motions are attenuated. Also, in order to preserve non-stationary

characteristics of the initial time history, it is essential to start with an acceleration time

history whose spectrum is as close to the target spectrum as possible in the period range

of interest. A close initial fit also ensures a speedy convergence to the design values

(Fahjan and Ozdemir, 2008). Although using real earthquake records has many

advantages, there may exist a lack of strong motion earthquake records to satisfy the

seismological and geological conditions and site-specific requirements defined in seismic

codes.

2.3.2 Artificial Records

Artificial accelerograms whose response spectra is closely compatible to the design

response spectra can be generated in either time or frequency domain (Gupta and

Krawinkler, 1999). Artificial spectrum-compatible accelerograms can be generated using

programs such as SIMQKE (Gasparini and Vanmarcke, 1979) and TARSCTHS

(Papageorgiou et al. 2002). The program SIMQKE computes a power spectral density

function from a specified smooth response spectrum and uses this function to derive the

amplitudes of sinusoidal signals which have random phase angles uniformly distributed

between 0 and 2π. The sinusoidal motions are then summed and an iterative procedure

can be invoked to improve the match the target response spectrum, by calculating the

16
ratio between the target and actual response ordinates at selected frequencies. The power

spectral density function is then adjusted by: the square of the ordinate ratio and a new

motion is generated. In order to get other characteristics of artificial spectrum-

compatible record, such as the duration, it is necessary to obtain supplementary

information about the expected earthquake motion apart from the response spectrum. The

computer code TARSCTHS (Papageorgiou et al., 2002) uses non-stationary stochastic

vector processes to generate artificial time histories from a user defined elastic response

spectrum. Here the iterative scheme is applied in the frequency domain where the phase

angles of the desired motion are randomly generated.

2.3.3 Synthetic Accelerograms

The accelerograms were found to be largely unavailable during the earlier decades due to

the absence of data recordings of the earthquakes, hence synthetic records were used

instead of real earthquake records, synthetic accelerograms are mainly made up

deterministic or stochastic ground-motion modeling methods. While short–period

motions behave stochastically, long-period motions mainly behave in a deterministic

manner, where the period of transition from deterministic to stochastic is assumed to be

1.0 s (Stewart.et al. 2001). A number of computer programs were developed for

generating synthetic ground-motions (e.g. Zeng et al, 1994; Beresnev and Atkinson,

1998; Boore, 2003). The simulation is based on stochastic point source approach of

which the specified Fourier spectrum of the ground motion is a function of magnitude

and distance. The simulation model also includes the source parameters characteristic for

the geographic region considered, and takes into account the effect of the magnitude and

distance on the duration of the ground motion summed (with a proper time delay) in the

17
time domain to obtain the ground motion from the entire fault. Using this method,

Atkinson (2009) simulated accelerograms for western and eastern Canada for earthquake

with various magnitudes a wide range of source-to-site distances.

2.4 Spectral Matching and GMR Scaling

2.4.1 Target Spectral Matching
Once an initial search in terms of magnitude, distance and site classification has been

performed, depending on the number of records retrieved, further pruning then needs to

be carried out to acquire the number of records deemed necessary to obtain stable results

from the inelastic dynamic analyses. If there are far more records than actually needed,

the obvious choice would be to apply a second sweep of the search using more restrictive

criteria, such as a smaller distance range or insisting on a close match with the site

classification. There are three methods for further modifying actual time histories to

match the target spectrum. Matching techniques are based on scaling of the selected time

history records in time domain after filtering the actual motion in frequency domain by its

spectral ratio with the design target spectrum; or elementary wavelets are added or

subtracted from the real time history to match a target design spectrum

2.4.2 Ground Motion Scaling in Time Domain

In this approach, recorded motion is simply scaled up or down by a constant scaling

factor uniformly to find out the best matches to the target spectrum over a period range of

interest, without changing its frequency content. It could be stated that the accelerograms

should only be scaled in terms of amplitude. There are procedures which minimize the

differences between the scaled motion’s response spectrum and target spectrum in a least-

square sense (e.g., Nikolaou 1998; Somerville et al. 1997a,b).

18
2.4.3 Spectral Matching in Frequency Domain
A frequency domain matching methodology uses an actual record to produce a similar

motion that matches almost perfectly a target (design) spectrum. In this method, an actual

motion is filtered in frequency domain by its spectral ratio with the design target

spectrum. Fourier spectral amplitudes of an input motion are modified while the Fourier

phases of that remain unchanged during the entire procedure. Preservation of phase

characteristics is important for non-linear time domain analyses, because the non-linear

solution can be sensitive to the phasing of the individual time history. In order to keep the

phases one applies to the signal a real-only "transfer function" (i.e., with a zero-imaginary

component), to rescale the Fourier amplitudes. The technique is repeated iteratively until

the desired matching is achieved for a certain range of periods. The more iterations

results with better compatibility with the target design spectrum (Ozdemir and Fahjan,

2007).

2.4.4 Spectral Matching in Time Domain

One approach for spectral matching is to adjust the original record iteratively in the time

domain to achieve compatibility with a specified target acceleration response spectrum

by adding wavelets having specified period ranges and limited durations to the input time

history. These wave packets are added at times where there is already significant

amplitude in that period range in the time history. This method preserves the overall

phasing characteristics and as the time varying (i.e., non-stationary) frequency content of

the ground motion (Somerville, 1998). The resulting records each have an elastic

response spectrum that is coincident (within a tolerance) with the target spectrum. This

procedure was first proposed by Kaul (1978) and was extended to simultaneously match

19
spectra at multiple damping values by Lilhanand and Tseng (1987). Although this

procedure is more complicated than the frequency domain matching procedure as in most

cases it can preserves the non-stationary character of the reference time history.

Abrahamson (1992) developed RSPMATCH software modifying the Lilhanand and

Tseng (1987) algorithm that preserves non-stationary character of the reference ground

motion for a wider range of time histories. Mukherjee and Gupta (2002) proposed a

method in which the accelerogram is divided into a finite number of time-histories of

distinct frequency bands, the time-histories are then iteratively scaled to match the target

spectrum.

20
Figure 2.1: The use of real earthquake accelerograms (Bommer and Acevedo, 2004)

21
2.5 Modal Pushover Scaling (MPS )
MPS scaling procedure was developed and proposed by Kalkan and Chopra (2011). This

method serves the purpose of scaling earthquake records near a fault site where inelastic

spectral deformation dominates predominantly over the corresponding elastic spectral

deformation (Bozorgnia and Mahin 1998; Alavi and Krawinkler 2000; Baez and Miranda

2000; Chopra and Chintanapakdee 2004). It is known to explicitly consider the strength

of the structure obtained from the first-mode pushover curve and determine the scaling

factors for each record to match a target value of the deformation of the first-mode

inelastic SDF system estimated by standard procedures. The MPS method is further

considered to be appropriate for analysis of first-mode dominated structures and is found

to be sufficient to accurately estimate the seismic response of low-rise to mid-rise

structures.

The intensity-based scaling procedures provides scale factors for a small number of

ground-motion records, so that nonlinear Response History Analysis (RHA) of a structure

for these scaled records remains reliable to estimate the median value of the seismic

demand parameters (SDPs) such that the record-to record variations in the EDP

(Engineering Demand Parameters) is kept low. However, none of the procedures like

scaling based on PGA (Peak Ground Acceleration), intensity or peak velocity take into

account the properties of the structure to be analyzed. As the intensity-based scaling

depends on elastic responses of the structure, it is found to produce inaccurate estimates

22
with wide dispersion in Seismic EDP values for structures responding in the nonlinear

range (Nau and Hall 1984; Miranda 1993; Vidic et al. 1994; Shome and Cornell 1998).

In this context a recent study conducted by Kalkan and Chopra (2011) considering three

sets of 7 ground motions scaled by MPS procedure and the code recommended

ASCE/SEI 7-05 procedure — for a 4 , 6 , and 13 storey Steel Moment Resisting Frame

(SMRF) buildings showed that the median values of EDPs like interstorey drift, and floor

displacement obtained from MPS procedure were within admissible dispersion levels of

about 20%, whereas the the EDPs from the code based procedure overestimated the

demand ranging by 20% to 50% in the 4- and 6-storey buildings and about 50% for the

13-storey building. Thus, the MPS method is deemed an accurate and efficient procedure

as compared to the ASCE/SEI 7-05 scaling method. Extension of the MPS method to

include higher vibration modes is expected to provide improved estimates for mid-rise

and high-rise buildings (Tothong and Cornell 2008; Tothong and Luco 2007; Luco and

Cornell 2007).

2.6 Steel Moment Resisting Frame (SMRF)

A large number of modern high-rise buildings have Steel Moment Resisting Frame

(SMRF) as the primary earthquake resisting system. This type of construction was

considered as an efficient way to resist lateral forces induced during earthquakes since

the steel elements are expected to be able to sustain large plastic deformation in bending

and shear. However, the failure of more than 150 SMRF during the 1994 Northridge

earthquake and the 1995 Kobe earthquake primarily in the form of brittle fractures at

welded beam to column connections raised serious concerns regarding the seismic

23
behavior of code compliant SMRF structures. The critical issues were then broadly

classified into the following three points (Gupta and Krawlinker, 1999):

 The observed behavior of SMRF structures was found to be largely deviant from

the expected code compliant designs.

 The immediate need for predicting the seismic demand for the very large number

of existing SMRF structures in different geographic locations and under different

levels of shaking to enable retrofit and rehabilitation if required.

 The immediate need to predict the structural safety at various seismic hazard

levels due to potential connection fractures.

The solution for the existing problem highlighted can be brought about through in-depth

understanding of the basic factors controlling the seismic behavior of SMRF structures.

The answers or solutions developed should provide an estimate of the structural

performance and the reliability for the very large inventory of existing SMRF systems in

order to facilitate the decision process for the seismic rehabilitation of these structures to

acceptable performance levels, thus there is a pressing need for a systematic evaluation of

SMRF structures in order to better understand the core structural behavior characteristics

and address performance expectations at different hazard levels .

2.7 Review of NBCC 2010 Code Provisions

According to NBCC (2010) the minimum lateral earthquake force V, is calculated by

using the following equation -

S (2.0) M V I EW
V …………….………………...……………….. 1.1
Rd R0

24
where S(Ta) is the spectral acceleration corresponding to the building’s fundamental

period Ta; MV is the factor to account for multistorey effect, IE is the importance factor,

W is the total weight of the building, Rd ductility related force modification factor, R0 is

the over strength related force modification factor. Table 1.4 shows the revised value of

higher mode factor MV in NBCC 2010.

Table 1.4: Values of higher mode factor Mv for various structural systems (NBCC 2005)

TYPE OF LATERAL RESISTING

Sa(0.2)/Sa(2.0) SYSTEMS Mv for Ta<1.0 Mv for Ta>2.0
Moment Resisting Frame or " coupled
1.0 1.0
walls "
Braced Frame 1.0 1.0
<8.0
Walls, wall-frame systems , other systems 1.0 1.2
Moment Resisting Frame or " coupled 1.0 1.2
walls "

Braced Frame 1.0 1.5

>8.0
Walls, wall-frame systems , other systems 1.0 2.5

Note : Linear Interpolation should be used for intermediate values

The design spectral response acceleration values S(Ta) is given by the following formula,

which holds good for linear interpolation for intermediate values.

S(Ta) = FaSa(0.2) for T≤ 0.2 s

= FvSa(0.5) or FaSa(0.2) whichever is smaller for Ta= 0.5 s

= FvSa(1.0) for Ta= 1.0 s ……………………………………………(1.2)

= FvS (2.0) for Ta =2.0 s

= FvSa(2.0)/2 for Ta≥4.0 s

25
where S(Ta) is the 5% damped spectral response acceleration expressed as a ratio to the

acceleration due to gravity, g for a period Ta; Fa is an acceleration based site coefficient,

and Fv accounts for velocity based site coefficient. The lateral load distribution along the

height of a building is given by the Eq. 1.3

(V  Ft )Wx hx
Fx  n
…………………………… ……. (1.3)
W h
i 1
i i

Fx is the lateral force applied at level x, n is the total number of storeys, hx and hi are the

heights above the Ground level to level x and i, respectively. Ft is given by the Eq. 1.4

considering the portion of V concentrated at top storey.

Ft = 0.07TaV < 0.25 V …………………………………………. (1.4)

Ft= 0 ( Ta < 0.7 s) …………………….…………………………… (1.4a)

The graph of spectral acceleration versus period for Vancouver as given in NBCC 2010 is

shown in Figure 1.3, and the design values of spectra for Vancouver are shown in Table

1.2. The spectral values in between the periods reported in Table 1.2 are obtained by

linear interpolation.

Table 1.5: Design Spectra of NBCC 2005 (Adams and Atkinson, 2003)

Location Sa(0.2) Sa(0.5) Sa(1.0) Sa(2.0) Sa(≥4.0)

Vancouver 0.96 0.66 0.34 0.18 0.09

26
 1
0.9
0.8
0.7
0.6
S(g)

0.5
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6

T(s)
Figure 2.2: Design Spectra for Vancouver (NBCC 2010)

2.8 Summary
Performing fully dynamic nonlinear structural analysis is time-consuming (and therefore

costly) in engineering practice. The use of spectrum-compatible records, which allow

fewer analysis runs to be made, will perhaps be preferred by design engineers. There is

no consensus yet on the number of real accelerograms required to obtain stable measures

of inelastic response from time history analysis. It is generally recommended that a suite

of seven to ten records are sufficient (Bommer and Acevedo, 2004). Currently, there is a

lack of knowledge in the area of damage potential due to the effect of spectral matching

and scaling of GMR (Leger and Tremblay, 2009). One of the goals of the present

research is evaluate the currently available scaling methods for GMRs used in RHA and

investigate their implications on the seismic response.

27
 Chapter 3

Design of Steel Moment Resisting Frames

3.1 Introduction

The buildings chosen for the performance evaluation and the research presented here are of steel

moment resisting frames, located in Vancouver, Canada. The Vancouver region in Canada is

classified as high seismic zone as compared to the other parts of the country. Four buildings of

five, ten, fifteen and twenty storey height, symmetrical along the vertical center line of the steel

frames are designed according to the seismic provisions of NBCC 2010. Each building has a

series of frames in the east-west (E-W) direction and three bays in the north-south (N-S)

direction. The center to center spacing of the frames in the E-W direction is 6 meters whereas in

the N-S direction the two exterior bays are of 9 meters and the interior bay is 6 meters. The first

storey height in the buildings is 4.85 meter and the remaining floors are spaced at 3.65 meter

each. A typical layout plan is shown in Figure 3.1, the elevation views are shown in Figures 3.2

and 3.3. The building frames along the north-south direction have been chosen for the design but

the effect of accidental torsion is not considered in the process as the building is configured in a

symmetrical and regular geometric manner.

28
 9m

6m
9m

6 @6m c/c

Figure 3.1 - Layout Plan of the Building

9m 6m 9m
9m 6m 9m
10 15 10 20 30 40
5
4
0
9 14 9 19 29 39
8 18 28 38
[email protected]

3 8 13 [email protected]
2 7 12 7 17 27 37

1 6 11 6 16 26 36
5 15 25 35
4 14 24 34
4.85 m

5 Storey SMRF
3 13 23 33
2 12 2231 32
1 11 21 31

10 Storey SMRF 4.85 m

Figure 3.2 - Elevation of 5 & 10 Storey frames

29
 9m 6m 9m

9m 6m 9m

[email protected]
[email protected]

4.85 m
4.85 m

15 Storey SMRF 20 Storey SMRF

Figure 3.3 - Elevation of 15 & 20 Storey Frames.

30
The equivalent static force-based design process as per the code has been adopted for the design

which involves the calculation of member forces prior to designing the members as per the

specified provisions. NBCC 2010 specifies a seismic hazard level under which probability of

exceedance is of 2% in 50 years.

3.2 Structural analysis and design

A number of commercial and non-commercial software programs are available for linear and

nonlinear analysis of structural systems. Some of the popular ones include DRAIN-2DX

(Prakash et al., 1993), DRAIN-3DX (Static and Dynamic Analysis of Inelastic 3D Structures),

DRAIN-BUILDING (Static and Dynamic Analysis of Inelastic Building Structures), SAP-2000,

ETABS and PERFORM 3D (Computer and Structures, 2012). To facilitate the nonlinear

dynamic analysis of the 2D frame structures considered for the research, the DRAIN-2DX

software has been employed since it is known to produce reliable results and simple to use for

plane structures. The program has been used in the present research to carry out the response

spectrum analysis of the frames in the unstressed state, the modal analysis to compute the

fundamental frequency of the frames, the pushover analysis to determine the maximum

deflection and strength of the structure as well as the nonlinear dynamic analysis in the time

domain. The modeling of the structure in the DRAIN 2DX software is done by defining the

planar coordinates of the frame. The beam members of the same floor level are grouped in the

same section type and the column sections are changed at an interval of five floors, i.e. columns

are spliced at every fifth floor. The modeling of the frame has been done using Element type- 2

(Beam-Column Element) available in DRAIN-2DX element library. It also allows for defining

the effect of axial force on bending strength by considering the P-M interaction curve and the

31
yield surface. Element Type 2 is shown in the Fig 3.4, which possess Linearized Geometric

Stiffness and allows for axial-flexural interaction. The connections of beam to column are

modeled as rigid zones as defined in FEMA 440 under predefined connections.

Plastic hinge Rigid Zone

Elastic Vj Node J
element
Mj
Vi
Pj Yi

Node I Mi
Xj
Pi
Yi
Xi
Y
DRAIN 2DX Beam-Column Elastic element
X

Figure 3.4 - Model of DRAIN-2DX Element Type-2 (Prakash et al., 1995)

3.3 Design of Steel Moment Resisting Frames ( SMRF )

The equivalent static lateral load procedure for the seismic load as prescribed by NBCC 2010 has

been used in the preliminary design the buildings which then revised using the modal and

response spectrum analysis. Building frames are designed to satisfy the NBCC 2010

requirements and the steel structural elements have been designed as per CSA S16-09 (CSA,

2010). The following loadings have been considered in the design of the buildings: gravity

32
loads (dead load (D), live load (L)) and seismic load (E). The dead loads comprise the self-

weight of the frame elements and other non-structural components the live loads are obtained as

per the specification from NBCC 2010. Table 3.1 gives the values of Dead loads and Live loads.

The total weight of the building has also been calculated using the static design procedure and is

found to remain constant at each iteration of the static design process. Live load at the roof is

mainly the snow load (S).

Table 3.1:Design loads.

Dead Load (kPa) Live Load (kPa)
Roof Floor Roof Interior typical floor Corridor
3.4 4.05 2.32 2.4 4.8

Linear static analysis of frames has been performed using DRAIN-2DX to determine the

member forces. Load combination of the forces has been used to evaluate the design force for

both beam and column of the frames. The combinations of different loads are given in Equations

3.4 and 3.5.

1.25D+1.5L ...........................................................3.4

1.0D± 1.0E +(0.5L+0.25S) ...................................3.5

In static analysis-based design it has been checked that the structure designed to be safe for the

combination of gravity loads also remains safe to withstand the earthquake loads. In the case

where structure is designed for gravity load fails to withstand the seismic load, the design has

been further modified to satisfy the both load combinations in Equations 3.4 and 3.5. During the

design process, the empirical fundamental periods of the frames has been calculated by using the

Equation 3.6.

Ta = 0.085(hn)3/4 (NBCC, 2010 ) …………………………………….. 3.6

33
where, Ta is the empirical fundamental period and hn is the total height of the building frame.

This period has been used to calculate the equivalent seismic force for the first iteration of the

static design. After designing of frames by using the empirical fundamental period, a detail

modal analysis of the frames has been carried out. The fundamental periods of frames obtained

from the modal analysis if found to be more than Ta obtained from the empirical period using

Equation 3.6, the seismic force is revised using the modal period or 1.5Ta, whichever is smaller

(NBCC, 2010 ). A summary of the periods of different frames is presented in the Table 3.2

Table 3.2: Fundamental Periods of the Buildings.

By Empirical Equation Period from Modal 1.5Ta, s

Frame Height Analysis,
(Eq.3.6), (Ta), s
5 storey 0.787 1.412 1.181
10 Storey 1.293 2.528 1.939
15 Storey 1.739 3.571 2.609
20 Storey 2.149 4.789 3.224

The design base shear is calculated by using Equation 1.1 as provided in NBCC 2010. The base

shear is then distributed along the height of the building in the form of inverted triangle as per

code requirement, and the lateral force is computed for each storey level according to the weight

and the height at the storey level. Seismic force at the specified storey level is computed by using

Equation 1.3 and the non-linear static pushover analysis is performed including the effect of P-Δ

to determine the capacity and yielding sequence of the structure. The buildings are considered to

be of normal importance and the frames are assumed to be ductile.

The equivalent base shear of the four buildings is given by the Eq.1.1 in which the parameters

are IE (importance factor )=1.0, MV (factor for higher mode effect )=1.0, Rd (ductility factor) =5.0

34
and the R0 (overstrength factor)=1.5. The soil is of type- C is assumed which is dense soil with

soft rock, hence Fv=Fa=1.0 (site specification factor). Further, the design spectral acceleration

value S(T) is equal to the spectral acceleration value Sa(Ta) provided in the code. The design

values of the base shear of the four buildings as determined using the empirical period are as

shown in Table 3.3.

Table 3.3: Base shear of SMRF.

Steel Moment Base Shear V (KN)

Resisting Frame Bare Frame
5 storey 154.70
10 Storey 192.44
15 Storey 293.75
20 Storey 400.96

However, the fundamental period determined by the modal analysis and 1.5Ta whichever is

smaller is used for recalculating the base shear. If any variation is found in the base shear, the

design of the buildings is revised with the new base shear and the sections of the frames are

modified suitably.

3.4 Modal Analysis

The software DRAIN-2DX has been employed to obtain the fundamental frequency and the

mode shapes of the frames. The mode shapes for different frames are shown in Figs. 3.5 and 3.6.

35
 F IR S T M O D E SEC ON D M OD E T H IR D M O D E F O UR T H M O D E
11 11 11 11

10 10 10 10

9 9 9 9

8 8 8 8

Node Number
7 7 7 7

6 6 6 6

5 5 5 5

4 4 4 4

3 3 3 3

2 2 2 2

1 1 1 1
- 1.0 - 0 .5 0 .0 - 1. 0 - 0 .5 0 .0 0 .5
1. 0 - 1. 0 - 0 .5 0 .0 0 .5 1. 0 - 1.0 - 0 .5 0 .0 0 .5 1.0
MODAL DISPLACEMENT MODAL DISPLACEMENT M O D A L D I SPLA C EM EN T M OD A L D ISPLA C EM EN T

(a)
F IR S T M O D E SEC ON D M OD E T H IR D M O D E F O UR T H M O D E
6 6 6 6

5 5 5 5
Node Number

4
4 4 4

3
3 3 3

2
2 2 2

1
1 1 1
- 1.0 - 0 .5 0 .0
- 1. 0 0 .0 1. 0 - 1. 0 0 .0 1. 0 - 1. 0 0 .0 1. 0
M OD A L D I SPLA C EM EN T M O D A L D I SPLA C EM EN T M O D A L D I SPLA C EM EN T M O D A L D I SPLA C EM EN T

(b)

Figure 3.5: Mode Shapes of 10 and 5 Storey Building Frames; (a) Mode Shapes of 10 Storey

Building Frame, (b) Mode Shapes of 5 Storey Building Frame.

36
 SEC O N D M O D E T HI R D M O D E F OU R T H M OD E
21 21 21

17 17 17

13 13 13

9 9 9

5 5 5

1 1
1
-1.0 -0.5 0.0 0.5 1.0 -1.0 -0.5 0.0 0.5 1.0 -1.0 -0.5 0.0 0.5 1.0
M o da l D is pla c e m e nt Modal Displacem ent Modal Displacem ent

(a)

F IR S T M O D E SEC ON D M OD E T H IR D M O D E F O UR T H M O D E
16 16 16 16

13 13 13 13
Node Number

10 10 10 10

7 7 7
7

4 4 4
4

1 1 1
1
- 1.0 - 0 .5 0 .0 0 .5
1.0 - 1. 0 - 0 . 6 - 0 . 2 0 . 2 0 .6 1. 0 - 1.0 - 0 .5 0 .0 0 .5 1.0
- 1.0 0 - 0 .50 0 .0 0
M ODAL DISPLACEM ENT M ODAL DISPLACEM ENT M O D A L D IS P LA C E M E N T M O D A L D IS P LA C E M E N T

(b)
Figure 3.6: Mode Shapes of 20 and 15 Storey Building Frames; (a) Mode Shapes of 20 Storey

Building Frame, (b) Mode Shapes of 15 Storey Building Frame

Table 3.4 depicts a sample calculation of the base shear post the modal analysis.

37
 Table 3.4: A sample calculation of base shear after modal analysis.

Modal 1.5 Ta Selected S(1.0)g S(2.0), g S(Design), g Mv (1.0) Mv (2.00)

Period Period
4.789 3.223 3.223 0.340 0.180 0.180 1.00 1.10
Mv Weight, Factor Factor Base shear after modal Base shear before modal
(Design) W (kN) Rd R0 analysis, Vm (kN) analysis, Vs (kN)

1.10 15187 1.50 5.00 400.96 601.44

In the steel frame design it has been checked that the selected design base shear is greater than

or equal to the base shear calculated for spectral acceleration S(2.0)g and less than 2/3 of base

shear corresponding to acceleration S(0.2)g. Ductile frames of Type-D are designed as per the

guidelines presented in the CISC (2010) Handbook of Steel Construction and the steel sections

used in the design are of CSA G40.21 grade with yield strength, Fy=350 MPa and modulus of

elasticity (E) = 200x103 Mpa ,for both beam and column. The columns in the ductile frames are

designed as beam-column elements to avoid yielding and flexural buckling.

The column strengths are further computed by using the Equation 3.7 as prescribed in CISC

(2010) which is applicable to plane and unbraced frames.

Cf 0.85U 1x M fx
  1.0 ……………………………3.7
Cr M rx

In the Equation 3.7, the constants U1x is = 1.0 in case of the unbraced frames. The factored

moment (Mfx) and the factored axial compressive force (Cf) are obtained from the analysis. The

resistive bending moment (Mrx) and the resistive compressive axial force (Cr) for the individual

columns are taken from the CISC (2010) Handbook.

38
The steel beams are designed to comply to the limit states specified in CAN/CSA-S16-01(2001)

and the computed factored resistance is compared to the specified factored resistance using the

condition Vr>Vf and Mr>Mf, where the factored beam shear (Vr) and moment resistant (Mr) are

obtained from the CISC (2010) Handbook. The design iteration is carried out till the condition is

satisfied. The deflection in beams has been checked for live and dead loads to satisfy the

serviceability limit state, and the deflection has been calculated by using Equation 3.8.

I reqd  WCd Bd
………………….………………. 3.8
  I reqd I  m

where Ireqd is the required moment of inertia of area, I is the gross moment of inertia, Δm is the

specified maximum deflection, Δ is the computed deflection, Cd is the deflection constant, Bd is

a constant pertaining to the load and support, and W is the total live load on the beam. In light of

the shake-down condition under which the frame system behaves elastically after initially

yielding in case of cyclic loading, the mandatory check for capacity-based design the column

and beam strength at the shake-down condition have been computed by using the formulae

given in CISC (2010) Handbook.

  d 
M rc   1.1 R y M pb  Vh  x  c   ……………………3.9
 2 


 Cf 
M rc  1.18  M pc 1    M pc ………………………3.10
 C y 
 

where Mrc and Mpb are the plastic moment of resistance of the column and the beam

respectively. Ø is the resistance factor, Vh is the shear acting upon the plastic hinge location

39
when plastic hinging occurs, Cf is the factored axial compressive load of column, Cy is the axial

compressive load at yield. Ry is a factor applied to yield stress Fy to estimate the probable yield

stress and Fy is the minimum specified yield stress. It is noted that the shake-down condition in

frames causes columns to carry all the loads, resulting in formation of plastic hinges mainly in

the beams at a certain specified distance from column center line and where the distance

depends on the type of connection of beam and column. The distance of plastic hinge from the

center of the column for the connection chosen for the presented frame is x+dc/2 (Fig.2.1 and

2.2), in which, dc is the depth of column and x is the distance between the plastic hinge and the

face of the column. It has been checked that all joints of every frame considered here have

satisfied this capacity design criteria. As a part of the dynamic analysis, a response spectrum

analysis of each frame has been performed to determine the base shear and the design base shear

is reduced further according to the NBCC provision. The finalized sections for different elements

of the frames are presented in the Table 3.6 and 3.7.

Table 3.5 - Sections of Columns.

Building Column Storey Storey Storey Storey

Height Row 1 to 5 6 to10 11 to 15 16 to 20
5 Storey External W310x179
Internal W310x253
10 Storey External W310x283 W310x158
Internal W310x314 W310x202
15 Storey External W310x283 W310x253 W310x179
Internal W360x314 W360x260 W310x283
20 Storey External W310x283 W310x253 W310x202 W310x179
Internal W360x314 W360x287 W360x262 W360x262

40
 Table 3.6 - Section of Beams.
Building Height
Storey Level
5 Storey 10 Storey 15 Storey 20 Storey
Top Storey W310x79 W310x79 W310x107 W310x107
Other Storey W310x86 W310x107 W310x129 W310x129

A flow chart of the above design methodology is presented in the Figure 3.8. It has been adapted

from (Hannan, 2006) and has been revised for the present work. The flowchart is described here

briefly.

Step1 – Select the member sections for the Steel moment resisting frame based on experience

and proceed further with the design and recalculation process.

Step 2 – Compute the empirical fundamental period of the structure along with other design

parameters to finally obtain the design base shear, distributed base shear as along the height of

the structure to get the lateral force.

Step 3 - Perform the Static analysis for the given loads and load combinations using the

DRAIN-2DX program and obtain the shear, bending and axial forces for individual member .

Step 4 - Check the obtained member forces with the code specified values , if the members pass

the check , further perform the modal analysis of the structure and revise the base shear with the

new fundamental period and proceed with Step 5, if the members fail the test redesign the

structure with new sections and repeat the design procedure from Step 1.

41
 Choose Initial member
sections

Compute the empirical fundamental period,

design parameters and the design Base shear.
along with the distributed lateral force from
the the base shear

Perform the Static analysis for

different loads by DRAIN-2DX program

Obtain member forces (Shear, Bending

Moment, Axial force) for different loads

Determine the design member forces from

the different load combinations

Check the member forces against the

code specified values.

No OK
Yes

Perform MODAL ANALYSIS and Recalculate base shear to revise the

obtain the fundamental period of design with new base shear
the structure

Check the ductility demand for strong column/weak beam criterion

Pushover No
OK
analysis
Yes
Dynamic analysis Determine the revised base shear and
(Response Spectrum) drift demand
No

End Finalize Member Design is OK

Sections Yes

Figure 3.7: Flow-chart for the design and evaluation of Steel Moment Resisting Frames

42
Step 5 – Check for ductility demands of the individual members in the structure designed, if the

members pass the ductility tests proceed with Step 6, else redesign the structure from Step 1.

Step 6 – Conduct the Non-linear static pushover analysis and response spectrum analysis on the

structure designed and compute the base shear to obtain revised values, calculate the drift

demand and check if the drift demands are within the code specified or acceptable limits.if the

drift demands are within limits revise the sections of the members if required and finalize the

design sections, if the results are unsatisfactory redesign the structure from Step 1.

3.5 Modal Analysis using a 3D model in ETABS Software

A 5 storey building was considered to be analyzed using the ETABS software in order to

compare the modes from the DRAIN 2DX modal analysis. A 3D model of the building is

developed using in the ETABS software and modal characteristics of the building model have

been compared to the 2D model developed in the DRAIN-2DX model in order to establish the

validity of the 2D models. ETABS is a reputed software in the as per the industry standards

known for its innovative features and reliability in Building analysis and design. It provides the

user with an Integrated Building Analysis and Design Environment. The software can analyze

variety of structures including Moment Resisting Frames, Braced Frames, Staggered Truss

Systems, Frames with Reduced Beam Sections or Side Plates, Rigid and Flexible Floors, Sloped

Roofs, Ramps and Parking Structures, Mezzanine Floors Composite or Steel Joist Floor Framing

Systems. ETABS is easy to use in designing a simple building or for performing a dynamic

analysis of a complex high-rise building that utilizes non-linear dampers for inter-storey drift

control.

43
 3.5.1 Modeling

The modeling phase in ETABS involves the representation of the entire structure by elements to

which physical and material properties are assigned. This building has 5 stories with 19.45m

meter height. The dimension in X direction is 36 m and in Y direction is 24 m. Fig 3.8 shows the

plan vie of the building in ETABS and Fig.3.9 shows the 3d view of the model structure in

ETABS

Fig.3.8 – Plan view of the building in ETABS

44
 Fig.3.9 3-Dimensional view of the structure in ETABS

3.5.2 Loading

The modeling phase in ETABS involves the representation of the entire structure by elements to

which physical and material properties are assigned. This building has 5 stories with 19.45m

meter height. The dimension in X direction is 36 m and in Y direction is 24 m, the loads given in

Table 3.1 has been adopted for the ETABS analysis and the Load combination has been chosen

as per the equations 3.4 and 3.5, the base shear distribution or the lateral load distribution in the

5 storey SMRF at each storey level has been given in the Table 5.8 .

3.5.3 ETABS Analysis

Comparison of Mode periods from the DRAIN 2DX software and the ETABS software has been

given in the Table 3.7. It is observed from Table 3.7 that the modal frequencies of the building

45
obtained from the 3D model (ETABS) and the 2D model (DRAIN 2DX) are in good agreement.

This indicates that the 2D models can provide acceptable results. This expected in all the

buildings as all of them have symmetrical plan. For the static and dynamic analysis, the 2D

models are also expected to give acceptable results since the loads are applied symmetrically and

structures, which are symmetrical would deform in a symmetrical manner. The effect of

accidental torsion is expected to be minimal as suggested in NBCC 2010. Hence the static and

dynamic analysis of all the buildings considered here are performed using the 2D models in the

DRAIN 2DX software.

Table 3.7 – Modal periods from DRAIN 2DX and ETABS .

Modal Periods DRAIN 2DX ETABS

1 1.4123 1.396

2 0.42427 0.3782

3 0.21304 0.20024

4 0.12495 0.13452

5 0.084307 0.11047

3.6 Summary

The model of the building and the layout plan selected is symmetrical along the X and the Y axis

hence a 2D analysis holds good for analyzing the effect of gravity, lateral and earthquake

induced forces, ductile moment resisting frames are used in the building with direction of

secondary beams running from right to left, and to allow for non-linear deformations. For the

design of the building the frame or the bay in the N-S direction has been considered and Class I

columns have been used in the design and has been shown to be in agreement with the CSA-S16-

46
09 standards, as the building is symmetrical in both the directions the columns have been spliced

at every 5 floors instead of 3 floors and it is found to be optimal for design purpose, the rotation

demands and the load carrying capacities of beams and columns of the ductile moment resisting

frame have been checked to satisfy the design . The building was further modeled in ETABS

under gravity, lateral and earthquake forces to verify and the check the results from the DRAIN

2D analysis and the results were found to be in agreement from both the cases .The designed

members have been strictly checked to satisfy the guidelines from CSA (2009) for CSA-S16-09.

47
 Chapter 4

Selection and Scaling of

Ground Motions
4.1 Introduction

Non-linear dynamic analysis in time domain performed as part of performance-based

design requires recorded acceleration as input data. The ground motion records selected

for the analysis are expected to possess all the characteristics of a real earthquake records

anticipated at the given site. Seismic waves, however are found to traverse in a complex

path from the source or the plane fault to different sites, and are considered to have

random characteristics in space and time. The code guidelines, on the other hand, are

found to be simple and inadequate to serve as a guide for selecting earthquake records as

it underestimates the potential effect of selecting appropriate records to be used for

performance evaluation of structures. Furthermore, when time history analysis is

performed, the response of a structure is captured for a suite of different ground motion

records to obtain a reasonable engineering solution.

47
4.2 Selection of Ground Motion Records (GMR)

The presently available literature on the selection of earthquake records are deficient in

engineering standards and the responsibility of selecting appropriate representative

records for dynamic analysis rests upon the design engineer who in turn depends on the

data of the seismic hazard at the site of the interest. Very few building codes which

requires the use of dynamic analysis of structures, may not necessarily address the critical

aspect of the number and type of records to be used in the analysis. The codes that

specify the same are rare and the most common recommendation is for 3 records where

maximum structural response must be used as per Eurocode 8 (EC8, 2004) and ASCE 7-

05 (ASCE, 2005). However, Reyes and Kalkan (2011) suggested that seven or more

records are preferable. The following figure (Fig. 4.1) shows a flowchart outlining the

available options for Seismic Hazard Analysis (SHA) procedures that can be performed

by an engineer. In the Fig.4.1 refers M,R,Ԑ refer to Magnitude (M), Source to site

distance (R), and the soil profile at the specific site, respectively; and Xi, Mi and Ri

indicate the selected and required number of records based on soil properties, Magnitude

and Source to site distance, respectively .

48
 Selection of records 1 GMR Selection
through DSHA
0.9
through PSHA

Accelaration in g,
Accelaration in g,
0.8
0.7
0.6

S(g)
0.5
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6

T(s)
Period in second, T(s)
Period in second, T(s)
NBCC 2010 SPECTRUM

Disaggregation

(M, R, Ԑ)

Site
Synthetic Classification Selection of GMRs in
terms of spectral
matching

Selection of GMRs in terms of

seismological parameters

Select X records Xi, Mi, Ri

Acceleration in g,
S(g)

T Period in sec, s

Figure 4.1: Available Selection procedures for strong motion records Bommer and

Acevedo (2004)

49
The primary selection of records itself is usually carried out from the available data banks

and are generally based on the engineering characteristics of an earthquake like

magnitude and distance, strong–motion criteria, and site soil conditions. For a detailed

study of seismicity on a particular area or for a structure of importance, a Seismic Hazard

Analysis is required to be carried out. The hazard estimation, if carried out

deterministically by assuming the design earthquake scenario in terms of magnitude,

source to site distance, and the site-soil conditions, the process is referred to as

Deterministic Seismic Hazard Analysis (DSHA). On the other hand, if the earthquake

scenario is explicitly calculated by the method of any disaggregation techniques

available, the process is referred to as Probabilistic Seismic Hazard Analysis (PSHA). In

the DSHA approach, the strong motion parameters are estimated for the most severe

ground motions at the specified site, considering the nature of the site soil geology, the

distance from the site to the fault zone, and the data from the past earthquakes. The most

important aspect of the DSHA process is to carefully estimate the “maximum credible or

the “design scenario earthquake “based on the seismic zones and the seismo-tectonic and

geological features of the source zone. As process is said to involve inherent ambiguities,

it has a very low probability of occurrence, and in some cases, the hazard levels are found

to be impractical to be used in the analysis for cost based economic feasibility studies.

The PSHA process overcomes the limitations of DSHA in predicting the probability of

occurrence. It is the most common method used for SHA originally developed by Cornell

(1968). The probabilistic approach serves as an excellent method for risk management

and in economic feasibility studies by taking account of the frequency or probability of

exceedance of the earthquake against the design life of a structure or facility. PSHA gives

50
the cumulative seismicity of a particular site for a given period to estimate the strong

motion parameters of an earthquake, the result of which is a site specific uniform hazard

spectrum. The required engineering parameters of an earthquake i.e., Magnitude (M),

Source to site distance (R) and ground motion deviation which are not available from the

resulting hazard curves, are obtained by the process of disaggregation of the design

earthquake scenarios. The method proposed by McGuire (1995) for disaggregation has

been widely used. The selection process for the ground motion records, in general, can

also be classified broadly under the following categories.

a. Selection based on Magnitude (M) and Distance (R)

b. Selection based on Site Soil Conditions

c. Selection based on Spectral Matching of Strong motion parameters

 Evaluation of a/v (i.e., peak ground acceleration to velocity) ratio

 Effect of duration of ground motion shaking

4.2.1 Selection based on Magnitude (M) & Distance (R)

It is the most commonly used parameter for initial ground motion selection, where the

magnitude of the earthquake selected is recommended to have the same value or +/- 0.2

to that corresponding to the target spectrum. The magnitude of the earthquake is found to

have considerable effect on the demand of the structure as it influences the duration and

shape of the response spectrum strongly. Although there are many techniques available

for altering the shape of the response spectrum, it is necessary to keep the shape of the

response spectrum in close agreement to that of the target spectrum. For this reason, the

magnitude of the earthquake selected is also an important parameter, which is usually

51
used in conjunction with the source to site distance from the fault zone to the site in

consideration to form a pair of selection parameters in the initial search of records.

However, the spectral shape is found to be less sensitive to distance (R) than to that of

magnitude. A collection of ground motion records sorted with magnitude and source-

distance (M, R) are usually referred to as a bin of records. In this methodology the

variation between the record and scenario magnitudes is recommended to be closely

spaced. Stewart et al. [2001] suggested a magnitude half-bin width of +0.25 M, while

Bommer and Acevedo [2004] recommended + 0.20 M. Therefore, while searching for

real records, the search parameters for the magnitude is spaced closely, and widened for

the distance range, if required.

Nevertheless, recent studies have questioned the effectiveness of (M,R) based selection

method because of the deviating and unreliable results in structural response observed

after direct use of earthquake record sets based on this particular criterion as input to non-

linear dynamic analysis. The source to site distance derived from the earthquake

scenarios has been proven to be an inadequate predictor of structural response. However,

in spite of the noted shortcomings, the method is largely familiar and in use by structural

engineers.

4.2.2 Selection based on Site Soil Conditions

One of the important parameters for selection of earthquake records is the soil profile at

the site of the interest. The soil strata classification parameter generally refers to the top

30 m and is said to influence the amplitude and shape of the response spectra to a large

extent. Boore (2004) has stated that soil strata much below 30 m also influences the

52
response spectra, the parameter is more often used in conjunction with the M, R

selection criteria. In such a scenario, it is observed that the M, R, S selection criteria

greatly reduces the number of records selected for a dynamic analysis; in which case, the

selection of records from similar soil database is recommended.

4.2.3 Selection based on Spectral Matching of strong motion parameters

As recommended in most building codes, one of the most important criteria for a selected

earthquake record is that it is a representative ground motion observed at a suitable

source to site distance. This specification directly relates to the compatibility of the

record to the target spectrum rather than the seismological criteria at the site specified.

This situation has given rise to the selection based on spectral matching as a prominent

method wherein selection of real accelerograms is obtained on the basis of degree of

compatibility to the corresponding ‘target’ spectrum as provided in the the relevant

building code or through a seismic hazard analysis. Ambraseys et al. [2004] proposed Eq.

(4.1) as a means to verify spectral compatibility of a given record with the target

spectrum for the European strong-motion databank.

( ) ( )
√∑ ( ) (4.1)

In the above equation, N is the number of periods at which the spectral shape is specified

Sα0(Ti) is the spectral acceleration of the record at period Ti, Sαs(Ti) is the target spectral

acceleration at the same period, while PGA0 and PGAs are the peak ground acceleration

of the record and the zero-period anchor point of the target spectrum, respectively. A

53
small value of Drms indicates a close match between the spectral shape of the recorded

motion and the target spectrum. In general, the value of Drms depends on the size of the

earthquake record databank and the number of records required. It is also dependent on

the period range of interest that must be specified for spectral matching, with a shorter

range being preferable to a longer one. Furthermore, the need to efficiently match the

target spectrum over the longer period range, which is of primary interest in many

structural engineering problems, led Beyer and Bommer [2007] to modify Eq. (4.1) by

proposing a scale factor, a for each record that minimizes the root-mean-square

difference Drms between the scaled geometric mean spectrum of the real record and the

target spectrum.

( ) ( )
√ ∑ ( ) (4.2)
( )

In the equation 4.2 ( ) represents the pseudo-acceleration ordinate of the real

spectrum j at period Ti, ( ) represents the value of the spectral ordinate of the

code spectrum at the same period, and N is the number of values used within a pre-

defined range of periods. Another procedure for selection of earthquake records is the

selection of real time histories whose spectral ordinates match to that of target spectrum

for the period range considered in a way that scaling is not required (Idriss, 1993).

Anderson and Naeim (1993) selected 120 records from a database with their plots of

constant strength inelastic response spectra; these ground motions are suitable for

engineering use and recommended for analysis purposes.

54
4.2.3.1 Evaluation of a/v ratio :

Based on seismotectonic features of a particular region, the ratio of the maximum

acceleration to that of the maximum velocity has been noted to be a complimentary

measure of the selection process. Tso et.al. [1992] and Sawada et.al. [1992] concluded

that this parameter is related to the earthquake magnitude, distance from source, and the

frequency contents of the accelerograms. They grouped the accelerograms based on the

ratio of the peak acceleration (in g) to the peak velocity (in m/s) ratio (a/v ratio) into Low

(a/v ≤ 0.8), Intermediate (0.8 < a/v ≤ 1.20) and High (a/v > 1.20).

4.2.3.2 Effect of duration of ground motion shaking

Strong motion duration is considered to be an additional and complimentary parameter in

the selection of ground motion records. The duration of ground shaking mainly depends

on the duration of rupture zone and the magnitude of the earthquake. Hannock and

Bommer [2006] pointed out that a structure undergoing stiffness and/or strength

degradation due to fatigue damage and absorbed hysteretic energy are more likely to

undergo large damage due to a long duration of shaking. However ASCE Standards 04-

98 (ASCE, 2000) recommends that duration of the ground motion should be

representative of the design or scenario earthquake records.

The selection procedure and the methodology presented here allows the engineer to make

a rational decision on using an appropriate earthquake record in time domain for the non-

linear dynamic time-history analysis. The ground motion records are usually selected

considering a few parameters and various other limiting factors like time, resources and

available data.

55
4.3 Scaling of the selected GMRs

For time history analysis, a ground motion record is selected such that the record is

compatible to the design spectrum. There are a number of methods available for scaling a

ground motion record in order to obtain a record that would represent the seismicity of a

location as expressed in the design response spectrum. The commonly used methods for

scaling or deriving a design spectrum compatible ground motion records are listed below

with a brief description.

4.3.1 PGA Scaling

In this scaling technique the input accelerogram is multiplied by a scalar quantity to

match the peak ground acceleration as that of the site specific target spectrum. Fig 4.2

shows the Peak Ground acceleration of the input ground motion record and Fig 4.3 shows

the Peak ground acceleration of the design spectrum, the scale factor is given by the

following formula (Eq. 4.3).

(4.3)

56
 1

0.9 Peak Ground Ground Motion

Acceleration Record
0.8

0.7

0.6

0.5
S(g)

0.4

0.3

0.2

0.1

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5
T, s

Fig 4.2. Peak Ground Acceleration in the selected GMR

NBCC Code Spectrum, 2010

1
0.9
0.8
Peak Ground
0.7
Acceleration
0.6
S(g)

0.5
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6

T(s)

Fig.4.3 Peak Ground Acceleration from the NBCC Code Spectrum

57
4.3.2 Ordinate Scaling Method

In this scaling technique the input accelerogram is multiplied by a scalar quantity to

match the spectral ordinate at the fundamental period of vibration of the structure as that

of the target spectrum .Fig.4.4 shows the ordinate at T1 of the structure on the input

ground motion record and Fig .4.5 shows the ordinate at T1 of the structure on the design

spectrum, the scale factor is given by the formula (Eq. 4.4):

(4.4)

Ordinate Scaling Method

0.25

0.2 Ground Motion Record

Accelaration in g, S(g)

0.15

0.1
Ordinate at T1
of input GMR
0.05

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5

Period in second, T(s)

Fig 4.4. Ordinate at T1 on the input GMR

58
 NBCC Code Spectrum, 2010
1
0.9
0.8
0.7
0.6
S(g)

0.5 Ordinate at T1
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6

T(s)

Fig 4.5. Ordinate at T1 on the NBCC Code Spectrum

4.3.3 Least Square Method

This scaling technique was proposed by Somerville et al., (1997a, b), under this method

the input accelerogram is multiplied by a scalar that minimizes the weighted sum of the

errors (differences) between the accelerogram response spectrum and the target spectrum,

the weights used are 0.3, 1.0, 2.0, 4.0 at the period corresponding to the first, second,

third and fourth modes (i.e., T1, T2, T3, T4), respectively. Fig. 4.6 shows the co-ordinates

of the least square method on the input ground motion record at 0.3s, 1s, 2s and 4s

respectively and Fig .4.7 shows the the co-ordinates of the least square method on the

design spectrum at 0.3s, 1s, 2s and 4s respectively

59
 1
Least Square Scaling Method

0.9

0.8 Co-ordinates for Least Square

0.7 Scaling r4,r3, r2, r1
Accelaration in g, S(g)

0.6

0.5
r4
0.4

0.3

0.2
r3 r2
0.1 r1
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5
T4 T3 T2 T1
Periods in sec, T(s)

Fig 4.6.Least Square Scaling Ordinates on the input Ground Motion Record

1
0.9
Co-ordinates for Least Square
0.8
Scaling S4, S3, S2, S1
0.7
0.6
S(g)

0.5
0.4 S4
0.3
0.2
S3

0.1 S2
S1
0
0
T4 T3
1
T2
2 T1 3 4 5 6

T(s)
Fig 4.7 .Least Square Scaling Ordinates on the NBCC Spectrum

60
Scale factor in the Least-square scaling method is given by the following expression (Eq.

4.5):

(4.5)

4.3.4 Partial Area Method

In this scaling technique the area under the acceleration response spectrum between the

second mode period, T2 and 1.2 times the first mode period, T1 be the same as that of the

target spectrum (Naumoski et al., 2004). Fig .4.8 shows the area under the input GMR

between T2 and 1.2 T1 and Fig.4.9 shows the area under the target spectrum between T2

and 1.2 T1. The scale factor is given by the following ratio (Eq. 4.6).

(4.6)

0.45

0.4 Ground Motion

0.35
record
Accelaration in g, S(g)

0.3

0.25

0.2 Area under the input

0.15
record between
ordinates 1.2 T1 and T2
0.1

0.05

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5

Periods in sec, T(s)

Fig 4.8 .Partial Area Scaling Ordinates on the input Ground Motion Record

61
 NBCC Code Spectrum, 2010
1
0.9
0.8
0.7
0.6 Area under the NBCC
S(g)

0.5 spectrum between T2

0.4 and 1.2 T1 ordinates
0.3
0.2
0.1
0
0 1 2 3 4 5 6

T(s)

Fig.4.9 .Partial Area Scaling Ordinates on the NBCC Code Spectrum

4.3.5 PSa Scaling Method

This scaling technique requires the area under input spectrum and target spectrum to be

equal between the period range 0-2 s (Naumoski et al., 2004).

62
1.2

1
Area under the
0.8 input GMR Ground Motion
Record
0.6

0.4

0.2

0
0 0.5 1 1.5 2 2.5

Fig 4.10 .PSa Scaling Ordinates on the input Ground Motion Record

NBCC Code Spectrum, 2010

1
0.9
0.8
0.7
Area under the target spectrum
0.6
S(g)

between 0-2 s
0.5
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6

T(s)

Fig 4.11 .PSa Scaling Ordinates on the NBCC Spectrum

63
Fig .4.8 shows the area under the input GMR between 0 and 2s and Fig.4.9 shows the

area under the target spectrum between 0 and 2 s. The scale factor is given by the

following expression (Eq. 4.6)

(4.6)

4.3.6 ASCE-7 Scaling Method

This technique requires that the average value of spectral ordinates should not be smaller

than those of the target spectra for the period range 0.2T1 to 1.5T1 where T is the

fundamental vibration (i.e., same as T1) of the structure.

0.45

0.4

0.35
Ground Motion
Accelaration in g, S(g)

0.3
record
0.25

0.2 Area under the input

0.15
GMR between 0.2T1 and
1.5T1
0.1

0.05

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5

Periods in sec, T(s)

Fig 4.12 ASCE-7 Scaling Ordinates on the input Ground Motion Record

64
 NBCC Code Spectrum, 2010
1
0.9
0.8
0.7
0.6
S(g)

Area under the target

0.5 spectrum between
0.4 0.2T1 and 1.5T1
0.3
0.2
0.1
0
0 1 2 3 4 5 6

T(s)

Fig 4.13 ASCE Scaling Ordinates on the NBCC Spectrum

Fig .4.12 shows the area under the input GMR between 0.2 T1 and 1.5 T1 and Fig.4.13

shows the area under the target spectrum between 0.2 T1 and 1.5 T1. The scale factor is

given by the following expression (Eq. 4.7).

(4.7)

4.3.7 Spectrum Matching Technique

In this method spectrum matching is done by modifying the frequency contents of the

input accelerogram to match its response spectrum to the target spectrum. There are

different software programs such as SeismoMatch (Abrahamson [1992] and Hancock et

al. [2006]) or Synth (Naumoski et al. 2004) are available for matching matching the

frequency of input spectrum to that of the target spectrum and generating the

corresponding time history signal for the ground acceleration.

65
4.3.8 Spectrum-compatible artificial earthquake record

In this technique the input accelerogram which is pre matched with the site specific target

spectrum are generated through simulation (e.g. Atkinson, 2009). Hence, these records

are directly used as input accelerograms in time history analysis. Fig. 4.14 and Table 4.5

provides a brief description of artificial records used in scaling of the ground motion

records to be used in time history analysis, the scale factor is not applicable in the case of

artificial earthquake records as the records are spectrum compatible beforehand.

4.4 GMRs used in the present study

The scaling and matching techniques for the research has been carried out for 30 (thirty)

ground motion records. Among these, eight records are synthesized and compatible to the

seismic hazard spectrum for Vancouver, Canada developed by Atkinson, [Beresnev and

Atkinson, 1998; Atkinson 2009], Fig.4.14 shows the 8 synthesized records from Atkinson

of which 4 are short duration and 4 are long duration records (Bagchi 2001; Tremblay et

al. 2001) and Table 4.1 represents the characteristics of the 8 synthesized ground motions

and Figure 4.15 show their response spectra. In addition, twenty two real ground motion

records obtained from the data base of Pacific Earthquake Engineering Research Center

(PEER, 2006) have been selected by comparing the peak acceleration-peak velocity ratio

of seismic motion (a/v) to be compatible with the seismicity of Vancouver, where a/v of a

potential ground motion is expected to be close to 1 (Table 4.2). It is noted that the a/v

ratio controls the spectral shape of the seismic motion. The response spectra of the

ground motion records are shown in Figure 4.16.

66
 550 550 Long period 1
Short period 1
350 350
150 150
-50 -50
0 3 6 9 12 15 18 0 2 4 6 8 10 12 14 16 18
-250 -250
-450 -450
550 550 Long period 2
Short period 2
Acceleration, cm/sec2

350 350
150 150
-50 -50
0 3 6 9 12 15 18 0 2 4 6 8 10 12 14 16 18
-250 -250
-450 -450
550 550 Long period 3
Short period 3
350 350
150 150
-50 -50
0 3 6 9 12 15 18 0 2 4 6 8 10 12 14 16 18
-250 -250
-450 -450
550 550
Short period 4 Long period 4
350 350

150 150

-50 -50
0 3 6 9 12 15 18 0 2 4 6 8 10 12 14 16 18
-250 -250

-450 -450

(a) Time, s (b)

Figure 4.14: Time History of Atkinson’s synthesized ground motions.

(a) Short period ground motions (b) Long period ground motions.

67
 Table 4.1: Characteristics of Atkinson’s Synthesized Ground Motion.

Record LP1 LP2 LP3 LP4 SP1 SP2 SP3 SP4

Peak Acceleration
266.2 279.4 248.6 271.7 523 527 567 380
(cm/sec2)

Duration (s ) 18.24 18.24 18.24 18.24 8.55 8.55 8.55 8.55

LP = Long Period, SP = Short Period

5
Synthesized
Design
Accelaration in g, S(g)

0
0 1 2 3 4 5 6
Periods in sec, T(s)

Figure 4.15: Spectra of selected Synthesized Ground Motion Records along with the
NBCC 2005/2010 design spectrum

5
(g)

4.5 NBCC 2005

a

4
Spectral Acceleration , S

3.5
3
2.5
2
1.5
1
0.5
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6
Period of Vibration ,T a

Figure 4.16: Spectra of selected Ground Motion Records along with the NBCC
2005/2010 design spectrum

68
 Table 4.2: Summary of Real Ground Motion

Record Location and year PGA (g) Peak Velocity a/v

No. (m/sec)
1 Imperial Valley (1940) 0.348 0.334 1.04
2 Kern Country (1952) 0.179 0.177 1.01
3 Kern Country( 1952) 0.156 0.157 0.99
4 Borrego Country (1968) 0.046 0.042 1.09
5 San Fernando (1971) 0.150 0.149 1.01
6 San Fernando ( 1971) 0.211 0.211 1.00
7 San Fernando(1971) 0.165 0.166 0.99
8 San Fernando (1971) 0.180 0.205 0.88
9 San Fernando (1971) 0.199 0.167 1.19
10 Record No.S-882 Gazli USSR 0.07 0.07 1.00
11 Record No.S-634 Coalinga 0.078 0.068 1.15
12 Monte Negro-2 (1979) 0.171 0.194 0.88
13 Report Del Archivo: 0.105 0.112 0.94
SUCH850919AL.T
14 Report del Archivo: 0.123 0.105 1.17
VILE850919AT.T
15 Kobe, Japan (1995) 0.061 0.049 1.24
16 Kobe, Japan (1995) 0.694 0.758 0.92
17 Kobe, Japan (1995) 0.707 0.758 0.93
18 Kobe, Japan (1995) 0.144 0.150 0.96
19 Northridge, CA (1994) 0.469 0.571 0.82
20 Northridge, CA (1994) 0.510 0.493 1.03
21 Northridge, CA (1994) 0.088 0.072 1.22
22 Northridge, CA (1994) 0.080 0.082 0.98

69
As some of the scaling methods considered here utilize the modal periods of a structure,

the first and second modal periods for all the building frames used here are reported in

Table 4.7. The scale factors obtained using different methods are given in Tables 4.8

through 4.17. In these tables, the scale factors exceeding 5 have been identified and NA

(Not Applicable) and the corresponding ground motion records are excluded from the

analysis. A set of spectrum compatible records have also been generated by frequecncy-

domain spectrum matching procedure using the Siesmo-match software. The spectra of

these artificial records have been shown in Figure. 4.17.

Table 4.3: Modal periods of the building frames

SMRF First period (T1), s Second period (T2), s

5 storey 1.41 0.42

10 storey 2.53 0.82

15 storey 3.57 1.17

20 storey 4.79 1.58

70
 Table 4.4: Scale Factors using PGA Method

Scaling Factor (SF) = S(g)NBCC/S(g)Field

EQ Records S(g) S(g) SF
Design Spec GMR Spec
naver1 0.960 0.9060 1.06
naver2 0.960 0.5290 1.81
naver3 0.960 0.4240 2.26
naver6 0.960 0.5210 1.84
naver7 0.960 0.6110 1.57
naver8 0.960 0.7150 1.34
naver9 0.960 0.4870 1.97
naver10 0.960 0.7300 1.32
naver11 0.960 0.2070 4.64
naver12 0.960 0.2650 3.62
naver13 0.960 0.6430 1.49
naver14 0.960 0.3990 2.41
naver15 0.960 0.4230 2.27
kobejap1 0.960 0.2250 4.27
kobejap3 0.960 2.4860 0.39
kobejap4 0.960 1.0240 0.94
northdr1 0.960 2.3860 0.40
northdr2 0.960 1.2770 0.75
northdr3 0.960 1.4360 0.67
northdr4 0.960 0.2560 3.75
SF = Scale Factor; GMR = Ground Motion Record; Spec = Spectral Ordinate

71
 Table 4.5: Scale Factors using Ordinate Method (5 and 10 storey)

5 Storey Building 10 Storey Building

GMRs S(g) S(g) SF S(g) S(g) Field SF

Design GMR Design
Spec Spec
naver1 0.270 0.1822 1.48 0.1486 0.174 0.86
naver2 0.270 0.1252 2.16 0.1486 0.057 2.60
naver3 0.270 0.1218 2.22 0.1486 0.044 3.40
naver6 0.270 0.1184 2.28 0.1486 0.037 4.03
naver7 0.270 0.2652 1.02 0.1486 0.091 1.63
naver8 0.270 0.1227 2.20 0.1486 0.089 1.67
naver9 0.270 0.0793 3.41 0.1486 0.053 2.80
naver10 0.270 0.1187 2.27 0.1486 0.086 1.73
naver11 0.270 0.0400 6.74 0.1486 0.010 14.23
naver12 0.270 0.1044 2.59 0.1486 0.015 10.10
naver13 0.270 0.1370 1.97 0.1486 0.096 1.55
naver14 0.270 0.0747 3.61 0.1486 0.028 5.23
naver15 0.270 0.0730 3.70 0.1486 0.037 4.03
kobejap2 0.270 0.0897 3.01 NA NA NA
kobejap3 0.270 0.9191 0.29 0.1486 0.201 0.74
kobejap4 0.270 0.5113 0.53 0.1486 0.133 1.11
northdr1 0.270 0.5976 0.45 0.1486 0.155 0.96
northdr2 0.270 0.3954 0.68 0.1486 0.119 1.24
northdr3 0.270 0.5705 0.47 0.1486 0.171 0.87
northdr4 0.270 0.0900 3.00 NA NA NA

72
 Table 4.6: Scale Factors using Ordinate Method (15 and 20 storey)

15 Storey Building 20 Storey Building

GMRs S(g) S(g) Field SF S(g) Design S(g) SF
Design Field
naver1 0.1051 0.069604 1.51 0.088 0.029 3.03
naver2 0.1051 0.03558 2.95 0.088 0.26712 0.33
naver3 0.1051 0.031148 3.37 0.088 0.03724 2.36
naver6 0.1051 0.032284 3.26 0.088 0.02576 3.42
naver7 0.1051 0.039968 2.63 0.088 0.06704 1.31
naver8 0.1051 0.05542 1.90 0.088 0.05716 1.54
naver9 0.1051 0.068432 1.54 0.088 0.04416 1.99
naver10 0.1051 0.085308 1.23 0.088 0.04428 1.99
naver13 0.1051 0.050864 2.07 0.088 0.02552 3.45
naver15 0.1051 0.023716 4.43 0.088 0.02288 3.85
kobejap3 0.1051 0.111036 0.95 0.088 0.05204 1.69
kobejap4 0.1051 0.055864 1.88 0.088 0.02464 3.57
northdr1 0.1051 0.055012 1.91 0.088 0.02852 3.09
northdr2 0.1051 0.07016 1.50 0.088 0.0354 2.49
northdr3 0.1051 0.064876 1.62 0.088 0.047992 1.83

73
 Table 4.7: Scale Factors using Least Square Method (5 and 10 storey)

5 Storey Building 10 Storey Building

GMRs S(g) S(g) SF S(g) S(g) SF
Design GMR Design GMR
Spec Spec Spec Spec
naver1 0.70318 0.75583 0.93 0.7467 0.79673 0.94
naver2 0.70318 0.3693 1.90 0.7467 0.35705 2.09
naver3 0.70318 0.35356 1.99 0.7467 0.34086 2.19
naver6 0.70318 0.2676 2.63 0.7467 0.26213 2.85
naver7 0.70318 0.4288 1.64 0.7467 0.49056 1.52
naver8 0.70318 0.2823 2.49 0.7467 0.33742 2.21
naver9 0.70318 0.345 2.04 0.7467 0.36318 2.06
naver10 0.70318 0.3557 1.98 0.7467 0.37429 1.99
naver11 0.70318 0.15198 4.63 0.7467 0.13767 5.42
naver12 0.70318 0.17096 4.11 0.7467 0.15627 4.78
naver13 0.70318 0.43135 1.63 0.7467 0.43225 1.73
naver14 0.70318 0.27939 2.52 0.7467 0.26642 2.80
naver15 0.70318 0.22938 3.07 0.7467 0.22119 3.38
kobejap2 0.70318 0.15849 4.44 0.7467 0.17206 4.34
kobejap3 0.70318 2.21181 0.32 0.7467 2.0103 0.37
kobejap4 0.70318 0.92385 0.76 0.7467 0.9252 0.81
northdr1 0.70318 1.55988 0.45 0.7467 1.42684 0.52
northdr2 0.70318 1.0694 0.66 0.7467 0.98174 0.76
northdr3 0.70318 1.32772 0.53 0.7467 1.26662 0.59
northdr4 0.70318 0.21299 3.30 0.7467 0.20523 3.64

74
 Table 4.8: Scale Factors using Least Square Method (15 and 20 storey)

15 Storey Building 20 Storey Building

GMRs S(g) S(g) SF S(g) S(g) SF
Design GMR Design GMR
Spec Spec Spec Spec
naver1 0.75733 0.68733 1.10 0.80887 0.57974 1.40
naver2 0.75733 0.33997 2.23 0.80887 0.3048 2.65
naver3 0.75733 0.32565 2.33 0.80887 0.28554 2.83
naver6 0.75733 0.25759 2.94 0.80887 0.22759 3.55
naver7 0.75733 0.55776 1.36 0.80887 0.53254 1.52
naver8 0.75733 0.3782 2.00 0.80887 0.37404 2.16
naver9 0.75733 0.37347 2.03 0.80887 0.33529 2.41
naver10 0.75733 0.39561 1.91 0.80887 0.378 2.14
naver13 0.75733 0.38812 1.95 0.80887 0.35815 2.26
naver14 0.75733 0.21712 3.49 0.80887 0.17459 4.63
naver15 0.75733 0.19663 3.85 0.80887 0.18815 4.30
kobejap3 0.75733 1.74361 0.43 0.80887 1.33105 0.61
kobejap4 0.75733 0.85218 0.89 0.80887 0.70161 1.15
northdr1 0.75733 1.14477 0.66 0.80887 0.90938 0.89
northdr2 0.75733 0.86556 0.87 0.80887 0.67825 1.19
northdr3 0.75733 1.09215 0.69 0.80887 0.86771 0.93
northdr4 0.75733 0.18547 4.08 0.80887 0.16002 5.05

75
 Table 4.9: Scale Factor using Partial Area Method (5 and 10 storey)

Records PADS PARS SF PADS PARS SF

naver1 0.5078 0.5801 0.88 0.5387 0.5041 1.07
naver2 0.5078 0.2693 1.89 0.5387 0.2349 2.29
naver3 0.5078 0.2672 1.90 0.5387 0.2094 2.57
naver6 0.5078 0.1965 2.58 0.5387 0.1661 3.24
naver7 0.5078 0.3181 1.60 0.5387 0.3234 1.67
naver8 0.5078 0.2168 2.34 0.5387 0.2266 2.38
naver9 0.5078 0.2588 1.96 0.5387 0.2201 2.45
naver10 0.5078 0.2344 2.17 0.5387 0.2263 2.38
naver11 0.5078 0.1284 3.96 NA NA NA
naver12 0.5078 0.1284 3.95 0.5387 0.1151 4.68
naver13 0.5078 0.3095 1.64 0.5387 0.2745 1.96
naver14 0.5078 0.2158 2.35 0.5387 0.1538 3.50
naver15 0.5078 0.1636 3.10 0.5387 0.1293 4.17
kobejap1 0.5078 0.1040 4.88 NA NA NA
kobejap2 0.5078 0.1194 4.25 0.5387 0.1222 4.41
kobejap3 0.5078 1.7074 0.30 0.5387 1.3367 0.40
kobejap4 0.5078 0.7027 0.72 0.5387 0.6765 0.80
northdr1 0.5078 1.2699 0.40 0.5387 0.9380 0.57
northdr2 0.5078 0.7828 0.65 0.5387 0.6336 0.85
northdr3 0.5078 1.0297 0.49 0.5387 0.8565 0.63
northdr4 0.5078 0.1625 3.12 0.5387 0.1371 3.93
PADS – Partial area under the design spectrum, PARS – Partial area under the record
spectrum

76
 Table 4.10: Scale Factor using Partial Area Method (15 and 20 storey)

GMRs PADS PARS SF PADS PARS SF

naver1 0.4818 0.4238 1.14 0.4312 0.3750 1.15
naver2 0.4818 0.2134 2.26 0.4312 0.2006 2.15
naver3 0.4818 0.1791 2.69 0.4312 0.1772 2.43
naver6 0.4818 0.1593 3.02 0.4312 0.1423 3.03
naver7 0.4818 0.3679 1.31 0.4312 0.3556 1.21
naver8 0.4818 0.2350 2.05 0.4312 0.2528 1.71
naver9 0.4818 0.2090 2.30 0.4312 0.2304 1.87
naver10 0.4818 0.2683 1.80 0.4312 0.2729 1.58
naver12 0.4818 0.0861 5.59 0.4312 0.0496 8.69
naver13 0.4818 0.2664 1.81 0.4312 0.2381 1.81
naver14 0.4818 0.1287 3.74 0.4312 0.1115 3.87
naver15 0.4818 0.1326 3.63 0.4312 0.1338 3.22
kobejap1 0.4818 0.1094 4.40 0.4312 0.0821 5.25
kobejap2 0.4818 0.1094 4.40 0.4312 0.0821 5.25
kobejap3 0.4818 0.9604 0.50 0.4312 0.6718 0.64
kobejap4 0.4818 0.5422 0.89 0.4312 0.3749 1.15
northdr1 0.4818 0.6592 0.73 0.4312 0.4237 1.02
northdr2 0.4818 0.5151 0.94 0.4312 0.4037 1.07
northdr3 0.4818 0.6114 0.79 0.4312 0.4650 0.93
northdr4 0.4818 0.1102 4.37 0.4312 0.0843 5.11
PADS – Partial area under the design spectrum, PARS – Partial area under the record
spectrum

77
 Table 4.11: Scale Factors using PSa Scaling

GMRs S(g) S(g) SF

Design Spec GMR Spec
naver1 0.967 0.8909 1.09

naver2 0.967 0.4430 2.18

naver3 0.967 0.4212 2.30

naver6 0.967 0.3501 2.76

naver7 0.967 0.5512 1.75

naver8 0.967 0.3802 2.54

naver9 0.967 0.4190 2.31

naver10 0.967 0.4457 2.17

naver12 0.967 0.2155 4.49

naver13 0.967 0.5272 1.83

naver14 0.967 0.3135 3.09

naver15 0.967 0.2692 3.59

kobejap3 0.967 2.4648 0.39

kobejap4 0.967 1.1017 0.88

northdr1 0.967 1.7930 0.54

northdr2 0.967 1.3123 0.74

northdr3 0.967 1.5093 0.64

northdr4 0.967 0.2473 3.91

78
 Table 4.12: Scale Factors using ASCE-7 method (5 and 10 storey)

5 Storey Building 10 Storey Building

GMRs S(g) S(g) SF S(g) S(g) SF

Design GMR Design GMR
Spec Spec Spec Spec

naver1 0.70318 0.75583 0.93 0.7467 0.79673 0.94

naver2 0.70318 0.3693 1.90 0.7467 0.35705 2.09

naver3 0.70318 0.35356 1.99 0.7467 0.34086 2.19

naver6 0.70318 0.2676 2.63 0.7467 0.26213 2.85

naver7 0.70318 0.4288 1.64 0.7467 0.49056 1.52

naver8 0.70318 0.2823 2.49 0.7467 0.33742 2.21

naver9 0.70318 0.345 2.04 0.7467 0.36318 2.06

naver10 0.70318 0.3557 1.98 0.7467 0.37429 1.99

naver13 0.70318 0.43135 1.63 0.7467 0.43225 1.73

naver14 0.70318 0.27939 2.52 0.7467 0.26642 2.80

naver15 0.70318 0.22938 3.07 0.7467 0.22119 3.38

kobejap2 0.70318 0.15849 4.44 0.7467 0.17206 4.34

kobejap3 0.70318 2.21181 0.32 0.7467 2.0103 0.37

kobejap4 0.70318 0.92385 0.76 0.7467 0.9252 0.81

northdr1 0.70318 1.55988 0.45 0.7467 1.42684 0.52

northdr2 0.70318 1.0694 0.66 0.7467 0.98174 0.76

northdr3 0.70318 1.32772 0.53 0.7467 1.26662 0.59

northdr4 0.70318 0.21299 3.30 0.7467 0.20523 3.64

79
 Table 4.13: Scale Factors using ASCE-7 method (15 and 20 storey)

15 Storey Building 20 Storey Building

GMRs S(g) S(g) SF S(g) S(g) SF

Design GMR Design GMR
Spec Spec Spec Spec

naver1 0.75733 0.68733 1.10 0.80887 0.57974 1.40

naver2 0.75733 0.33997 2.23 0.80887 0.3048 2.65

naver3 0.75733 0.32565 2.33 0.80887 0.28554 2.83

naver6 0.75733 0.25759 2.94 0.80887 0.22759 3.55

naver7 0.75733 0.55776 1.36 0.80887 0.53254 1.52

naver8 0.75733 0.3782 2.00 0.80887 0.37404 2.16

naver9 0.75733 0.37347 2.03 0.80887 0.33529 2.41

naver10 0.75733 0.39561 1.91 0.80887 0.378 2.14

naver13 0.75733 0.38812 1.95 0.80887 0.35815 2.26

naver14 0.75733 0.21712 3.49 0.80887 0.17459 4.63

naver15 0.75733 0.19663 3.85 0.80887 0.18815 4.30

kobejap3 0.75733 1.74361 0.43 0.80887 1.33105 0.61

kobejap4 0.75733 0.85218 0.89 0.80887 0.70161 1.15

northdr1 0.75733 1.14477 0.66 0.80887 0.90938 0.89

northdr2 0.75733 0.86556 0.87 0.80887 0.67825 1.19

northdr3 0.75733 1.09215 0.69 0.80887 0.86771 0.93

80
 1.4

1.2
Sa, g
1

0.8

0.6 NBCC 2010

0.4

0.2

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

Period, s

Fig 4.17 – Spectra of records matched with frequency matching with the Seismomatch
software

4.5 Summary
This chapter provides a brief description of the more commonly known procedures for

selecting earthquake ground motion records to use in the nonlinear dynamic time history

analysis along with the familiar engineering parameters through which selection can be

done through classification. The chapter also describes on how to implement different

methods of scaling of GMR with explanatory sketches. The history of selecting

earthquake records and the uncertainties one may across while selecting a suite of records

for time history analysis is discussed. Further, the scale factors for each of the scaling

methods have been tabulated.

81
 Chapter 5

Static Pushover Analysis

5.1 Introduction

Static pushover analysis is a nonlinear analysis procedure in which structural deformations due

to incrementally increased lateral forces are determined. Static lateral loads are applied using

predetermined load magnitudes that represent approximately the relative inertia forces generated

at the specific site. Strength displacement, deformation demands in the structure can be

calculated at global, storey, and element levels through the pushover analysis. The pushover

analysis is implemented in recent seismic guidelines for retrofitting of existing building

structures [FEMA 273 (1997), ATC-40 (1996)]. Pushover analysis has been shown to provide a

reliable and consistent estimate of the deformation response for structures that respond to

hazards primarily in the first mode. The biggest advantage however, lies in its ability to provide

information regarding yielding sequence and capacity of a structure. The DRAIN-2DX software

has been used for analysis of plane two dimensional models of the frames. The pushover analysis

has been performed using inverted triangular load distribution patterns for all frames. The base

shears considered are the normalized base shear (Herrera et al 2003) and the base shear

coefficient is defined as the ratio of the seismic base shear (V) to the weight (W) tributary to the

frame of the building. Using the results from the pushover analysis a plot of base shear vs. roof

displacement called the pushover curve (Akbas et al. 2003) is obtained. From the pushover curve

one can estimate the capacity of the building and also trace the different stages of deformation

like beam and column yielding.

83
 First Yield of Beam First Yield of Column Point of Instability or 2.5%
Interstorey drift

(a) (b)

(c) (d)

Figure 5.1 Pushover Curves for (a) 5 Storey SMRF (b) 10 Storey SMRF (c) 15 Storey SMRF

(d) 20 Storey SMRF

84
The pushover graphs of different frames are shown in Figure 5.1. On the graph the point of first

yielding of beam and column is shown along with the point of instability of the Steel Moment

Resisting Frame.

The pushover analysis has been carried out by considering 5% strain hardening, P-∆ effect is

considered in the analysis to account for large deformations and the second order effect. The

point of instability has been marked on the pushover curve to indicate the point where the

structure fails and the slope of the pushover graph tends to shift in the downward direction. In

SMRFs that are found to be fail beyond the 2.5% interstorey drift, the point corresponding to

2.5% interstorey drift is marked on the graph as the point of failure. The capacity of the frames

is calculated from the pushover graph by estimating the yield displacement due to seismic load.

In the analysis the gravity load (D+0.5L) is applied corresponding to lateral load. The pushover

curve for the 5, 10, 15, 20 storey frames are shown in the Figure 5.1.

The normalized design base shear of 5, 10, 15 and 20 storey buildings are found to be 0.042,

0.0253, 0.0255 and 0.0264. The numbering sequence of beam and column is shown in the Figure

3.1. The first yielding in the 5 storey frame starts from the beam no. 6 at a normalized base shear

of 0.159, the first yielding of the column starts at the normalized base shear of 0.273 in column

no.6 for bare frame. Plastic hinge formation in the 10 storey building occurs first at beam no.12

for normalized base shear of 0.0886 similarly the first yielding in a column occurs at normalized

base shear of 0.154. In the 15 storey building the first yielding of beam occurs at normalized

base shear 0.064 at beam no.19. and the column yielding occurs at a normalized base shear of

0.10 at column no.31. In the 20 storey frame building the first beam yielding occurs at the

normalized base shear of 0.048 in beam no.23 and at 0.066 in column no.41. The summary of

base shear and the displacements at the first beam and column yielding are given in Table 5.1.

85
The available capacity and the deformation demand are also calculated in the pushover analysis.

The base shear, lateral roof displacement and the interstorey drifts at the point of instability are

also determined as percentage of total height of building from the pushover analysis and

presented. In the pushover analysis it has also been observed that the formation of plastic hinges

occur in the beams first, and in the columns they occur at the lower storey, which satisfies the

requirements of the capacity based seismic design.

Table 5.1 Base Shear Coefficient at the first beam and first column yielding

First Beam Yielding Fist Column Yielding

SMRF Level Base Shear Beam Index Base Shear Column Index
Coefficient Coefficient

5 Storey 0.159 No. 6 0.273 No. 6

10 Storey 0.0886 No.12 0.154 No. 11

15 Storey 0.064 No.19 0.074 No. 31

20 Storey 0.048 No.23 0.066 No. 41

The pushover curves presented in Figure 5.1 are idealized as bilinear curves to obtain the yield

points of the building frames. The corresponding yield displacements and normalized base shear

are given in Table 5.2. Table 5.3 shows the values of the base shear coefficient and

corresponding roof displacement at failure.

86
 Table 5.2 Design base shear and yield points of the SMRFs.

Design base shear, Normalized base

Yield displacement,
SMRF KN shear at yield
Dy, in %H
Fy/W
5 storey 178.8 0.22 0.75
10 storey 306.2 0.12 0.90
15 storey 366.4 0.09 1.00
20 storey 401.0 0.063 0.90
H=Height of the building.

Table 5.3 Displacement at failure (i.e., point of instability or 2.5% drift) of the frames.

SMRF Normalized Base Shear, Roof Displacement, Interstorey Drift,

Fu/W Du in %H du in %h
5 Storey 0.275 1.495 2.50

10 Storey 0.145 1.690 2.50

15 Storey 0.105 2.095 2.42

20 Storey 0.065 1.750 2.35

H = height of the building, h= storey height

The roof displacement from the pushover analysis has also been observed for the parameter-

Maximum (Mean (M) + Standard deviation (SD)) interstorey drift obtained from the time history

analysis presented in the next chapter. Table 5.4 shows the values of roof displacement are used

to compare the dynamic analysis and the static analysis.

87
 Table 5.4: Roof displacement (% H) at Maximum M+SD of interstorey drift.

SMRF Roof displacement (% H) Roof displacement (% H)

(Dynamic) (Pushover, at failure)
5 Storey 1.183 1.495
10 Storey 1.220 1.690
15 Storey 1.144 2.095
20 Storey 1.483 1.750
M=Mean value, SD=Standard deviation.

5.2 Performance estimation based on the Pushover Analysis

5.2.1 Capacity Demand Diagram Method

The concept of capacity spectrum method proposed by Freeman et al. (1975) was later

developed and proposed by Chopra and Goel (1999) called as Capacity Demand Diagram

method. This method involves transforming a multiple degrees of freedom (MDOF) system

into an equivalent single degree of freedom (SDOF) system using a suitable factor which is

determined based on the deformed modal shape, Φ. The first modal deformation shape has been

used here as the deformed shape of the structure to develop the capacity diagram. The masses

and the modal displacements from the first mode, Φ are tabulated for 20, 15, 10, and 5 storey

frames as follows (Tables 5.5 through 5.8, respectively).

88
 Table 5.5 Base shear distribution in 20 storey SMRF

Storey Height Weight hiWi F (KN) Mass (m) Phi Φ

hi(m) (Wi, KN) (KNm)

20 74.20 590.87 43842.55 113.59 60.23 1.00

19 70.55 758.00 53476.86 28.20 77.27 0.99

18 66.90 758.00 50710.17 26.74 77.27 0.97

17 63.25 758.00 47943.47 25.28 77.27 0.95

16 59.60 758.00 45176.77 23.82 77.27 0.92

15 55.95 758.82 42456.15 22.39 77.35 0.89

14 52.30 759.65 39729.52 20.95 77.44 0.85

13 48.65 759.65 36956.81 19.49 77.44 0.81

12 45.00 759.65 34184.10 18.02 77.44 0.76

11 41.35 759.65 31411.39 16.56 77.44 0.71

10 37.70 762.37 28741.30 15.15 77.71 0.66

9 34.05 765.09 26051.34 13.74 77.99 0.60

8 30.40 765.09 23258.76 12.26 77.99 0.53

7 26.75 765.09 20466.18 10.79 77.99 0.47

6 23.10 765.09 17673.60 9.32 77.99 0.40

5 19.45 767.13 14920.71 7.87 78.20 0.33

4 15.80 769.17 12152.93 6.41 78.41 0.26

3 12.15 769.17 9345.45 4.93 78.41 0.19

2 8.50 769.17 6537.97 3.45 78.41 0.12

1 4.85 786.73 3815.62 2.01 80.20 0.06

Sum 15104.39 588851.63 400.96 1539.69

89
 Table 5.6 Base shear distribution in 15 storey SMRF

Storey Height Weight hiWi F (KN) Mass Phi Φ

hi(m) (Wi, (KNm) (m)
KN)

15 55.95 742.96 41568.74 83.20 75.73 1

14 52.3 759.51 39722.11 36.89 77.42 0.934763

13 48.65 759.51 36949.92 34.31 77.42 0.869526

12 45 759.51 34177.72 31.74 77.42 0.80429

11 41.35 759.51 31405.53 29.16 77.42 0.739053

10 37.7 761.40 28704.85 26.65 77.61 0.673816

9 34.05 763.30 25990.33 24.13 77.81 0.608579

8 30.4 763.30 23204.29 21.55 77.81 0.543342

7 26.75 763.30 20418.24 18.96 77.81 0.478105

6 23.1 763. 30 17632.20 16.37 77.81 0.412869

5 19.45 766.23 14903.29 13.84 78.11 0.347632

4 15.8 769.17 12152.93 11.28 78.41 0.282395

3 12.15 769.17 9345.45 8.68 78.41 0.217158

2 8.5 769.17 6537.97 6.07 78.41 0.151921

1 4.85 784.81 3806.32 3.53 80.00 0.086685

Sum 11454.14 346519.90 366.38 1167.60

90
 Table 5.7 Base shear distribution in 10 storey SMRF

Storey Height Weight hiWi F (KN) Mass Phi Φ

hi(m) (Wi, (KNm) (m)
KN)

10 37.7 727.54 27428.19 75.73 74.16 1

9 34.05 747.03 25436.33 44.53 76.15 0.903183

8 30.4 747.03 22709.67 39.75 76.15 0.806366

7 26.75 747.03 19983.02 34.98 76.15 0.709549

6 23.1 747.03 17256.36 30.21 76.15 0.612732

5 19.45 755.53 14695.01 25.72 77.02 0.515915

4 15.8 764.03 12071.61 21.13 77.88 0.419098

3 12.15 764.03 9282.92 16.25 77.88 0.322281

2 8.5 764.03 6494.22 11.37 77.88 0.225464

1 4.85 764.71 3708.83 6.49 77.95 0.128647

Sum 7527.97 159066.20 306.16 767.38

Table 5.8 Base shear distribution in 5 storey SMRF

Storey Height Weight hiWi F (KN) Mass (m) Phi Φ

hi(m) (Wi, KN) (KNm)

5 19.45 583.95 11357.81 55.28 59.53 1

4 15.8 747.22 11806.05 47.21 76.17 0.812339

3 12.15 747.22 9078.71 36.31 76.17 0.624679

2 8.5 747.22 6351.36 25.40 76.17 0.437018

1 4.85 754.66 3660.08 14.64 76.93 0.249357

Total 3580.26 42254.01 178.83 364.96

91
To transform MDOF to a SDOF system, the transformation factor Γ is applied and given by the

below formulae (Chopra and Goel, 1999)

m* =∑miΦi …………………………………………………………5.1

m*
Γ = ……………………………………..…………………… 5.2
 mi  i2
where mi is the mass of ith storey, m* is the mass of the equivalent single-degree-of –freedom

system, Φi is the assumed displacement at the ith storey and Γ is the transformation factor

applied to transform the multi-degree-of-freedom system to the corresponding SDOF system.

The transformation factor and the mass m* for the different frames are tabulated below

Table 5.9 - Values of Γ and m* of SMRF

SMRF Γ m*

20 Storey 1.306 950.0

15 Storey 1.454 631.3

10 Storey 1.432 430.1

5 Storey 1.394 221.5

The pushover analysis has been done according to code by applying an inverted triangular force

distribution in which the total equivalent earthquake force has been considered. The hazard

spectra considered if taken from NBCC 2010 and has been used as design spectra to determine

demand diagram The pushover curve from the MDOF system is idealized in a bilinear form,

which is then converted to the capacity curve of the equivalent SDOF system by dividing the

roof displacement at yield by the transformation factor, Γ, and the yield base shear by the

equivalent mass, m*. Based on the data provided in Tables 5.2, 5.3 and 5.9, the yield and

ultimate base shear and displacement values of the equivalent SDOF system for each building

are computed and presented in A-D (Acceleration-Displacement) format in Table 5.10.

92
 Table 5.10 - Yield and ultimate base shear and displacement values of the equivalent SDOF
systems in A-D format using the CDD method

SMRF Say*, g Dy*, mm Sau*, g Du*, mm Ductility (µ)

20 Storey 0.102 499 0.105 970 1.94

15 Storey 0.166 498 0.194 1046 2.10

10 Storey 0.214 455 0.259 854 1.88

5 Storey 0.363 390 0.453 779 2.00

The hazard spectra conceived in NBCC (2010) has been further used to determine the demand

curve in A-D format, given by the following equation

2
  Tn 
D   A (Fajfar 2000) ………………………… 5.6
R y  2 

Where D is the roof displacement, µ is the ductility, Ry ductility reduction factor, Tn is the period

of vibration and A is the spectral acceleration. For Krawinkler and Nasser proposed Ry-µ-Tn

relation the following Equations are used.

R y  C   1  1
1/ C
(Chopra and Goel 1999) …………………………… 5.7

Ry ductility reduction factor, µ is the ductility capacity and C is a constant as described in the

Equation 5.8.

Tna b
C Tn ,     (Chopra and Goel 1999) ………………… 5.8
1  Tn Tn
a

where b and α are constant of the material property, Tn is the period of vibration in sec.

A number of R-µ-T relations are readily available in the literature given by, Newmark and Hall

(1982); Krawinkler and Nassar (1992); Vidic et al. (1994); and Miranda and Bertero (1994).

Figure 10 shows the demand spectra constructed using different R-µ-T for the four different

building frames and it is noted to produce similar demand spectra.

93
 0.20
KN (mu=5)
0.18 VFF (mu=5)
NH (mu=5)
0.15 MB (mu=5)
All (mu=1)
Acceleration (A), g

0.13

0.10

0.08

0.05

0.03

0.00
0 100 200 300 400 500 600 700 800

Displacement (D), mm

Figure 5.2. Seismic demand curves using different R-µ-T relations (KN Krawinkler and Naser
(1992);VFF Vidic et al. (1994); and NH Newmark and Hall (1982))

For the application of the CDD method, the Capacity Curve (CC) of the equivalent SDOF system

of a building frame is plotted using the data given Table 5.10 and overlaid with the Demand

Curves (DC) generated using Equation 5.8 or other methods as shown in Figure 5.2. The

Capacity-Demand diagrams developed for different building frames are shown in Figure 5.3

through 5.6.

Figure 5.3 shows the capacity-demand diagram of the SDOF system diagram for the twenty-

storey building where the capacity curve CC1 is overlaid with the demand curve for different

values of ductility µ (mu). From the intersection of capacity diagram and the elastic demand

diagram as indicated by Point 1 in Figure 5.3, the displacement demand to the SDOF system is

found to be 450 mm. For the corresponding MDOF system, the displacement demand is

estimated to be (450*1·306) or 588 mm (0·79% of the building height). Assuming an average

94
ratio of storey drift to roof drift to be 1·6 as suggested by Gupta and Krawinkler (2000), the

maximum interstorey drift demand can be estimated to be 1·27%, which is somewhat lower than

the mean interstorey drift obtained from the dynamic time history analysis (about 1·6 to 1·7%

depending on the method of scaling). Considering the Vision 2000 Committee prescribed level

of performance (SEAOC, 1995), the CDD method indicates the achievement of the LS (Life

Safety) performance of the structure, which is consistent with the dynamic analysis presented

earlier.

0.20
DC, mu = 5
0.18 DC, mu = 1.94
DC, mu = 1
0.15 CC1
CC2
Acceleration, A g

0.13

0.10
1 2
0.08
3
0.05

0.03

0.00
0 200 400 600 800 1000
Displacement, D mm

Figure 5.3. Application of the CDD, N2 and DBSD methods (20 storey building)

For the 15, 10 and 5 storey buildings the CDD plots are shown in Figures 5.4, 5.5 and 5.6,

respectively. From the intersection of the capacity curve CC1 and the elastic demand curve (DC-

mu=1) as indicated by Point 1 in each of the above figures, the SDOF demand and the

corresponding MDOF demand for each building have been obtained the same ways as described

above in the context of the twenty storey building. The results have been summarized in Table

95
5.11 which indicates that the estimated interstorey drift demand in each building is within 2.5%

as required by NBCC 2010.

0.20
DC, mu = 5

0.18 DC, mu =
2.1
0.15 DC, mu = 1
Acceleration A, g

0.13

0.10 1

3
0.08 2

0.05

0.03

0.00
0 200 400 600 800 1000
Displacement D, mm

Figure 5.4 Application of the CDD, N2 and DBSD methods (15 storey building)

0.30
0.28 DC, mu = 5
DC, mu = 1.88
0.25
DC, mu = 1
0.23
CC1
0.20 CC2
Acceleration A, g

0.18
0.15 1
0.13
0.10 2
0.08 3

0.05
0.03
0.00
0 200 400 600 800 1000
Displacement D, mm

Figure 5.5 Application of the CDD, N2 and DBSD methods (10 storey building)

96
 0.40
DC, mu = 5
0.35 DC, mu = 2
DC, mu = 1
0.30 CC1
CC2
Acceleration A, g

0.25

0.20
1
0.15 2
3
0.10

0.05

0.00
0 200 400 600 800 1000
Displacement, D mm

Figure 5.6. Application of the CDD, N2 and DBSD methods (5 storey building)

Table 5.11 - The SDOF and MDOF demand using the CDD method

SMRF SDOF MDOF Roof Drift, Estimated

Displacement, Displacement, %H Interstorey
mm mm Drift, %h

20 Storey 450 588 0.79 1.27

15 Storey 328 477 0.85 1.36

10 Storey 280 401 1.06 1.70

5 Storey 178 248 1.28 2.04

97
5.2.2 N2 Method

The N2 method was proposed by Fajfar (2000) which is known to be an extension of the

capacity-demand diagram method. The capacity diagram is constructed the same way as

described in the earlier section, however, in this method the yield base shear is divided by the

factor m*Γ, instead of m*. The elastic period, T* of the idealized bilinear system is given by the

equation (5) (Fajfar, 2000).

m * D *y
T  2
*
(Fajfar, 2000) ..………………………………………….
Fy

Based on the above description, the yield and ultimate shear, and the elastic period of the

equivalent SDOF structures have calculated and shown in Table 5.12.

Table 5.12 - Yield and ultimate base shear in acceleration format and the elastic period of the
equivalent SDOF using the N2 method

SMRF Say*, g Sau*, g T*, s

20 Storey 0.078 0.081 4.51

15 Storey 0.114 0.134 3.47

10 Storey 0.150 0.181 2.92

5 Storey 0.260 0.325 2.08

The capacity curve for the equivalent SDOF system using the N2 method has been shown by

CC2 in Figure 5.3. The acceleration at the yield point (Say) is determined to be 0·078 g. As the

capacity curve CC2 does not intersect the elastic demand curve, the intersection of the line

following the slope of the capacity curve with the demand curve as indicated by Point 2 in Figure

5.3, is considered for estimating the elastic acceleration (Sae), which is found to be 0·088 g. The

98
corresponding elastic displacement (Sde) is found to be 590 mm. The ductility reduction factor

for the structure is given by (Rµ = Sae /Say) which equals to 1·14. The elastic period of system (T*

= 4·51 s as shown in Table 5.11) is greater than the critical period, Tc = 0·2 s which further

supports the ductility value of Rµ = 1·14. In the case of T* > Tc, the elastic displacement and

the inelastic displacement is considered equal, i.e. Sa = Sae (Fajfar, 2000). Therefore, the inelastic

displacement demand of SDOF system is 590 mm. Hence, the inelastic displacement demand of

the MDOF system is (590*1·306) or 771 mm (1·04% of building height), and the ultimate

displacement of the building is about 874 mm (1·18% of building height). The ductility demand

of the structure determined here is less than the ductility capacity of 1·94 as indicated in Table

5.10. Therefore, the system has adequate ductility and the design is said to be satisfactory.

For the 15, 10 and 5 storey buildings the N2 Capacity curves are shown and CC2 in Figures 5.4,

5.5 and 5.6, respectively. From the intersection of the capacity curve CC2 and the elastic demand

curve (DC-mu=1) as indicated by Point 2, the elastic acceleration and the corresponding

displacements are determined. In each of the above figures (Figs. 5.4-5.6), the capacity curve

CC2 intersects the elastic demand curve before the yield point is reached. Therefore, the ductility

reduction factor Rµ in these cases is considered to be 1. Also, the elastic period in each case is

larger than the critical period of 0.2 s as in the case of the twenty storey building. In that case, the

inelastic demand is assumed to be the same as the elastic demand as recommended in Fajfar

(2000). A summary of the SDOF and MDOF demands is given in Table 5.13. The ductility

demand of the structures determined here is less than their ductility capacities as indicated in

Table 5.10. Therefore, the systems are deemed to have adequate ductility and the design is

considered satisfactory.

99
 Table 5.13 - The SDOF and MDOF demand using the N2 method

SMRF SDOF MDOF Roof Drift, Rµ

Displacement, Displacement, %H
mm mm

20 Storey 590 771 1.04 1.14

15 Storey 380 553 0.99 1.0

10 Storey 329 471 1.25 1.0

5 Storey 223 311 1.60 1.0

5.2.3 DBSD Method

This method is proposed by Humar and Ghorbanie-Asl (2005). The preliminary target roof

displacement is checked against the following limits

a. Maximum roof displacement permissible as per code provisions

b. Roof displacement magnitude at the P-∆ instability limit in pushover analysis.

c. Roof displacement at which the element’s ductility demand exceeds its ductility

strength.

The design of the structure is done normally by computing the base shear of the multi-

degree-of-freedom (MDOF) system and is then transformed to SDOF by using a modification

factor.

The steps followed in this design method are as follows:

100
 Step-I. Compute the ductility strength (µ) from the initial yield displacement and the

ultimate displacement. If the ductility capacity recommended by the code is lower than

the calculated one than the code permitted, design is OK.

 Step-II. Compute the ultimate displacement of SDOF (δu).

 Step-III. Construct the inelastic spectrum for ductility of µ along with the acceleration-

displacement (A-D) of the spectrum using Equations 5.6 to 5.8.

 Step-IV. Determine the inelastic acceleration from the A-D spectrum for equivalent

ultimate displacement (δu) and compute the design base shear from the inelastic

acceleration obtained from the A-D spectrum. The design base shear is obtained using

Equation 5.9 (Humar and Ghorbaine-Asl., 2005).

Ay m*
V …………………… 5.9
R0

Wherein m* is the mass of equivalent SDOF system given by dividing the mass of

multi-degree-of-freedom system by a modification factor Γ as given by Equation 5.2, R0

is the over strength related force reduction factor taken as 1·5 for the steel moment

resisting frame according to NBCC (2010); and Ay is the spectral acceleration.

 Step-V. Design the structure for the base shear calculated in Step-IV using the procedure

provided in the code (i.e., NBCC 2010).

 Step-VI. Perform the Static Pushover analysis on the designed structure to obtain a

refined value of the yield and ultimate displacements. Repeat Steps I to V until the design

base shear is acceptable.

101
The preliminary design of the building frames has been initially carried using the NBCC 2010

code based procedure, and the pushover curves are plotted for the buildings (Figure 5.1). The

design base shear for each building computed as per code procedures is given in Table 5.2 .The

yield and ultimate displacements provided in Table 5.2 and 5.3, respectively. The ductility

capacity of each building as given in Table 5.10 are found to be lower than the code permitted

ductility for a fully ductile moment resisting frame which is 5. Hence the computed ductility

capacity as given in Table 5.10 is used in the design. The transformation factor Γ for each

building is provided in Table 5.9 which is utilized for calculating the yield (dy) and ultimate

displacements (du) of the SDOF systems as listed in Table 5.10. The inelastic demand curve for

the obtained for the corresponding ductility for each building as shown in Figures 5.3 through

5.6, in which the intersection of the inelastic demand curve and the capacity curve CC1 at Point

3 indicates the inelastic demand acceleration (A) and corresponding displacement demand. The

revised design base shear is obtained using Equation 5.9 and listed in Table 5.14.

Table 5.14 - Estimation of the design base shear using the DBSD method

SMRF Ay, g m*, t Vy, KN Vd, KN

20 Storey 0.060 950.0 372.8 401.0

15 Storey 0.073 631.3 301.4 366.4

10 Storey 0.087 430.1 244.7 306.2

5 Storey 0.127 221.5 183.9 178.8

It is found from Table 5.14 that the base shear estimated using the DBSD method is lower than

that used in the equivalent static load-based design and hence the code-based static design

procedure is satisfactory, yet conservative, in all cases except the five storey building. In the case

102
of the five storey building, Vy as determined from the DBSD method is less than 3% higher than

the original design base shear Vd, and thus the original design is still satisfactory.

5.2.4 Yield Point Spectra Method (Aschheim, 2004)

Yield point Spectra (YPS) method is a direct and simplified method of Capacity Spectrum

method, it is found to be useful when the performance objectives are specified in terms of

ductility and peak displacement of structure, in such a scenario this method can be easily used to

predict the performance of the structure. In this method a graphical procedure is followed to

evaluate the performance of the structure which is a plot of base shear verses yield

displacement. The yield point is constituted by the yield strength (Fy) and yield displacement

(Dy). The following Equations (Eq. 5.10 and 5.11) are used to determine the yield point spectra.

2
 T 
Dy     S a (Aschheim 2004) ………………………………5.10
 2 

S ae
Sa 
R
(Fajfar 2000) …………………………………………5.11

Wherein Sa is the inelastic spectral acceleration, Sae is the elastic spectral acceleration, µ is the

ductility capacity of the structure, Rµ is the ductility reduction coefficient, T is the period of

vibration and Dy is the yield displacement. In the above equations the R-µ-T relationship

proposed by Miranda and Bertero (1994) has been used. Equation 5.12 is applied to compute the

value of Rµ.

  16T 
R    1    exp  
   ……….5.12

103
In the above equation µ is the ductility capacity of the structure, Rµ is the ductility reduction

coefficient, T is the period of vibration.

YPS method can also be used to obtain the vibration properties for the target displacement and

the ductility demands whereas other PBSD methods can only be used to determine the

displacement properties from the given vibration data.

0.40
Ductility 1
Ductility 2
0.35
Ductility 4
Ductulity 8
0.30
Base Shear Coefficient, Cy

CP demand
Design region LS demand
0.25

0.20
Life safety
P Collpase
(LS) Demand
0.15 prevention (CP)

0.10
A
0.05
D C B
0.00
0 200 400 600 800 1000 1200
Yield Displacement, mm

Fig 5.7 Yield point spectrum of NBCC 2010 response spectrum for Vancouver
(20 storey building)

Figure 5.7 shows the application of the YPS procedure to the 20-storey building frame. The

seismic hazard spectra for Vancouver as provided in NBCC (2010) are applied to obtain the

yield point spectra for different values of ductility in (Figure 5.7). In order to predict the

performance of the structure using the yield point spectra, the following two performance

objectives are considered: (a) Collapse Prevention as per Vision 2000 Committee (SEAOC,

1995) recommendation and NBCC (2010) requirement of 2·5% interstorey drift and (b) Life
104
Safety objective based on Vision 2000 Committee’s (SEAOC, 1995) recommendation of 1·5%

interstorey drift. Correspondingly, the CP level of performance indicated by the performance

demand curve ABCD in Figure 5.7 has been plotted. Gupta and Krawinkler (2000) stated that

the storey drift to roof drift ratio for low to medium rise building varies between 1·2- 2·0.

Considering the average value of 1·60, the roof displacement corresponding to an interstorey

drift of 2·5% is computed to be 1159 mm, Under the first modal displacement vector the

transformation factor for SDOF is 1·306. Hence, the maximum roof displacement of the SDOF

structure is worked out as 1159/1·306 or 887 mm. therefore Point A in the graph corresponds

to an elastic yield displacement of 887 mm, point B on the YPS represents a ductility of value 2

and yield displacement of 887/2 or 444 mm, point C on the YPS represents a ductility 4 and an

yield displacement of 887/4 or 222 mm, and point D corresponds to a yield displacement of

887/8 111 mm with ductility factor of 8. Similarly, the LS performance demand curve is

constructed in the similar fashion (Figure 5.7) to represent a interstorey drift of 1·5% which

corresponds to a roof displacement of 535 mm. The base shear coefficient and roof displacement

at yield for the equivalent SDOF systems are found to be 0·102 and 499 mm, respectively, as

represented by the point P in Figure 5.7. As point P lies above both LS and CP levels of

performance demand curve, the design satisfies both of these levels of performance objectives.

105
 0.40
Ductility 1
Ductility 2
0.35
Ductility 4
Ductulity 8
Base Shear Coefficient, Cy 0.30
CP demand
Design region LS demand
0.25 P
Life safety
0.20 (LS)
Collpase
Demand prevention (CP)
0.15

0.10

0.05 A
D C B
0.00
0 200 400 600 800 1000 1200
Yield Displacement, mm

Fig 5.8 Yield point spectrum of NBCC 2010 response spectrum for Vancouver (15 storey
building)
Figure 5.8 shows the plot of YPS for the 15 storey building in which the Collapse Prevention

curve and the Life safety demand curves have been further plotted for the corresponding

displacement values of 874 mm and 525 mm, respectively. As before, Point P indicating the

yield capacity lies above both LS and CP levels of performance demand curve.

106
 0.40
Ductility 1
Ductility 2
0.35
Ductility 4
Design region Ductulity 8
Base Shear Coefficient, Cy 0.30 P CP demand
LS demand
0.25
Life safety
0.20 (LS)
Demand Collpase
prevention (CP)
0.15

0.10
A
0.05 D C B
0.00
0 200 400 600 800 1000 1200
Yield Displacement, mm

Fig 5.9 Yield point spectrum of NBCC 2010 response spectrum for Vancouver (10 storey
building)
Figure 5.9 shows the plot of YPS for the 10 storey building in which the collapse prevention

curve and the Life safety demand curve have been plotted for the corresponding displacement

values of 589 mm and 353 mm, respectively. In this case, Point P indicating the yield capacity

lies well above both LS and CP levels of performance demand curve.

107
 0.40
Ductility 1
Ductility 2
0.35
P Ductility 4
Design region Ductulity 8
Base Shear Coefficient, Cy 0.30
CP demand
Life safety
LS demand
0.25 (LS) Demand

D
0.20
Collpase
prevention (CP)
0.15
C
0.10 B A
0.05

0.00
0 200 400 600 800 1000 1200
Yield Displacement, mm

Fig 5.10 Yield point spectrum of NBCC 2005 response spectrum for Vancouver (5 storey
building)

Figure 5.10 shows the plot of YPS for the 5 storey building in which the yield point spectra for

the collapse prevention curve and the Life safety demand curves have been plotted for the

corresponding displacement values of 304 mm and 182 mm, respectively. Here the yield

capacity point (Point P) lies high above the CP and LS demand curves indicating a conservative

design.

5.3 Summary

The static pushover analysis performed and the corresponding capacity curves have been

obtained and interpreted with commonly known performance-based design methods. It is noted

that all the methods are considered here confirm that the existing design based on the code

procedure is adequate and many a times it is conservative. The pushover curves are also

compared to the results obtained from the Time history analysis to determine the performance

108
achievements of the buildings. The pushover curves indicate that buildings fail to achieve the

desired ductility in case of a real life scenario. Among different PBSD methods considered here

to test the adequacy of the code-based design of the buildings for achieving different levels of

performance, only the BDSD method is found to provide a direct estimation of modified forces

to be used the refinement of the design. The other methods are helpful in comparing the

displacement demand to capacity for a given level of performance objective. While there

differences in the ways how the existing PBSD methods such as those considered here work,

there is one common aspect in them, that is, they all require an MDOF structure to be converted

to an equivalent SDOF structure by using a transformation technique. Also, they produce

consistent results for buildings considered here.

109
 Chapter 6

Non-linear Time-history Analysis

6.1 History

The basic performance objective of a designed structure under structural engineering

practice was widely accepted as to provide adequate life-safety while keeping the

construction cost economical. However, structures designed for critical facilities, like

nuclear plants, manufacturing units and major business operations for which an

interruption in operation or damage to the facility might result in severe socio-economic

problems impacting the society at large. Hence, the present practice demands that

engineered structures conform to a predictable performance in accordance to pre-defined

performance objectives. It is also necessary to perform a cost-benefit-risk assessment

considering the seismic hazards, resulting in controlling the earthquake-related

investment for the structure during the design life span.

Traditionally, the seismic design codes recommends the design of a structure based on

limits on stresses and member forces calculated from prescribed levels of applied lateral

shear forces. Further, the performance objectives are in these codes are not clearly stated

except to provide for life safety (strength and ductility) and damage control

(serviceability drift limits). However, recent earthquakes such as the 1994 Northridge

and 1995 Hanshin– Awaji (Kobe) earthquakes were noted to cause high level of

109
damage to structures, resulting in economic loss due to loss of use, public services line

disruption, and cost of structural repair works were unexpectedly high.

It is noted that in the current code design procedures, there are uncertainties involved

concerning the seismic demand and seismic capacity of the structure. Structural failures

observed after the 1994 Northridge and 1995 Kobe earthquakes have exposed the

weakness of the prevalent design and construction procedures notably in steel moment

frames and pressed the need for new approaches for evaluation of building performance

and design.

In the Performance-Based Seismic Design (PBSD) philosophy the design criteria are

expressed in terms of achieving the stated performance levels corresponding to the stated

levels of seismic hazard. In the design process, attempts must be made to address the core

issue of the large uncertainty inherent in defining seismic demands and improving

performance in the evaluation and design process.

6.2 Non-linear Time History Analysis

Non-linear Time History Analysis, often referred to as Non-linear Dynamic analysis is

considered to be the most suitable method to estimate deformation and forces in an

structure subjected to seismic ground excitation (Ghobarah, 2001). For the analysis to be

reliable and credible, it is necessary to ensure that a) appropriate site-specific ground

motion with specified hazard level are selected with confidence; b) the structural model is

representative and realistic; c) analysis procedures and interpretation tools are reliable;

110
and d) identification of modes and sequence of element and component failure are

predictable .

The main objective of dynamic analysis is to estimate the roof displacement and the

interstorey drift of the building subjected to the seismic hazard. Dynamic analysis

provided information on flexural yielding, and change of the pattern in strength and

stiffness distribution of the structure. The maximum ductility and the deflection in the

individual member can be determined through the results of a dynamic analysis. It also

allows for considering the P-delta effect in the structure caused due to gravity loading on

the lateral displacement. The DRAIN -2DX software has been employed to carry out the

dynamic analysis for the considered 30 earthquake records. The discussion and analysis

of the results in the following section are classified broadly into three categories a) Scaled

GMR records b) Spectrum compatible GMR records c) Atkinson’s artificial earthquake

records .

6.3 Discussion of Interstorey drift results

6.3.1 PGA Scaling method
The interstorey drift graphs of the 5, 10, 15 and 20 storey frames have been plotted from

the results of the Non-linear time history analysis for the selected GMRs scaled using the

PGA scaling method.

111
 Table 6.1 : PGA Scaling method - Summary of Interstorey drift for Real ground motion

SMRF Interstorey Drifts (%hs) of SMRF

GMRs scaled by Peak Ground Acceleration method
Max. of Mean Max. of (Mean+SD)
+ - + -
5 Storey 1.244185 -0.64124 1.607446 -0.81725
10 Storey 0.881252 -0.49075 1.176595 -0.71509
15 Storey 0.673171 -0.33707 0.94335 -0.46054
20 Storey 0.673202 -0.33178 0.832117 -0.42891
SD = Standard Deviation, hs = Storey height.

Table 6.1 gives the values of mean, Sum of mean and the Standard deviation values of

the interstorey drifts from the PGA Scaling method .

The Ground motion record NAVER2 is found to have the drift close to the positive mean

value in the 5 storey frame, in the 10 storey frame NORTHRIDGE1 is found to have the

drift close to the positive mean value, in the 15 storey frame the drift close to the positive

mean value is obtained by scaling the record NAVER15, finally in the 20 storey frame

NORTHRIDGE 4 ground motion record produces the modest drift in the building. Figure

6.1 shows the Interstorey drift graphs from the time history analysis of GMRs scaled

using PGA method for the 5 storey SMRF and the 10 storey SMRF, Figure 6.2 shows

the Interstorey drift graphs from the time history analysis of GMRs scaled using PGA

method for the 15 storey SMRF and the 20 storey SMRF

112
 5
MEAN
MEAN + SD
4

3
LEVEL

2
STORY

0
-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5
INTERSTORY D R I F T (%h)

(a)

10
MEAN
MEAN + SD
LEVEL

6
STORY

0
-1.5 -1 -0.5 0 0.5 1 1.5 2
INTERSTORY D R I F T (%h)

(b)
Figure 6.1 : Interstorey drift graphs from RHA of GMRs scaled using PGA method

(a) 5 storey SMRF (b) 10 storey SMRF

113
 15
LEVE MEAN
MEAN + SD

10
STORY

0
-2 -1.5 -1 -0.5 0 0.5 1 1.5
INTERSTORY D R I F T (%h)

(a)

20
MEAN
MEAN + SD
LEVEL

15
STORY

10

0
-1.5 -1 -0.5 0 0.5 1 1.5
INTERSTORY D R I F T (%h)

(b)
Figure 6.2: Interstorey drift graphs from RHA of GMRs scaled using PGA method
(a) 15 storey SMRF (b) 20 storey SMRF

114
 Table 6.2: Base Shear (KN) from PGA scaling method

EQ Records 5 Storey 10 Storey 15 Storey 20 Storey

naver1 165 203 278 415
naver2 165 158 283 421
naver3 150 178 301 430
naver6 152 169 300 467
naver7 164 177 288 470
naver8 168 171 274 422
naver9 163 165 295 428
naver10 159 173 283 435
naver11 163 193 290 414
naver12 158 191 277 429
naver13 153 199 291 403
naver14 162 178 287 411
naver15 161 184 274 389
kobejap1 169 178 271 404
kobejap3 155 189 286 425
kobejap4 161 200 272 395
northdr1 164 179 269 400
northdr2 168 184 279 405
northdr3 178 187 288 411
northdr4 175 198 305 389

Table 6.2 gives the base shear values in the 5, 10, 15 and 20 storey buildings after the

ground motions are scaled to the design spectrum using the PGA scaling method .

115
6.3.2 PSa Scaling method
The interstorey drift graphs of the 5, 10, 15 and 20 storey frames have been plotted from

the results of the Non-linear time history analysis obtained after the selected GMR for

the time history analysis is scaled using the PSaA scaling method .

Table 6.3: PSa Scaling Method - Summary of Interstorey Drift for Real Ground Motion

SMRF Interstorey Drifts (%hs) of SMRF

GMRs scaled by PSa scaling method
Max. of Mean Max. of (Mean+SD)
+ - + -
5 Storey 1.483261 -0.76901 1.737574 -0.90917
10 Storey 1.125759 -0.61232 1.383671 -0.78159
15 Storey 0.862494 -0.4247 1.14435 -0.54325
20 Storey 0.839577 -0.38752 1.064295 -0.47288
SD = Standard Deviation, hs = Storey height.

Table 6.3 gives the values of mean , Sum of mean and the Standard deviation values of

the interstorey drifts from the PGA Scaling method .The Ground motion record

NAVER3 is found to have the drift close to the positive mean value in the 5 storey

frame, in the 10 storey frame NAVER15 is found to have the drift close to the positive

mean value, in the 15 storey frame the drift close to the positive mean value is obtained

by scaling the record NAVER15, finally in the 20 storey fame NAVER6 ground motion

record produces the modest drift in the building. Figure 6.3 represents the Interstorey

drift graphs from the time history analysis of GMRs scaled using PSa method for the 5

storey SMRF and the 10 storey SMRF, Figure 6.4 represents the Interstorey drift graphs

from the time history analysis of GMRs scaled using PSa method for the 15 storey

SMRF and the 20 storey SMRF.

116
 5
MEAN
MEAN + SD
LEVEL 4

3
STORY

0
-3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5
INTERSTORY D R I F T (%h)
(a)

10
MEAN
LEVE

MEAN + SD
8

6
STORY

0
-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2
INTERSTORY D R I F T (%h)

(b)

Figure 6.3: Interstorey drift graphs from RHA of GMRs scaled using PSa method
(a) 5 storey SMRF (b) 10 storey SMRF

117
LEVE 15
MEAN
MEAN + SD

10
STORY

0
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
INTERSTORY D R I F T (%h)

(a)
20
MEAN
MEAN + SD
LEVEL

15
STORY

10

0
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
INTERSTORY D R I F T (%h)

(b)
Figure 6.4: Interstorey drift graphs from RHA of GMRs scaled using PSa method
(a) 15 storey SMRF (b) 20 storey SMRF

118
 Table 6.4 : Base Shear (KN) from PSa scaling method

GMRs 5 Storey 10 Storey 15 Storey 20 Storey

naver1 135 189 275 386

naver2 145 181 283 402

naver3 162 184 302 412

naver6 125 215 282 395

naver7 127 182 276 394

naver8 124 173 283 406

naver9 133 177 291 393

naver10 136 183 301 404

naver12 152 174 311 415

naver13 143 181 288 418

naver14 137 173 294 422

naver15 144 175 297 385

kobejap3 150 196 305 400

kobejap4 147 191 310 397

northdr1 132 200 313 420

northdr2 134 204 316 396

northdr3 136 210 322 411

northdr4 144 203 330 423

Table 6.4 gives the base shear values in the 5, 10, 15 and 20 storey buildings after the

ground motions are scaled to the design spectrum using the PSa scaling method .

119
6.3.3 Ordinate Scaling method
The interstorey drift graphs of the 5, 10, 15 and 20 storey frames have been plotted from

the results of the Non-linear time history analysis for the selected GMR scaled using the

Ordinate scaling method .

Table 6.5 – Ordinate Scaling Method - Summary of Interstorey Drift for Real Ground
Motion

SMRF Interstorey Drifts (%hs) of SMRF

GMRs scaled by Ordinate scaling method
Max. of Mean Max. of (Mean+SD)
+ - + -
5 Storey 1.367361 -0.71707 1.511227 -0.89374
10 Storey 1.302523 -0.69476 1.714063 -1.00643
15 Storey 1.408986 -0.62363 1.867248 -0.89513
20 Storey 1.099769 -0.59848 1.592114 -1.01768
SD = Standard Deviation, hs = Storey height.

Table 6.5 gives the values of mean, Sum of mean and the Standard deviation values of

the interstorey drifts from the Ordinate Scaling method. The Ground motion record

NAVER6 is found to have the drift close to the positive mean value in the 5 storey

frame, in the 10 storey frame NAVER9 is found to have the drift close to the positive

mean value, in the 15 storey frame the drift close to the mean value is obtained by scaling

the record NAVER6 , finally in the 20 storey fame NAVER13 ground motion record

produces the modest drift in the building . Figure 6.5 represents the Interstorey drift

graphs from the time history analysis of GMRs scaled using the ordinate method for the

5 storey SMRF and the 10 storey SMRF, Figure 6.6 represents the Interstorey drift

graphs from the time history analysis of GMRs scaled using Ordinate method for the 15

storey SMRF and the 20 storey SMRF.

120
 5
MEAN

LEVE
MEAN + SD
4
STORY

0
-4 -3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(a)

10
MEAN
LEVE

MEAN + SD
8

6
STORY

0
-4 -3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(b)

Figure 6.5: Interstorey drift graphs from RHA of GMRs scaled using Ordinate method
(a) 5 storey SMRF (b) 10 storey SMRF

121
 15

LEVEL
MEAN
MEAN + SD
10
STORY

0
-4 -3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(a)
20
MEAN
MEAN + SD
LEVEL

15
STORY

10

0
-4 -3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(b)
Figure 6.6: Interstorey drift graphs from RHA of GMRs scaled using Ordinate method
(a) 15 storey SMRF (b) 20 storey SMRF

122
 Table 6.6: Base Shear (KN) from Ordinate scaling method

GMR 5 Storey 10 Storey 15 Storey 20 Storey

naver1 123 220 278 381

naver2 144 197 294 421

naver3 150 221 284 407

naver6 143 189 293 402

naver7 132 186 297 401

naver8 134 194 286 423

naver9 145 203 299 443

naver10 157 207 315 410

naver11 153 208 322 402

naver12 151 192 283 395

naver13 148 183 285 386

naver14 154 185 288 379

naver15 169 191 287 388

kobejap2 172 177 274 412

kobejap3 175 184 283 414

kobejap4 173 179 286 411

northdr1 144 193 277 398

northdr2 146 192 287 396

northdr3 158 178 285 392

northdr4 149 186 281 390

Table 6.6 gives the base shear values in the 5, 10, 15 and 20 storey buildings after the

ground motions are scaled to the design spectrum using the Ordinate scaling method .

123
6.3.4 Partial area scaling method
The interstorey drift graphs of the 5, 10, 15 and 20 storey frames have been plotted from

the results of the Non-linear time history analysis for the selected GMR scaled using the

Partial area scaling method .

Table 6.7 – Partial Area Scaling Method -Summary of Interstorey Drift for Real Ground
Motion

SMRF Interstorey Drifts (%hs) of SMRF

GMRs scaled by Partial area scaling method
Max. of Mean Max. of (Mean+SD)
+ - + -
5 Storey 1.26888 -0.65818 1.58834 -0.81005
10 Storey 1.153488 -0.62655 1.396499 -0.80613
15 Storey 0.922696 -0.43006 1.232704 -0.59044
20 Storey 0.838254 -0.39392 1.15148 -0.55283
SD = Standard Deviation, hs = Storey height.

Table 6.7 gives the values of mean , Sum of mean and the Standard deviation values of

the interstorey drifts from the Partial area Scaling method .The Ground motion record

KOBJAP1 is found to have the drift close to the positive mean value in the 5 storey

frame, in the 10 storey frame NAVER10 is found to have the drift close to the positive

mean value, in the 15 storey frame the drift close to the positive mean value is obtained

by scaling the record NAVER15, finally in the 20 storey fame NAVER14 ground motion

record produces the modest drift in the building . Figure 6.7 represents the Interstorey

drift graphs from the time history analysis of GMRs scaled using Partial area method for

the 5 storey SMRF and the 10 storey SMRF, Figure 6.8 represents the Interstorey drift

graphs from the time history analysis of GMRs scaled using the Partial area method for

the 15 storey SMRF and the 20 storey SMRF

124
 5
MEAN
MEAN + SD
LEVEL 4

3
STORY

0
-4 -3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(a)

10
MEAN
LEVE

MEAN + SD
8
STORY

0
-4 -3 -2I N T E R-1S T O R Y0 D R I F1T (%h) 2 3 4

(b)

Figure 6.7: Interstorey drift graphs from RHA of GMRs scaled using Partial area scaling

(a) 5 storey SMRF (b) 10 storey SMRF

125
 15
MEAN
LEVEL MEAN + SD

10
STORY

0
-4 -3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(a)
20
MEAN
MEAN + SD
LEVE

15
STORY

10

0
-4 -3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(b)
Figure 6.8: Interstorey drift graphs from RHA of GMRs scaled using Partial area scaling
(a) 15 storey SMRF (b) 20 storey SMRF

126
 Table 6.8: Base Shear (KN) from Partial area scaling method

GMRs 5 Storey 10 Storey 15 Storey 20 Storey

naver1 132 178 335 511

naver2 143 192 351 527

naver3 122 193 356 524

naver6 151 194 354 519

naver7 138 188 344 531

naver8 135 185 340 528

naver9 148 187 342 532

naver10 147 179 311 541

naver11 139 178 329 550

naver12 147 185 336 550

naver13 143 186 327 525

naver14 156 180 331 529

naver15 141 204 324 526

kobejap1 158 183 334 536

kobejap2 156 182 353 538

kobejap3 153 188 348 542

kobejap4 151 189 342 547

northdr1 148 213 352 552

northdr2 139 214 353 551

northdr3 143 197 357 522

northdr4 139 186 362 520

Table 6.8 gives the base shear values in the 5, 10, 15 and 20 storey buildings after the

ground motions are scaled to the design spectrum using the Partial area scaling method .

127
6.3.5 ASCE 2007 scaling method
The interstorey drift graphs of the 5, 10, 15 and 20 storey frames have been plotted from

the results of the Non-linear time history analysis for the selected GMR scaled using the

ASCE 2007 scaling method .

Table 6.9 –ASCE 2007 Scaling Method- Summary of Interstorey Drift for Real Ground
Motion

SMRF Interstorey Drifts (%hs) of SMRF

GMRs scaled by ASCE 2007 scaling method
Max. of Mean Max. of (Mean+SD)
+ - + -
5 Storey 1.22 -0.68 1.83 -0.81
10 Storey 1.08 -0.57 1.30 -0.73
15 Storey 0.89 -0.43 1.07 -0.57
20 Storey 1.00 -0.50 1.18 -0.64
SD = Standard Deviation, hs = Storey height.

Table 6.9 gives the values of mean, Sum of mean and the Standard deviation values of

the interstorey drifts from the ASCE 2007 Scaling method .The Ground motion record

NAVER10 is found to have the drift close to the positive mean value in the 5 storey

frame, in the 10 storey frame NAVER15 is found to have the drift close to the mean

value, in the 15 storey frame the drift close to the mean value is obtained by scaling the

record NAVER14, finally in the 20 storey fame NAVER13 ground motion record

produces the modest drift in the building . Figure 6.9 represents the Interstorey drift

graphs from the time history analysis of GMRs scaled using ASCE 2007 method for the

5 storey SMRF and the 10 storey SMRF, Figure 6.10 represents the Interstorey drift

graphs from the time history analysis of GMRs scaled using the ASCE 2007 method for

the 15 storey SMRF and the 20 storey SMRF

128
 5
MEAN
MEAN + SD
LE
4
STORY

0
-3 -2 -1 0 1 2 3
INTERSTORY D R I F T (%h)

(a)

10
MEAN
LE

8
MEAN + SD
STORY

0
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
INTERSTORY D R I F T (%h)

(b)

Figure 6.9: Interstorey drift graphs from RHA of GMRs scaled using ASCE 2007 scaling
(a) 5 storey SMRF (b) 10 storey SMRF

129
 15

LEVEL
MEAN
MEAN + SD

10
STORY

0
-1.5 -1 -0.5 0 0.5 1 1.5
INTERSTORY D R I F T (%h)

(a)
20
MEAN
MEAN + SD
LEVEL

15
STORY

10

0
-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2
INTERSTORY D R I F T (%h)

(b)
Figure 6.10: Interstorey drift graphs from RHA of GMRs scaled using ASCE 2007
scaling (a) 15 storey SMRF (b) 20 storey SMRF

130
 Table 6.10: Base Shear (KN) from ASCE 2007 scaling method

GMRs 5 Storey 10 Storey 15 Storey 20 Storey

naver1 223 264 334 512

naver2 218 273 346 523

naver3 219 268 342 526

naver6 226 272 348 533

naver7 236 281 346 519

naver8 238 279 352 523

naver9 246 266 357 527

naver10 241 262 353 533

naver13 239 271 348 531

naver14 247 275 332 535

naver15 255 273 337 538

kobejap3 210 283 339 545

kobejap4 249 282 352 555

northdr1 256 274 354 523

northdr2 254 276 351 535

northdr3 255 278 357 547

Table 6.10 gives the base shear values in the 5, 10, 15 and 20 storey buildings after the

ground motions are scaled to the design spectrum using the ASCE 2007 scaling method .

131
6.3.6 Least Square scaling method
The interstorey drift graphs of the 5, 10, 15 and 20 storey frames have been plotted from

the results of the Non-linear time history analysis for the selected GMR scaled using the

Least square scaling method .

Table 6.11 - Summary of Interstorey Drift for Real Ground Motion from Least Square
Scaling Method

SMRF Interstorey Drifts (%hs) of SMRF

GMRs scaled by Least Square scaling method
Max. of Mean Max. of (Mean+SD)
+ - + -
5 Storey 1.06 -0.60 1.86 -0.91
10 Storey 1.10 -0.62 1.49 -0.89
15 Storey 0.81 -0.36 1.46 -0.55
20 Storey 0.54 -0.25 1.02 -0.46
SD = Standard Deviation, hs = Storey height.

Table 6.11 gives the values of mean , Sum of mean and the Standard deviation values

of the interstorey drifts from the Least Square Scaling method .The Ground motion

record NAVER10 is found to have the drift close to the mean value in the 5 storey frame,

in the 10 storey frame NRD2 is found to have the drift close to the mean value, in the 15

storey frame the drift close to the mean value is obtained by scaling the record NAVER8,

finally in the 20 storey fame NRD2 ground motion record produces the modest drift in

the building. Figure 6.11 represents the Interstorey drift graphs from the time history

analysis of GMRs scaled using Least square method for the 5 storey SMRF and the 10

storey SMRF, Figure 6.12 represents the Interstorey drift graphs from the time history

analysis of GMRs scaled using the Least square method for the 15 storey SMRF and

the 20 storey SMRF

132
 5
MEAN

LEVE
MEAN + SD
4

3
STORY

0
-4 -3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(a)

10
MEAN
MEAN + SD
LEVEL

6
STORY

0
-4 -3 -2 -1 0 1 2 3
INTERSTORY D R I F T (%h)

(b)

Figure 6.11 : Interstorey drift graphs from RHA of GMRs scaled using Least square

scaling: (a) 5 storey SMRF (b) 10 storey SMRF

133
 15
MEAN
LEV MEAN + SD

10
STORY

0
-3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(a)

20
MEAN
MEAN + SD
LEVEL

15
STORY

10

0
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5
INTERSTORY D R I F T (%h)

(b)
Figure 6.12: Interstorey drift graphs from RHA of GMRs scaled using Least square
scaling (a) 15 storey SMRF (b) 20 storey SMRF

134
 Table 6.12: Base shear (KN) from Least square scaling method

GMR 5 Storey 10 Storey 15 Storey 20 Storey

naver1 232 261 357 541

naver2 238 265 343 534

naver3 221 266 352 532

naver6 224 273 358 536

naver7 237 275 342 528

naver8 225 277 341 525

naver9 235 278 340 541

naver10 238 264 356 543

naver11 229 273 361 546

naver12 227 273 358 538

naver13 228 271 355 544

naver14 233 282 364 522

naver15 246 288 362 528

kobejap2 251 283 358 531

kobejap3 254 280 361 536

kobejap4 253 274 363 538

northdr1 257 277 372 529

northdr2 253 274 376 533

northdr3 251 279 374 537

northdr4 249 283 379 545

Table 6.12 gives the base shear values in the 5, 10, 15 and 20 storey buildings after the

ground motions are scaled to the design spectrum using the Least square scaling method .

135
6.4 Discussion on the interstorey drift results from spectrum compatible
records
6.4.1 Spectral matching using Seismo Match software
The interstorey drift graphs of the 5, 10, 15 and 20 storey frames have been plotted from

the results of the Non-linear time history analysis for the selected GMR matched with

the target spectrum using the Seismomatch software.

Table 6.13- Summary of Interstorey Drift for Real Ground Motion from Seismo Match
Scaling method

SMRF Interstorey Drifts (%hs) of SMRF

Spectral matching using Seismo match
Max. of Mean Max. of (Mean+SD)
+ - + -
5 Storey 0.99 -0.56 1.46 -0.80
10 Storey 0.68 -0.40 0.82 -0.54
15 Storey 0.52 -0.30 0.63 -0.37
20 Storey 0.50 -0.32 0.56 -0.38
SD = Standard Deviation, hs = Storey height.

Table 6.13 gives the values of mean , Sum of mean and the Standard deviation values

of the interstorey drifts from the spectral matched records from the Seismomatch

software .The Ground motion record NAVER8 is found to have the drift close to the

mean value in the 5 storey frame, in the 10 storey frame NRD3 is found to have the drift

close to the mean value, in the 15 storey frame the drift close to the mean value is

obtained by scaling the record KOBJAP4, finally in the 20 storey fame NAVER1 ground

motion record produces the modest drift in the building. . Figure 6.13 represents the

Interstorey drift graphs from the time history analysis of GMRs scaled using the

Seismomatch software for the 5 storey SMRF and the 10 storey SMRF, Figure 6.14

represents the interstorey drift graphs from the time history analysis of GMRs scaled

using the Seismomatch software for the 15 storey SMRF and the 20 storey SMRF.

136
 5
MEAN
MEAN + SD
LEVEL
4

3
STORY

0
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
INTERSTORY D R I F T (%h)
(a)
10
MEAN
MEAN + SD
LEV

6
STORY

0
-1.5 -1 -0.5 0 0.5 1 1.5
INTERSTORY D R I F T (%h)

(b)
Figure 6.13: Interstorey drift graphs from RHA of Spectral matching using Seismo match

(a) 5 storey SMRF (b) 10 storey SMRF

137
 15

LEVE
MEAN
MEAN + SD

10
STORY

0
-1 -0.5 0 0.5 1 1.5
INTERSTORY D R I F T (%h)
(a)

20
MEAN
MEAN + SD
LEVEL

15
STORY

10

0
-1 -0.5 0 0.5 1 1.5
INTERSTORY D R I F T (%h)

(b)
Figure 6.14: Interstorey drift graphs from RHA Spectral matching using Seismo match

(a) 15 storey SMRF (b) 20 storey SMRF

138
 Table 6.14 - Base Shear (KN) from Seismo Match Spectral matching

GMRs 5 Storey 10 Storey 15 Storey 20 Storey

naver1 226 318 355 532

naver2 224 295 350 534

naver3 236 315 352 540

naver6 233 316 349 542

naver7 235 269 339 545

naver8 238 274 348 547

naver9 243 278 351 541

naver10 241 273 352 537

naver11 246 272 356 524

naver12 248 279 361 485

naver13 238 284 364 515

naver14 234 289 366 522

naver15 253 276 364 531

kobejap2 244 294 371 543

kobejap3 246 273 376 551

kobejap4 248 275 352 539

northdr1 254 278 357 538

northdr2 257 277 365 533

northdr3 249 285 344 528

northdr4 254 288 333 541

Table 6.14 gives the base shear values in the 5, 10, 15 and 20 storey buildings after the

ground motions are scaled to the design spectrum using the Seismomatch software .

139
6.5 Discussion on the interstorey drift results from Atkinson’s artificial

records

6.5.1 Spectral matching using Atkinson’s artificial records

The interstorey drift graphs of the 5, 10, 15 and 20 storey frames have been plotted

from the results of the Non-linear time history analysis obtained using the Atkinson’s

artificial records.

Table 6.15 - Summary of Interstorey Drift for Real Ground Motion from Atkinson’s
artificial earthquake records

SMRF Interstorey Drifts (%hs) of SMRF

Spectral matching using Atkinson’s artificial records
Max. of Mean Max. of (Mean+SD)
+ - + -
5 Storey 2.61 -0.99 3.48 -1.32
10 Storey 1.92 -0.61 2.71 -0.88
15 Storey 1.32 -0.43 1.82 -0.64
20 Storey 2.40 -0.45 3.76 -0.59
SD = Standard Deviation, hs = Storey height.

Table 6.15 gives the values of mean, Sum of mean and the Standard deviation values

of the interstorey drifts while using the Atkinson’s artificial earthquake records . Figure

6.15 represents the Interstorey drift graphs from the time history analysis using

Atkinson’s artificial records for the 5 storey SMRF and the 10 storey SMRF, Figure

6.16 represents the Interstorey drift graphs from the time history analysis using the

Atkinson’s artificial records for the 15 storey SMRF and the 20 storey SMRF

140
 5
MEAN
MEAN + SD
4
LEVEL
3

2
STORY

0
-4 -3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(a)

10
MEAN
LEVE

MEAN + SD
8

6
STORY

0
-4 -3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(b)

Figure 6.15: Interstorey drift graphs from RHA of Spectral matching using Atkinson’s

artificial earthquake records (a) 5 storey SMRF (b) 10 storey SMRF

141
 15

LEVE
MEAN
MEAN + SD

10
STORY

0
-4 -3 -2 -1 0 1 2 3 4
INTERSTORY D R I F T (%h)

(a)
20
MEAN
MEAN + SD
LEVEL

15
STORY

10

0
-4 -3 -2 -1 0 1 2 3 4 5 6
INTERSTORY D R I F T (%h)

(b)
Figure 6.16: Interstorey drift graphs from RHA Spectral matching using Atkinson’s

artificial earthquake records (a) 15 storey SMRF (b) 20 storey SMRF

142
 Table 6.16: Base Shear (KN) from Atkinson’s artificial records

GMR 5 Storey 10 Storey 15 Storey 20 Storey

LP1 145 185 298 412

LP2 140 215 283 396

LP3 137 190 271 411

LP4 136 203 265 423

SP1 128 206 305 432

SP2 125 207 315 429

SP3 121 212 320 425

SP4 135 218 335 430

Table 6.16 gives the base shear values in the 5, 10, 15 and 20 storey buildings after the

ground motions are scaled to the design spectrum using the Atkinson’s artificial records .

6.6 Summary

The results of the response history analysis using scaled and artificial ground motion

records are summarized in Tables 6.17 through 6.20. In these tables ISD indicates the

mean values of the interstorey drift envelope, SD is standard deviation, Disp is the

measure of dispersion which is defined here as the standard deviation expressed as a

percentage of the mean value, and BS indicates the base shear obtained from the dynamic

time history analysis.

143
 Table 6.17: Summary of interstorey drift and base shear for the 5-storey frame

ISD, SD, Dispersion, ISD+SD, Min BS, Max BS,

Scaling method %h %h %ISD %h kN kN
PGA 1.24 0.36 29.2 1.61 150 178
PSa 1.48 0.25 17.1 1.74 124 162
Ordinate 1.37 0.14 10.5 1.51 123 173
Partial Area 1.27 0.32 25.2 1.59 122 158
ASCE-7 1.22 0.61 50.0 1.83 210 255
Least Square 1.06 0.80 75.5 1.86 221 257
Spectrum Match 0.99 0.47 47.5 1.46 224 254
Atkinson 2.61 0.87 33.3 3.48 121 145

Table 6.18 : Summary of interstorey drift and base shear for the 10-storey frame

ISD, SD, Dispersion, ISD+SD, Min BS, Max BS,

Scaling method %h %h %ISD %h kN kN
PGA 0.88 0.30 33.5 1.18 165 203
PSa 1.13 0.26 22.9 1.38 173 215
Ordinate 1.30 0.41 31.6 1.71 183 220
Partial Area 1.15 0.24 21.1 1.40 178 214
ASCE-7 1.08 0.22 20.4 1.30 264 283
Least Square 1.10 0.39 35.5 1.49 261 288
Spectrum Match 0.68 0.14 20.6 0.82 272 318
Atkinson 1.92 0.79 41.1 2.71 185 218

Table 6.19 : Summary of interstorey drift and base shear for the 15-storey frame

ISD, SD, Dispersion, ISD+SD, Min BS, Max BS,

Scaling method %h %h %ISD %h kN kN
PGA 0.67 0.27 40.1 0.94 269 305
PSa 0.86 0.28 32.7 1.14 275 330
Ordinate 1.41 0.46 32.5 1.87 274 322
Partial Area 0.92 0.31 33.6 1.23 311 362
ASCE-7 0.89 0.18 20.2 1.07 332 357
Least Square 0.81 0.65 80.2 1.46 340 379
Spectrum Match 0.52 0.11 21.2 0.63 333 376
Atkinson 1.32 0.50 37.9 1.82 271 335

144
 Table 6.20 : Summary of interstorey drift and base shear for the 20-storey frame

ISD, SD, Dispersion, ISD+SD, Min BS, Max BS,

Scaling method %h %h %ISD %h kN kN
PGA 0.67 0.16 23.6 0.83 389 470
PSa 0.84 0.22 26.8 1.06 385 423
Ordinate 1.10 0.49 44.8 1.59 379 443
Partial Area 0.84 0.31 37.4 1.15 511 552
ASCE-7 1.00 0.18 18.0 1.18 512 555
Least Square 0.54 0.48 88.9 1.02 522 545
Spectrum Match 0.50 0.06 12.0 0.56 485 547
Atkinson 2.40 1.36 56.7 3.76 396 432

It is clear from these tables that the results of the dynamic time history analysis involves a

significant uncertainty irrespective of the method used for scaling the ground motion

records. Most methods produce large dispersion (i.e., more than 30%) in majority of the

cases. Even the artificial records produce 33% to 57% dispersion of the interstorey drift

values. Among all the methods used for scaling or matching the ground motion records to

the design level of seismic hazard, the frequency domain matching of the spectral shape

as performed here by Seismo Match seem to produce the best results by limiting the level

of dispersion to less than 30% except in the case of the 5-storey frame. In case of the 5-

storey frame, the ordinate method of scaling produces the best result (10.5% dispersion).

In cases of the 10, 15 and 20 storey buildings, the spectral matching of the ground motion

records by SeismoMatch produces the best results with the following levels of dispersion:

20.6%, 21.2% and 12.0%, respectively.

The ranges of the base shear (BS) obtained from the dynamic analysis using different

methods of scaling of the ground motion records show a similar level of variability.

145
However, in comparison to the design base shear as reported in Table 5.2, the base shear

from the dynamic analysis are found to be in the acceptable range. For example, the

design base shear from the 5-storey building frame is 178 kN as shown in Table 5.2,

while the minimum and maximum values of the base shear obtained from the time history

analysis are found to be 121 kN and 254 kN, respectively (considering all methods as

shown in Table 6.17). Similar observation can be made for the other buildings.

The scaling methods show different options available to the engineer to employ in scaling

the GMR used in time history analysis. However, in controlling the response parameters

and to minimize the effect of the scaling method employed engineering judgment has to

be exercised under adequate supervision to obtain the design spectrum from the available

ground motion. There is wide range of variability in the response quantities (e.g.,

interstorey drift) irrespective of the ground motion scaling techniques used. Among all

the methods of scaling and spectral matching of the ground motion records, the frequency

domain spectral matching seem to produce the best results as the dispersion in the results

are observed to be lower than that in other cases. However, the interstorey drift obtained

from the time history analysis using different scaling methods show a uniform and

consistent pattern of deformation in low rise to medium rise frames, whereas a greater

dispersion of the results has been observed in tall buildings. Although a similar level of

variability is observed in the base shear obtained from the dynamic time history analysis,

they are consistent with the design base shear from the corresponding buildings.

146
 Chapter 7

Observations and Conclusions

7.1 Observations
In the present research the seismic response of steel buildings with moment resisting

fames has been carried out using static and dynamic analysis procedures. Different methods

available for the performance-based seismic design of buildings have been examined in the

context of the buildings designed according to the current building code of Canada. In the

nonlinear time history analysis, the effect of different available ground motion scaling techniques

on the seismic response of the buildings has been studied. Four identical type of steel moment

frames with varying height i.e., 5, 10, 15, 20 storey frames have been used in the study for the

evaluation of seismic performance of moment resisting steel frame buildings. The force-based

design provision or the equivalent static load method prescribed in the National Building Code of

Canada (NBCC 2010) has been adopted in the design of the steel frames. The design of the

building using the code based method is further evaluated through rigorous static and dynamic

nonlinear analysis to check the performance of the buildings. The Nonlinear static Pushover

analysis is carried out to evaluate the ductility capacity under seismic action. In this method the

frame is pushed to a targeted roof displacement by applying the seismic force as lateral force

varying in a inverted triangular shape, through the Nonlinear static pushover analysis the

ductility capacity is obtained using the yield displacement and ultimate displacement of the

SMRF. The ductility capacities as obtained from the pushover analysis obtained are 1.94, 2.10,

1.88 and 2.0 for twenty, fifteen, ten and five storey frames, respectively. Therefore, it is inferred

that the ductility capacity of the buildings are lower than that assumed in the building design. A
148
range of performance-based design methodologies based on the static procedure has been

examined in the context of the static pushover analysis and it is observed that the design

procedure as provided in the current code NBCC 2010 can be used for producing a building

design to achieve the life safety and collapse prevention levels of performance.

In the next step, Nonlinear Time history analysis (RHA) has been carried out for each of the real

ground motion used coupled with every scaling techniques considered for the study to evaluate

the seismic demand of the buildings to consider the effect of scaling technique used. The real

ground motions are scaled to make them compatible to the specified location of the buildings

beforehand whereas the synthesized ground motions are readily compatible to be used in the

RHA . The mean interstorey drift value (M) and the mean plus standard deviation (M+SD) are

computed from the RHA of the real ground motions. It is further noted that any of the eight

scaling technique tested in the study can be easily used for scaling real ground motions in

practice, the study also reveals that response from the scaled ground motions are observed to be

more coherent and less dispersed in 5 and 10 storey frames whereas the dispersion was found to

be deviant in 15 and 20 storey frames, the interstorey drift obtained from the tested scaling

methods confirmed to the code NBCC 2010 specified limit of 2.5% interstorey drift. However

the Least square scaling method and the synthesized records displayed interstorey drift exceeding

2.5% in certain ground motions. The value of mean plus standard deviation of the interstorey

drift for real ground motion varies from 1.55% of storey height to 2.5% of storey height. The

maximum value of interstorey drift for long period synthesized records varies from 1.20% to

2.18% and the maximum value of the interstorey drift for short period records varies from 1.93%

of height of the storey to 2.35%.

149
The NBCC 2010 specifies only one performance objective - collapse limit at 2.5% interstorey

drift. On the other hand, other standards like FEMA-273 (1997) and Vision 2000 (SEAOC,

1995) specify different level of performance with different performance objectives for a more

robust design. FEMA-273 (1997) specifies the interstorey drift limit for Immediate Occupancy

(IO) performance objective as less than 2%. NBCC 2010 drift specification of 2.5% for collapse

prevention (CP) is considered is considered to in line with those standards. However, it may

produce a conservative design in some cases such as the low rise building (e.g., the five storey

building considered here). The Non-linear static analysis carried out in the light of the

performance-based design of structures also shows that NBCC design procedure is conservative

in some cases.

7.2 Conclusions

It is important to note that while the techniques utilized in the present work are available in the

literature, the main contribution of the present work lies in the study of the effect of ground

motion scaling methods in estimating the seismic response of a set of buildings designed

according to the latest version of the Canadian code, and the interpretation of the seismic

response of these buildings estimated using static procedures in the context of performance-

based design. The present study provides an important insight into the sensitivity of the

estimated dynamic response of buildings to ground motion scaling. It also shows how the

existing performance-based methods can be utilized in the design and evaluation of the new

buildings based on the static pushover analysis. Based on the present study, the following

conclusions are drawn.

150
 The code prescribed force-based design procedure is found to be slightly

conservative. However, it can be used to predict or ensure the performance of the

structure for life safety and collapse prevention.

 Non-linear static and nonlinear dynamic analysis in time domain are required to

design the building in order to achieve the stated performance objective, the code

based equivalent static load procedure holds good for a large class of conventional

and simple structures with no complexities in design or geometry .

 The ductility capacity assumed in the force-based design is also unrealistic and the

assumed ductility is hard to achieve

 It is a noted observation in pushover analysis that the roof displacement

corresponding to instability or 2.5% interstorey drift is greater than the roof

displacement corresponding to maximum of M+SD. Where, M & SD are the mean

and standard deviation of the interstorey obtained from Response History Analysis

 The chosen method of scaling the real ground motion has a direct effect on the

seismic demand of the building which usually varies widely for different methods of

scaling. Hence a suitable scaling technique has to be chosen so as to keep the seismic

demand and the related dispersion in control.

 Artificial or synthesized records may be chosen in absence of real records as they are

known to give distorted dispersion of the seismic demand. However, the artificial

records may not necessarily produce better results. It may produce the same level of

uncertainty as for the scaled real records.

 Ground motions produced from a suit of real records by changing them in the

frequency domain to match the design spectrum seem to work better and the level of

151
 dispersion produced by these records is found to be lower than the cases when other

scaling methods are used.

7.3 Scope for future work

The scope for further research lies in exploring ways to the possibility of new scaling techniques

that can control the dispersion in the response more effectively. Also, other types of buildings

should be studied to understand the effect of ground motion scaling such that a coherent scaling

technique can be developed in the future.

152
REFERENCES

Abrahamson, N.A. (1992). Software program – RSPMatch, Non-Stationary Spectral Matching.

Seismological Research Letters, 63:1, 30.

Abrahamson, N.A., Hancock , J., Watson-Lamprey J., Markatis, A., McCOY, E., MENDIS, R.,

2006. Program SeismoMatch v2 – software capable of adjusting earthquake

accelerograms to match a specific design response spectrum, using the wavelets

algorithm proposed by Abrahamson [1992] and Hancock et al. [2006].

http://www.seismosoft.com/en/SeismoMatch.aspx (cited October 2005).

Akbas B., Shen J., Kara FéI., Tugsal U.M., 2003, Seismic behavior and Pushover Analyses in

Steel Frames, fifth National Conference on Earthquake Engineering, 26-30 May,

Istanbul, Turkey, Paper No: AT-053

Alavi, B., and Krawinkler, H. (2000). “Consideration of near-fault ground motion effects in

seismic design.” Proc. of the 12th World Conf. on Earthquake Engineering, Paper No.

2665, New Zealand Society of Earthquake Engineering, Silverstream, New Zealand.

Ambraseys, N. N., Douglas, J., Rinaldis, D., Berge-Thierry, C., Suhadolc, P., Costa, G.,Sigb

jomsson, R. and Smit , P. [ZOO41 "Dissemination of European strong-mot ion data,Vol.

2,". CD-ROM Collection, Engineering and Physical Sciences Research Council,United

Kingdom.

ASCE, 2000. Standard ASCE 4-98, Seismic analysis of safety-related nuclear structures and

commentary, American Society of Civil Engineers, Reston, VA.

ASCE, 2005, ASCE Standard ASCE/SEI 7-05. Minimum Design Loads for Buildings and Other

Structures, Including Supplement No.1, American Society of Civil Engineers, Reston,

VA.

153
Aschheim M., 2004, Yield Point Spectra: A Simple Alternative to the Capacity Spectrum

Method, ASCE online research library http://www.ascelibrary.org (cited October 2005)

ATC, 1978. Tentative Provisions for the Development of Seismic Regulations for Buildings

(ATC 3-06), Applied Technology Council, Redwood City, CA. (A second printing, in

1984, includes 1982 amendments).

Atkinson, G.M. 2009. Earthquake time histories compatible with the 2005 National Building

Code of Canada uniform hazard spectrum. Canadian Journal of Civil Engineering, 36:

991-1000.

Bagchi, 2001, Evaluation of the Seismic Performance of Reinforced Concrete Buildings, Ph. D

thesis, Department of Civil and Environmental Engineering, Carlton University, Ottawa,

Canada

Beresnev, I., and G. Atkinson (1998). Stochastic finite-fault modeling of ground motions from

the 1994 Northridge, California earthquake. I.Validation on rock sites, Bull. Seism. Soc.

Am. 88, 1392–1401.

Beyer, K., and Bommer, J. 2007. Selection and scaling of real accelerograms for bi-directional

loading: a review of current practice and code provisions. Journal of Earthquake

Engineering, 11: 13-45.

Bommer, J.J. and Aceveo, A.B., 2004. The use of real earthquake accelerograms as input to

dynamic analysis. Journal of Earthquake Engineering, 8(1), 43–91.

Boore, D. (2003). Prediction of ground motion using the stochastic method, Pure Appl. Geophys.

160, 635–676.

Bozorgnia, Y., and Mahin, S. A. (1998). “Ductility and strength demands of near-fault ground

structure-specific scalar intensity measures 389 motions of the Northridge earthquake.”

154
 Proc. of the 6th U.S. National Conf. on Earthquake Engineering, Earthquake Engineering

Research Institute, Seattle.

Bozorgnia, Y.A., Bertero, V.V.A., 2004. Earthquake Engineering: From Engineering

Seismology to Performance-Based Engineering , Text Book , CRC Press.

Chopra A.K. and Goel R.K., 1999, Capacity-Demand-Diagram Method for Estimating Seismic

deformation of Inelastic Structures: SDF systems, Report No.PEER-1999/02 Pacific

Earthquake Engineering Research Center, University of California, Berkeley, USA

Chopra, A. K., and Chinatanapakdee, C. 2004. “Inelastic deformation ratios for design and

evaluation of structures: Single-degree-of-freedom bilinear systems.” J. Struct. Eng.,

130(9), 1309–1319.

CISC, 2010. Handbook of Steel Construction, Canadian Institute of Steel Construction,

Willowdale, Ontario, Canada.

Cornell, C.A. (1968). Engineering seismic risk analysis, Bull. Seism. Soc. Am., 58, 1583-1606.

CSA. 2009. Design of Steel Structures, CSA-S16-09, Canadian Standards Association, Toronto,

ON.

CSI, 2012a, ETABS v9.5 Integrated analysis, design and drafting of building systems, Computer

Software, Computers and Structures, Inc., http://www.csiberkeley.com/ (cited Dec

2012).

CSI, 2012b, PERFORM 3D v5 NONLINEAR ANALYSIS AND PERFORMANCE

ASSESSMENT FOR 3D STRUCTURES, Computer Software, Computers and

Structures, Inc., 2012, http://www.csiberkeley.com/ (cited Dec 2012).

155
DeVall R.H, 2003, Background information for some of the proposed earthquake design

provisions for the 2005 edition of the National Building Code of Canada, Canadian

Journal of Civil Engineering. 30: 279-286,

EC8, 2004. Eurocode 8: Design of structures for earthquake resistance, Part 1 – General rules,

seismic actions and rules for buildings, European Committee for Standardization,

Brussels, Belgium.

Fahjan, Y.M., Ozdemir, Z. and Keypour, H. 2007. Procedures for real earthquake time histories

scaling and applicationto fit iranian design spectra, 5th International Conference on

Seismology and Earthquake Engineering (SEE5), May 14-16, Tehran, Iran.

Fahjan Y. and Ozdemir Z., 2008. “Scaling of earthquake accelerograms for non‐ linear dynamic

analyses to match the earthquake design spectra”, The 14th World Conference on

Earthquake Engineering (14WCEE), Beijing, China.

Fajfar P.,2000, A Nonlinear Analysis Method for Performance Based Seismic Design,

Earthquake Spectra, Vol.16, No.3pp.573-592

FEMA-273, 1997, NEHRP Guidelines for the Seismic Rehabilitation of the Buildings, prepared

for the Building Seismic Safety Council (BSSC), Washington, D.C. USA

Gasparini, D.A., and Vanmarcke, E.H. 1976. SIMQKE: A program for artificial motion

generation. User's manual and documentation, Department of Civil Engineering,

Massachusetts Institute of Technology, Cambridge, Massachusetts.

Ghobarah A., 2001, Performance-based design in earthquake engineering: state of development,

Engineering Structures 23 : 878-884.

156
Gupta A. and Krawinkler H.,2000,Estimation of seismic drift demands for frame structures,

Earthquake Engineering and Structural Dynamics, Earthquake Engng Struct.Dyn.2000;

29:1287-1305

Hancock J, Bommer JJ. A state-of-knowledge review of the influence of strong-motion duration

on structural damage. Earthquake Spectra 2006;22(3):827–45.

Hannan M., 2006, Earthquake Resistant Design of a Twenty Story Building with Steel Moment

Resisting Frames, M.Eng. Project , Department of Building Civil and Environmental

Engineering, Concordia University, Montreal, Canada

Herrera R., Ricles J. M., Sause R., Lewis B., 2003, Seismic Performance Evaluation of Steel

Moment Resisting Frames With Concrete Filled Tube Columns ‘, Dept. of Civil Eng.,

Lehigh University, Bethlehem, Pennsylvania, http://www.ncree.org.tw (cited October

2005)

Humar, J. and Mahgoub M. A., 2003, Determination of seismic design forces by equivalent static

load method, Canadian Journal of Civil Engineering, Volume 30 287-307.

Humar J. and Ghorbanie-Asl M., 2005, A New Displacement-Based Design Method for Building,

33rd Annual General Conference of the Canadian Society for Civil Engineering, GC-136,

Toronto, Ontario, Canada.

Idriss, I.M, 1993. ”Procedures for selecting ground motions at rock sites, “Report to the U.S.

Department of Commerce . National Institute of Standards and Technology .

Kalkan, E. and Chopra, A. (2011). ”Modal-Pushover-Based Ground-Motion Scaling Procedure.”

J. Struct. Eng. 137, SPECIAL ISSUE: Earthquake Ground Motion Selection and

Modification for Nonlinear Dynamic Analysis of Structures, 298–310.

157
Katsanos, E.I., Sextos, A.G., and Manolis, G.D. 2010. Selection of earthquake ground motion

records: a state-of-the-art review from a structural engineering perspective. Soil

Dynamics and Earthquake Engineering, 30: 157-169.

Kaul, M.K. (1978). Spectrum consistent time-history generation, ASCE J. Eng Mech. EM4, 781-

788.

Leger, P., Tremblay, R. ‘Earthquake Ground Motions for Seismic Damage Assessment and Re-

Evaluation of Existing Buildings and Critical Facilities”, 2009, Book Section Page 193-

219 - Damage Assessment and Reconstruction after War or Natural Disaster ,

Organisation - NATO Science for Peace and Security Series C: Environmental Security

Edited by : Ibrahimbegovic, Adnan, Zlatar, Muhamed, Springer Netherlands 2009

Lilhanand, K. and Tseng, W. S. [I9881 "Development and application of realistic earthquake

time histories' compatible with multiple-damping design spectra," Proceedings of the

Ninth World Conference on Earthquake Engineering, Tokyo-Kyoto 2, 819-824.

Luco, N., and Cornell, A. C. (2007). “Structure-Specific Scalar Intensity Measures for Near-

Source and Ordinary Earthquake Ground Motions,” Earthquake Spectra, Vol. 23, No. 2,

pp.357–392.

McGuire, R. K. [1995] "Probabilistic seismic hazard analysis and design earthquakes: Closing

the loop,", Bulletin of the Seismological Society of America 85, 1275-1284.

Miranda, E. (1993). “Evaluation of site-dependent inelastic seismic designspectra.” J. Struct.

Eng., 119(5), 1319–1338.

Mukherjee, S. and Gupta, V.K., 2002. Wavelet-based generation of spectrum compatible time

histories. Soil Dynamic and Earthquake Engineering, 22(9), 799–804.

158
Nau, J. and Hall, W. (1984). ”Scaling Methods for Earthquake Response Spectra.” J. Struct.

Eng., 110(7), 1533–1548

Naumoski, N. 2001. Program SYNTH – Generation of artificial accelerograms compatible with a

target spectrum. User's manual, Department of Civil Engineering, University of Ottawa,

Ottawa, Ont., Canada.

Naumoski N., Saatcioglu M., and Amiri-Hormozaki K., 2004, Effects of Scaling of Earthquake

Excitations on the Dynamic Response of Reinforced Concrete Frame Buildings, 13th

World Conference on Earthquake Engineering, Vancouver, B.C., Canada, Paper No.

2917.

NBCC 2010, National Building Code of Canada, 2005, Canadian Commission on Building and

Fire Codes, National Research Council of Canada, Ottawa

Nikolaou, A.S. (1998). A GIS Platform for Earthquake Risk Analysis, Ph.D.Thesis, State

University of New York at Buffalo

Papageorgiou, A., Halldorsson, B. and Dong, G. (2002). TARSCTH (Target Acceleration

Spectra Compatible Time Histories), Engineering Seismology Laboratory (ESL) at the

State University of New York at Buffalo.

PEER, 2012, Pacific Earthquake Engineering Research Center: NGA Database,

http://peer.berkeley.edu (cited October 2005).

Prakash V., Powell, Graham H, Campbell, Scott D., 1993, Drain-2DX: Static and Dynamic

analysis of Inelastic Plane Structure, Department of Civil Engineering, University of

California, Berkeley.

Priestley, M.J.N [2000].” Performance Based Seismic design” . Proceedings, 12 th WCEE,

Auckland , New Zealand

159
Reitherman, R. (2008). International Aspects Of the History of Earthquake Engineering Part I

February 12, Draft https://www.eeri.org/site/images/awards/reports/reithermanpart1.pdf

(cited December, 2012).

Sawada T, Hirao K, Yamamoto H, Tsujihara O. Relation between maximum amplitude ratio

and spectral parameters of earthquake ground motion. In: Proceedings of 10th world

conference of earthquake engineering. Madrid, Spain, vol. 2, 1992.

SEAOC, 1995. Performance-based seismic engineering, Vision 2000 Committee, Report

prepared by Structural Engineers Association of California, Sacramento, CA

SEAOC, 2007. “Development of System Factors”, Seismology Committee, The SEAOC Blue

Book: Seismic Design Recommendations, Structural Engineers Association of California,

Sacramento, CA. Accessible via the world wide web at: http://www.seaoc.org/bluebook

Shome, N., and Cornell, A. C. (1998). “Normalization and scaling accelerograms for nonlinear

structural analysis.” Proc. of the 6th U.S. National Conf. on Earthquake Engineering,

Earthquake Engineering Research Institute, Seattle.

Somerville., P., N. C. Smith, and R. W. Graves, 1997a. Modification of Empirical Strong

Ground Motion Attenuation Relations to Include the Amplitude and Duration Effects of

Rupture Directivity, Seism. Res.Lett. 68, no. 1, 199–221.

Somerville, P., Smith, N., Punyamurthula, S. and Sun, J., 1997b. Development of ground motion

time histories for phase 2 of the FEMA/SAC steel project. SAC Joint Venture, Report

No.SAC/BD-97/04, California.

Somerville, P. G. (1998). Emerging Art: Earthquake Ground Motion. Geotechnical Earthquake

Engineering and Soil Dynamics III, ASCE Geotechnical Special Publication No. 75, Vol.

1, 1-38.

160
Stewart, J.P., S.J. Chiou, J.D. Bray, R.W. Graves, P.G. Somerville, N.A. Abrahamson. (2001).

“Ground motion evaluation procedures for performance-based design,” PEER-2001/09,

UC Berkeley.

Tso WK, Zhu TJ, Heidebrecht AC. Engineering application of ground motion A/V ratio. Soil

Dynamics and Earthquake Engineering 1992; 11:133–44.

Tothong, P., and Luco, N., 2007. Probabilistic seismic demand analysis using advanced

groundmotion intensity measures, Earthquake Eng. Struct. Dyn. 36, 1837–1860.

Tothong, P., and Cornell, A. C. (2008). “Structural performance assessment under near-

sourcepulse-like ground motions using advanced ground motion intensity measures,” Eq.

Eng. And Str. Dyn. Vol. 37, No. 7, pp. 1013-1037.

Vidic, T., Fajfar, P., and Fischinger, M. (1994). “Consistent inelastic design spectra: Strength

and displacement.” Earthquake Eng. Struct. Dyn.,23(5), 507–521.

Yun S.Y., Hamburger R.O., Cornell C. A., and Foutch D.A., 2002, Seismic Performance

Evaluation for Steel Moment Frames, Journal of Structural Engineering, Vol. 128, No. 4,

pp. 534-545

Yousuf, M, 2006. “Md Yousuf, M.A.Sc. Seismic Performance of Steel Frame Buildings”,

M.A.Sc. thesis, Department of Building, Civil and Environmental Engineering,

Concordia University, Montreal, Canada.

Yousuf, M. and Bagchi, A., 2009, “Seismic Design and Performance Evaluation of Steel-Frame

Buildings Designed using NBCC-2005”, Canadian J of Civil Engineering, 36: 280-294.

Yousuf, M. and Bagchi, A., 2010, “Seismic Performance of a Twenty Storey Steel Frame

Building in Canada”, J of Structural Design of Tall and Special Building, 19: 901-921.

161
Zeng, Y., J. Anderson, and G. Yu (1994). A composite source model forcomputing realistic

synthetic strong ground motions, Geophys. Res.Lett. 21, 725–728.

162