See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/283684633

Evaluating the spread plasticity model of IDARC for inelastic analysis of

reinforced concrete frames

Article in Structural Engineering & Mechanics · October 2015

DOI: 10.12989/sem.2015.56.2.169

CITATIONS READS

11 807

2 authors:

Mehdi Izadpanah Ali-Reza Habibi

Kermanshah University of Technology Shahed University
16 PUBLICATIONS 68 CITATIONS 43 PUBLICATIONS 220 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Evaluation of a simple and effective criterion to estimate damage to reinforced concrete moment resisting frames View project

Drift Based Damage Functions for Reinforced Concrete Moment Resisting Frames View project

All content following this page was uploaded by Ali-Reza Habibi on 30 June 2016.

The user has requested enhancement of the downloaded file.

Structural Engineering and Mechanics, Vol. 56, No. 2 (2015) 169-188
DOI: http://dx.doi.org/10.12989/sem.2015.56.2.169 169

Evaluating the spread plasticity model of IDARC for inelastic

analysis of reinforced concrete frames
Mehdi Izadpanaha and AliReza Habibi
Department of Civil Engineering, University of Kurdistan, Sanandaj, Iran

(Received April 5, 2015, Revised October 2, 2015, Accepted October 5, 2015)

Abstract. There are two types of nonlinear analysis methods for building frameworks depending on the
method of modeling the plastification of members including lumped plasticity and distributed plasticity. The
lumped plasticity method assumes that plasticity is concentrated at a zero-length plastic hinge section at the
ends of the elements. The distributed plasticity method discretizes the structural members into many line
segments, and further subdivides the cross-section of each segment into a number of finite elements. When a
reinforced concrete member experiences inelastic deformations, cracks tend to spread form the joint
interface resulting in a curvature distribution. The program IDARC includes a spread plasticity formulation
to capture the variation of the section flexibility, and combine them to determine the element stiffness
matrix. In this formulation, the flexibility distribution in the structural elements is assumed to be the linear.
The main objective of this study is to evaluate the accuracy of linear flexibility distribution assumed in the
spread inelasticity model. For this purpose, nonlinear analysis of two reinforced concrete frames is carried
out and the linear flexibility models used in the elements are compared with the real ones. It is shown that
the linear flexibility distribution is incorrect assumption in cases of significant gravity load effects and can be
lead to incorrect nonlinear responses in some situations.
Keywords: lumped plasticity; distributed plasticity; reinforced concrete; flexibility distribution; gravity
load effects

1. Introduction

The plasticity models are generally divided in two categories: lumped plasticity and spread
plasticity. In the former, there are some proposed models such as “two component model”, “one
component model”. The two component model is one of the first models proposed by Clough and
Johnston (1966) and consists of a linear elastic member in parallel with an elastic perfectly plastic
member. The most important lack of this model is that it cannot consider stiffness degradation.
Aoyama and Sugano (1968) extended this model. In their proposed model, each element is divided
into four parallel elements which consists of an elastic member and three elastoplastic members in
parallel. The model is able to consider different cracking and yielding characteristics at the two
critical end sections. Giberson (1967, 1969) suggested a one-component model. This model

Corresponding author, Associate Professor, E-mail: [email protected]

a
Ph.D. Student, E-mail: [email protected]

Copyright © 2015 Techno-Press, Ltd.

http://www.techno-press.org/?journal=sem&subpage=8 ISSN: 1225-4568 (Print), 1598-6217 (Online)
170 Mehdi Izadpanah and AliReza Habibi

consists of an elastic element with one nonlinear rotational spring at each end. The location of the
springs is considered in places that yielding is expected. The inelastic deformation of each element
is condensed into these rotational end springs. A major advantage of the model is that inelastic
member-end deformation depends solely on the moment acting at the end so that any moment-
rotation hysteretic model can be assigned to the spring. The weak spot of this model is ignoring the
curvature distribution effects on the member-end rotation. The above mentioned model was
amended by changing the location of the plastic hinges at the ends of the members to consider the
effects of rigid end zones (Al-Haddad and Wight 1986).The performance of the one-component
model is expected to be reasonably good for a relatively low-rise frame structure in which the
inflection point of a column locates reasonably close to mid-height (Otani 1980). Giberson (1967)
compared the one component model with the two-component model and showed that the one
component model was more proper than the two-component model. Kunnath and Reinhorn (1989)
proposed a concentrate plasticity model that was used in IDARC2D (Park et al. 1987) as an option
for concentrated plasticity.
Although simplicity and computability are two advantages of the above listed lump plasticity
models, due to their intrinsic zero-length plastic-zone assumption, they does not accurately
represent the distribution of plasticity within individual members of the frame. To overcome this
problem, discrete element models were proposed. In these models, the member can be subdivided
into short line segments along the length, with each short segment assigned a nonlinear hysteretic
characteristic. Each short segment is allocated a nonlinear hysteric stiffness characteristic. The
nonlinear stiffness can be assigned within a segment, or at the connection of two adjacent
segments. Wen and Janssen (1965) presented their models in this category. They introduced a
multi-spring model for dynamic analysis of a plane frame. Powell (l975) put forward a degrading
stiffness hysteresis model. In this model, shorter segments were recommended in a region of high
moment, and longer segments in a low-moment region. These discrete element models are more
accurate but they require more computational effort than other plasticity models. The latter model
used for nonlinear behavior is the spread plasticity model. The spread plasticity models provide a
more general framework for nonlinear structural analysis.
Against to the above mentioned discrete plasticity models, there are some continues models
developed based on prescribed distribution pattern of flexural flexibility along the length of
member. The parabolic-inflection distribution (Takizawa 1973) and linear-inflection distribution
(Park et al. 1987) are in this category. In parabolic-inflection model, an elastic flexibility at the
infection point is taken into account. This is an interesting concept for analyzing an inelastic
member. However, the parabolic-inflection flexibility distribution may not describe the actual
concentration of deformation at critical sections due to flexural yielding and deformation
attributable to slippage of longitudinal reinforcement within a beam-column connection
(Otani1980). The linear-inflection proposed by Park et al. (1987) was introduced in the original
version of IDARC2D developed by Reinhorn et al. (2009). Although the parabolic-inflection and
linear-inflection models can be efficient for some members, one of their defects is that they are
dependent on location of inflection point. To solve this problem and improve the plasticity models,
two spread plasticity models based on a linear flexibility distribution and a uniform flexibility
distribution were proposed that in these two models, the flexibility varies only in inelastic zones
while the rest of the member is elastic with constant flexibility (Kunnath and Reinhorn 1989).
Recently, Hajjar et al. (1998) put forward a distributed plasticity model for cyclic analysis of
concrete-filled steel tube beam-columns and composite frames. They presented the constitutive
formulation and cyclic analysis capability of a three-dimensional fiber-based distributed plasticity
 Evaluating the spread plasticity model of IDARC for inelastic analysis of reinforced concrete frames 171

Fig. 1 Deformation of reinforced concrete beam under gravity and earthquake loads (Otani1980)

finite element for square or rectangular concrete-filled steel tube (CFT) beam-columns. They also
used a distributed plasticity model for concrete filled steel tube beam-columns with inter layer slip
(Hajjar et al. 1998). Kim and Kurama (2008) used the spread plasticity model to reflect flexural
nonlinearity. The column and beam members were modeled using nonlinear beam-column
elements. Alva and de Cresce El Debs (2010) applied a lumped dissipation model in nonlinear
analysis of reinforced concrete structures. They considered the dissipation of energy of the
reinforced concrete members as a consequence of concrete damage and steel reinforcement
plasticity. As a simplification, it was supposed that the energy dissipation was restricted to plastic
hinges at the ends of the member, while the rest of the member remains elastic. He and Zhong
(2012) used the fiber section model to derive the nonlinear relation of section deformations and
internal forces and their interaction. Birely et al. (2012) presented a model to simulate the
nonlinear response of planar reinforced-concrete frames including all sources of flexibility. They
modeled nonlinearity by introducing a dual-hinge lumped-plasticity beam element comprised two
rotational springs in series; one spring simulates beam flexural response and one spring simulates
joint response. Roh et al. (2012) proposed a power spread plasticity model for inelastic analysis of
reinforced concrete structures and compared their plasticity model with linear plasticity model
used in IDARC2D. Kucukleret et al. (2014) extended their stiffness reduction, to capture fully the
detrimental influence of the spread of plasticity, residual stresses and geometrical imperfections on
the capacity of columns and beam-columns. Nguyen and Kim (2014) presented a displacement-
based finite element procedure for second-order spread-of-plasticity analysis of plane steel frames
with nonlinear beam-to-column connections under dynamic and seismic loadings.
Since the assumed flexibility distribution pattern in the IDARC is violated when stress levels
due to initial loads are relatively large, the main objective of this study is to evaluate the accuracy
of this model for nonlinear analysis of reinforced concrete structures. To do so, the real flexibility
of several structural elements with different load levels is determined and results are compared
with the assumed plasticity model.

2. Reinforced concrete plasticity model

As it was mentioned, the proposed plasticity models are divided to two main categories: lump
plasticity and spread plasticity. Now the main question is that which one is suitable for reinforced
concrete members? Inelastic deformation of reinforced concrete structural elements does not
172 Mehdi Izadpanah and AliReza Habibi

concentrate in a place and rather spread along the member as shown in Fig. 1. So it seems that the
proposed lumped plasticity models cannot simulate the real distribution of stiffness within a
reinforced concrete member.
Another choice in plasticity models is “discrete element method”. In this model, the member is
subdivided into short line segment with assigned a nonlinear hysteretic characteristic for each
short segment. The nonlinear stiffness can be assigned within a segment, or at the connection of
two adjacent segments as shown in Fig. 2.
The accuracy of this model is enhanced when the number of segments rises; although the
computational effort increases too. Choosing shorter segments should be based on moment
distribution along the member and places with high moment but this selection is difficult when the
moments are different at various load steps during nonlinear analysis. The other plasticity models
are “spread plasticity models”. In this regard, one of the most popular models is the linear
plasticity model used in IDARC2D (Reinhorn et al. 2009) (Fig. 3). This model has been used for
inelastic analysis of reinforced concrete structures by many researchers. In fact, the main problem
of these kinds of models (spread plasticity models) is that the flexibility distribution along the
element is assumed to be merely based on lateral load apart from the gravity load effects. This
assumption can lead to incorrect results in cases of significant gravity load moments.

Fig. 2 Discrete element model (a) multi-spring model (Birely et al. 2012) (b) multi-section model (Wen
and Janssen1965)

Fig. 3 Linear flexibility distribution model

 Evaluating the spread plasticity model of IDARC for inelastic analysis of reinforced concrete frames 173

Fig. 4 Tri-linear moment curvature curve

3. Moment curvature curve

The moment-curvature relation of every RC structural element has a definitive effect on the
behavior of the structure. The flexural stiffness of each member depends on moment curvature
relation directly. So in this study, the tri-linear moment curvature relation, as shown in Fig. 4, is
used to express the nonlinear behavior of reinforced concrete sections.
As it is evidence in the Fig. 4, the flexural stiffness can has three values based on the amount of
bending moments. The moment curvature relation of a structural element highly depends on its
cross-section. In this study, the column sections are limited to rectangle and that of beams can
assume rectangle, T or L shaped. The moment-curvature relations used in the present study are
similar to those utilized by Habibi and Moharrami (2010). The flexibility of each branch in the
Fig. 4 can be specified as follows
-1
1 M 
=  cr  (1a)
EI0  φ cr 

-1
1  M y - M cr 
=  (1b)
EIcr  φ y - φ cr 


-1
1  Mu - M y 
=  (1c)
EI y  φ u - φ y 
 

Where Eqs. (1a)-(1c) are used for zones of 1, 2 and 3 of M-φ curve, respectively. In Eq. (1) EI0,
EIcr and EIy are the flexural stiffness of different zones of moment curvature curve in Fig. 4 Mcr,
My and Mu are the cracking, yielding and ultimate moments; and φcr, φy and φu are corresponding
curvatures.
174 Mehdi Izadpanah and AliReza Habibi

Fig. 5 The considered member to extract formulation

4. Real flexibility distribution for reinforced concrete members

Since sections along a reinforced concrete element exhibit different flexibility characteristics
depending on the degree of inelasticity, in this study, the flexibility distribution assumed in the
IDARC program (Fig. 3) is evaluated. For this purpose, real flexibility distributions of elements
are determined and compared with those resulting from the IDARC. Real distribution is
determined considering uniform distributed load that is the most probable kind of gravity loads in
the building frames. In Fig. 5, the member under subjected loads and moments are taken into
account to develop the formulation.
For the member shown in Fig. 5, the amount of moment in each section can be calculated by
Eq. (2).

M(x) = V1x + M1 - 0.5Wx 2 (2)

Where, M(𝑥) is the moment in the section at a distance x from the left end. M1 is the bending
moment at the left end of the member. W is the amount of uniform distributed gravity load. V1 is
the shear force at the left end of the member and is determined according to the following equation

V1 = 0.5WL -
 M1 + M 2  (3)
L
Where, M2 is the bending moment at the right end of the member. L is the length of the
member.Now, to obtain the real plasticity of member, it is necessary that the flexibility of each part
of member be determined. To do so, by considering moment curvature relations in section (3) and
solving the following equations, different parts of member with various behaviors are specified.
(a) Locations of cracking moments in the length of the member
2 =M
V1x cr+ + M1 - 0.5Wx cr+ cr+ (4)

2 =M
V1x cr- + M1 - 0.5Wx cr- cr- (5)

(b) Locations of yielding moments in the length of the member

V1x y+ + M1 - 0.5Wx 2y+ = M y+ (6)

V1x y- + M1 - 0.5Wx 2y- = M y- (7)

 Evaluating the spread plasticity model of IDARC for inelastic analysis of reinforced concrete frames 175

Fig. 6 The process for determining real flexural flexibility

(c) Locations of zero moments in the length of the member

V1x 0 + M1 - 0.5Wx 02 = 0 (8)

Where, xcr+ and xcr− are the locations of positive and negative cracking moments (Mcr+, Mcr-), xy+
and xy- are the locations of positive and negative yielding moments (My+, My-), respectively. x0 is the
location of zero moments. It should be noted that each equation from 4 to 8 can have zero, one or
two roots.
176 Mehdi Izadpanah and AliReza Habibi

Fig. 7 Distribution of the flexural flexibility (a) step nth (b) step (n+1)th with neglecting the
previous step (c) step (n+1)th with considering the previous step (d) final distribution of
flexibility in which parts beside with same flexibility are assembled one

After determining the cracking and yielding points, the structural element is divided to several
parts. Flexibility of each part depends on the moments in two ends of it. To determine the real
flexural flexibility of each subdivided part, the following steps need to be performed:
1. Determining the number of zero moments in the moment diagram arranging from 0 to 2.
This step specifies the general shape of moment diagram and is used to assign flexibility of each
part.
2. Solving the equations 4 and 6 to assign the locations of positive cracking and yielding
moments.
3. Solving the equations 5 and 7 to assign the locations of negative cracking and yielding
moments.
4. Specifying the points determined in the two previous steps on the length of member and
delimiting the length of the member between two successive points as a part.
5. Determining the flexural flexibility for internal subdivided parts based on the amount of
moments in two ends. For the first part, the flexibility depends on M1 and the moments in its right
end; for the last part, the flexibility is related to M2 and the moments in its left end. For example, if
Mcr+≤M1≤My+ and the first limiting point corresponds with positive yielding moment and the
1
number of zero points is 1, the flexural flexibility would be . To better explain the core steps
EI cr
for determining the real flexural flexibility of each subdivided part, a flowchart is presented in Fig.
6. Since several different states can be occurred, to take all of them into account, a computer
program called Real Flexibility Model (RFM) was prepared. In this program, all various
conditions for flexural flexibility of a part of the member that can experience based on the
moments of its two ends are forecast.
It should be noted that in step by step nonlinear analysis of reinforced concrete structural
elements, the flexural flexibility in each step completely depends on the previous steps. Therefore,
in this study two main assumptions are considered: 1. The cracked part of member does not
change to a part with no crack 2. The yielded part of member does not change to a part with no
crack and no yield. To consider the above mentioned matter, the current state of each member is
compared with its previous state and then the flexural flexibility is obtained. Given that the
 Evaluating the spread plasticity model of IDARC for inelastic analysis of reinforced concrete frames 177

flexibility distributions of a member in two consequence steps of its analysis be like Figs. 7(a) and
7(b) (independent of previous steps); therefore, considering previous steps, the (n+1)th distributed
flexibility will be similar to Fig. 7(c). The last but not the least (Fig. 7(d)) is the final flexibility
distribution in which the parts with the same flexibility are taken as one part.

5. Evaluating the linear flexibility model of IDARC

Part In this section, the accuracy of the linear plasticity model used by the program IDARC is
assessed. In this regard, two building frames that have been previously studied by Habibi (2007)
and Habibi and Moharrami (2010) are chosen. Pushover analysis is carried out on these frames by
applying monotonically increasing lateral loads along with constant gravity loads. At each load
step, the base shear increment is applied to the structure with a predefined profile over the height
of the structure. The incremental lateral load vector can be computed as
 cv,1 
 
 cv,2 
ΔPE = ΔV b .   = ΔV b .Cv (9)
 M 
cv,ns 
 
Where ΔVb is the incremental base shear and Cv is the vector of lateral load distribution factors Cv,s
(s=1,…, number of stories), which is determined from FEMA273
ns
c v,s = Ws Hsk k
 Ws Hs ,s = 1,..., ns (10)
s=1
Where Ws is the portion of the building seismic weight at story level s; Hs is the vertical distance
from base of the building to story level s; ns is the number of stories; and k is a parameter that has
been recommended by FEMA273 as follows
1 T 0.5

k = 0.5T + 0.75 0.5  T  2.5 (11)
2
 T 2.5

Where T is the fundamental period of the building. This analysis is done by IDARC software
(Valles et al. 1996). Since the main objective of this research is evaluating the accuracy of linear
flexibility model (LFM), it is necessary that all required input parameters for each member in
IDARC2D and the proposed method of this study be same. To do so, after pushover analysis by
IDARC, the end moments of each member are obtained and are used to calculate the real
flexibility by the proposed method. To consider various performance levels, the states of frames
are taken into account regarding to overall drifts 0.5,1,2 and 4 percent in pushover analysis. The
former frame is a three-story, two-bay planar asymmetric frame of Fig. 8. The concrete is assumed
to have a cylinder strength of 20 Mpa, a modulus of rupture of 2.82 Mpa, a modulus of elasticity
of 22360 Mpa, a strain of 0.002 at maximum strength and an ultimate strain of 0.003. The steel
has a yield strength of 300 Mpa and a modulus of elasticity of 200,000 Mpa. A uniformly
distributed gravity load of 20 KN/m is applied on the beams of each story. Reinforcements have
the cover to the steel centroid of 50 mm. It is assumed that columns and beams have rectangular
cross sections Habibi (2007).
178 Mehdi Izadpanah and AliReza Habibi

Fig. 8 The three-story frame

Table 1 The moment-curvature properties of members chosen in example 1

𝑀𝑐𝑟 (+) 𝑀𝑐𝑟 (-) 𝑀𝑦 (+) 𝑀𝑦 (-) EI(+) EI(-) 𝐸𝐼𝑐𝑟 (+) 𝐸𝐼𝑐𝑟 (-) 𝐸𝐼𝑦 (+) 𝐸𝐼𝑦 (-)
Member
(KN.m) (KN.m) (KN.m) (KN.m) (KN. m2 ) (KN. m2 ) (KN. m2 ) (KN. m2 ) (KN. m2 ) (KN. m2 )
Column
C12 28.4 28.4 80.7 80.7 17795 17795 5013 5013 135.4 135.4
C33 17 17 55 55 17795 17795 4238 4238 200 200
Beam
B12 15 15 49 49 17795 17795 4100 3993 204 204
B21 15 15 49 49 17795 17795 4100 3993 204 204

Fig. 9 LFM and RFM for B12 (a) overall drift is 0.5% (b) overall drift is 1% (c) overall drift is 2% (d)
overall drift is 4%
 Evaluating the spread plasticity model of IDARC for inelastic analysis of reinforced concrete frames 179

Fig. 10 LFM and RFM for B21 (a) overall drift is 0.5% (b) overall drift is 1% (c) overall drift is 2% (d)
overall drift is 4%

As shown, numbering of elements in each story is from left to right. Therefore, the name of
each member consists of two indices (like Cij for columns and Bij for beams) the former shows the
story of the member and the latter corresponds to its place in each story. To cover varied situations
in the three story frame, two beams and two columns are chosen (C23, C33, B12, B21) The moment-
curvature properties of members chosen are presented in Table 1.
The linear and real flexibility distributions are shown in Figs. 9-12. In the Fig. 13, all the
bending moment curves resulting from pushover analysis for each member in drift 4% are shown
to express the range of end moments, cracked and yielded points and shape of moment diagrams.
Since the number of curves is too many and marked points are not clear in the Fig. 13; to explain
more, the bending moment curves of the first member (B12) is exhibited for some moments
(approximately 5 percent of moment diagrams (Fig. 13(a))). In Fig. 13, asterisk points are related
to moments that are equal to positive or negative cracked moments. Diamond shapes show the
location of positive or negative cracked moments and the circles are taken into account as the
place of zero moments. The parameters of plasticity models are explained in Table 2. In the
mentioned Table, there are two amounts for linear plasticity model presenting the yield penetration
coefficients (Fig. 3). For the real flexibility model, there are some coefficients expressing the
transformation points. It should be noted that these mentioned points are calculated from the left
side and all these lengths are divided to the length of members.
180 Mehdi Izadpanah and AliReza Habibi

Fig. 11 LFM and RFM for C12 (a) overall drift is 0.5% (b) overall drift is 1% (c) overall drift is 2% (d)
overall drift is 4%

Fig. 12 LFM and RFM for the element C33 (a) overall drift is 0.5% (b) overall drift is 1% (c) overall drift
is 2% (d) overall drift is 4%
 Evaluating the spread plasticity model of IDARC for inelastic analysis of reinforced concrete frames 181

Fig. 13 The bending moment diagrams for the selected members in example 1 (a) the 5 percent of all
moment diagrams for B12. (b-e) all of moment diagrams for B12, B21, C12 and C33 respectively

As shown in Figs. 9-12, in all the members at various performance levels, the linear plasticity
model could not reflect the real flexibility of them. For B12, it is clear that the linear flexibility
model (LFM) is more flexible than the real one. This can lead to more rotation in the right side of
member which directly affects the performance levels. It seems that for the right side of this
member, the power spread plasticity model proposed by Roh et al. (2012), may model the
flexibility more proper than LFM. As evident for B21, LFM in the end right side of the length of
member is stiffer than RFM. In the left side, the flexibility of LFM is lower than RFM in the part
near the end but, in the middle part, the flexibility is more. For the right side of this element, it
182 Mehdi Izadpanah and AliReza Habibi

seems that the power plasticity model with high power will be more accurate than LFM as long as
the length of the yielded part is restricted; but as shown, when the length of the yielded part
increases like the left side of B21 in drift 4%, the power plasticity model with low power can model
the flexibility well. The results presented for beam elements in the first example are completely
different in linear and real flexibility models. Since LFM does not take the gravity loads into
account, this model can make egregious errors. What is manifest for beam elements is that
although power spread plasticity can increase the accuracy of flexibility, the main problem is
neglecting gravity loads; therefore, for beams, the places of cracked and yielded points are
different from real locations. Also, the power in power spread plasticity simply depends on the end
moments and cannot consider the effects of gravity loads in the middle parts of the beam elements.
The flexibility models for C12 are different as well. Although for the overall drift 0.5%, LFM is
stiffer than RFM, for the other performance levels it is more flexible except for the yielded part
narrowed in the end of left side. For the last member C33, the LFM is stiffer than RFM in the two
prime performance levels. Unlike first and second performance levels, the linear flexibility
distribution for the both of last performance levels is more flexible than real flexibility distribution
except in the small end yielded part of right side that is stiffer. As shown, the LFM even the
column members that are just subjected to lateral loads, cannot reflect the real flexibility of them.
It seems because of lack of gravity loads on the column elements, the power flexibility model can
model the flexibility more accurately.
In this example, some performance levels were assessed and the accuracy of them was
challenged. It should be noted that the effect of errors occurred is noticeable in some analyses like
dynamic, cyclic and pushover methods because the stiffness or softness matrix of a structure
depends on the flexibility models used in them. The errors of the flexibility models accumulated in
the steps of the analysis are not negligible. This matter will be described more in the next example.
The latter example is a ten-story, two-bay planar frame of Fig. 14. For this example, the
concrete is assumed to have a cylinder strength of 30 Mpa, a modulus of rupture of 3.45 Mpa, a
modulus of elasticity of 27,400 Mpa, a strain of 0.002 at maximum strength and an ultimate strain
of 0.004. The other material properties are same with the first example. A uniformly distributed
gravity load of 20 KN/m is applied on the beams of each story. Reinforcements have the cover to
the steel centroid of 50 mm. It is assumed that columns and beams have rectangular cross sections
(Habibi and Moharrami 2010).

Table 2 Yield penetrations coefficients for LFM and the transformation points for RFM in example 1
Overall Linear flexibility model Real flexibility model
drift % 𝐶12 𝐶33 𝐵12 𝐵21 𝐶12 𝐶33 𝐵12 𝐵21
0.08,0.16,
0.03, 0.34, 0.036, 0.018,0.56,
0.5 0.46,0 0.5, 0.5 0.47 0.77,
0.42 0.38 0.58 0.78, 0.986
0.86,0.998
0.08,0.14,
0.04, 0.35, 0.037, 0.11,0.56, 0.77,
1 0.54,0 0.5, 0.5 0.025, 0.54 0.77,
0.58 0.38 0.41 0.97
0.86,0.99
0.04, 0.36, 0.037,0.28, 0.08,0.10, 0.772, 0.19,0.56, 0.76,
2 0.54,0 0.5, 0.5 0.06, 0.54
0.71 0.38 0.97 0.844,0.97 0.94
0.08,0.082,
0.04, 0.37, 0.08,0.54, 0.037,0.24, 0.22,0.56, 0.76,
4 0.54,0.03 0.5, 0.5 0.772,
0.75 0.38 0.96 0.92 0.94
0.833,0.96
 Evaluating the spread plasticity model of IDARC for inelastic analysis of reinforced concrete frames 183

Fig. 14 The ten-story frame

Table 3 The moment-curvature properties of members chosen in example 2

𝑴𝒄𝒓 (+) 𝑴𝒄𝒓 (-) 𝑴𝒚 (+) 𝑴𝒚 (-) EI(+) EI(-) 𝑬𝑰𝒄𝒓 (+) 𝑬𝑰𝒄𝒓 (-) 𝑬𝑰𝒚 (+) 𝑬𝑰𝒚 (-)
Member
(KN.m) (KN.m) (KN.m) (KN.m) (KN. m2 ) (KN. m2 ) (KN. m2 ) (KN. m2 ) (KN. m2 ) (KN. m2 )
Column
C21 91.4 91.4 719 719 339230 339230 200771 200771 2825 2825
C72 49.1 49.1 240 240 45705 45705 24820 24820 363 363
Beam
B32 87 87 474 474 217340 217340 97943 99473 1618 1618
B91 13.7 13.7 88.3 88.3 17207 17207 7642 7809 284 284

Table 4 Yield penetrations coefficients for LFM and the transformation points for RFM in example 2
Overall Linear flexibility model Real flexibility model
drift % 𝑪𝟐𝟏 𝑪𝟕𝟐 𝑩𝟑𝟐 𝑩𝟗𝟏 𝑪𝟐𝟏 𝑪𝟕𝟐 𝑩𝟑𝟐 𝑩𝟗𝟏
0.38,
0.5 0.129, 0 0.36, 0.32 0.26, 0.38 0.5, 0.5 0.126 0.35,0.67 0.998
0.726
0.4,
1 0.37, 0.13 0.40, 0.36 0.31, 0.42 0.5, 0.5 0.36, 0.87 0.40,0.63 0.98
0.66
0.005, 0.43, 0.42,
2 0.44, 0.20 0.43, 0.362 0.34, 0.42 0.5, 0.5 0.43, 0.79 0.96
0.63 0.65
0.015, 0.43, 0.426,
4 0.48, 0.25 0.43, 0.363 0.35, 0.43 0.5, 0.5 0.46, 0.74 0.95
0.635 0.64

In this frame, four members including two beams and two columns just like the first frame (B32,
B91, C21, C72) are chosen. The moment-curvature properties of the members are expressed in Table
3.
The plasticity models are presented in Figs. 15-18 and the related plasticity parameters are
described in Table 4.
184 Mehdi Izadpanah and AliReza Habibi

Fig. 15 LFM and RFM for B32 (a) overall drift is 0.5% (b) overall drift is 1% (c) overall drift is 2% (d)
overall drift is 4%

Fig. 16 LFM and RFM for B91 (a) overall drift is 0.5% (b) overall drift is 1% (c) overall drift is 2% (d)
overall drift is 4%
 Evaluating the spread plasticity model of IDARC for inelastic analysis of reinforced concrete frames 185

Fig. 17 LFM and RFM for C72 (a) overall drift is 0.5% (b) overall drift is 1% (c) overall drift is 2% (d)
overall drift is 4%

Fig. 18 LFM and RFM for C21 (a) overall drift is 0.5% (b) overall drift is 1% (c) overall drift is 2% (d)
overall drift is 4%
186 Mehdi Izadpanah and AliReza Habibi

Fig. 19 LFM and RFM for B91 (a) overall drift is 0.05% (b) overall drift is 0.1% (c) overall drift is 0.2%

As can be seen, in the second example, also, the highest difference between the real and linear
flexibility models is related to the beam elements. Neglecting the gravity loads in LFM brings
about errors in the location of the cracked and yielded moments (this problem exists in other
models like the power plasticity model disregarding the effect of gravity loads). For B32, the linear
flexibility model is stiffer than real one. The gap between the models for B91 is very large and the
LFM is more flexible than RFM. Occurring yield in the left side of C72 leads to a large contrast
between the two models. For the last member C21, in all the overall drifts, LFM is more flexible
than RFM. As noted, the accumulated errors of the flexibility models severely affect the outcomes
of analysis. For more explanation, several steps of pushover analysis of B91 are presented. These
steps are regarding the 0.05, 0.1 and 0.2 percent overall drifts. These drifts are chosen to show the
influence of neglecting the gravity loads (Fig. 19).
As shown, in the 0.05, 0.1 and 0.2 percent overall drifts, middle part of the beam is cracked.
The gap between LPM and RPM is very large. None of the plasticity models when ignoring the
gravity loads cannot correctly simulate the flexibility of beams in most steps of analyses. This
occurs in the steps in which the effect of gravity loads overcomes the lateral moments. Although in
some analyses like pushover analysis, the effect of lateral loads dominate the gravity loads by
increasing the overall drift, the stiffness matrix depends on the plasticity models in all steps of
analysis; therefore, the accumulated errors in initial steps of analysis can lead to big errors in the
next steps.
It should be noted that despite improving the accuracy of some models like “multi segment
model” when increasing the number of segment, these models have some problems. In the
 Evaluating the spread plasticity model of IDARC for inelastic analysis of reinforced concrete frames 187

mentioned model, the more accurate flexibility is obtained when shorter segments are chosen.
These shorter segments require more computational effort that is time consuming. Moreover, in
some steps of analyses, it is possible that the behavior of elements still is elastic or experience
crack in the small parts of elements and therefore it is not necessary that they be subdivided into
shorter segments.

6. Conclusions

In this study, the spread plasticity model of IDARC for reinforced concrete elements was
evaluated. For this purpose, first, a real spread plasticity model was developed that takes the
influence of gravity loads into account. In the proposed model, the flexibility of each member is
determined based on the locations of positive and negative yielded and cracked flexural moments
in current step of analysis and also by considering history of flexibility in previous steps. Then, the
real plasticity model was compared with the linear plasticity model used in IDARC that is one of
prevalent models for reinforced concrete elements. To do so, pushover analysis was carried out on
a 3-story frame and a 10-story frame. To cover wide range of behavior of the frames, the plasticity
models are compared in the 0.5%, 1%, 2% and 4% of overall drifts. The outcomes exhibit that the
linear plasticity model has glaring errors in modeling reinforced concrete elements and can model
the member stiffer or softer than real model. Some of these errors may be eliminated by using
other plasticity models like the power or the uniform spread plasticity. Meanwhile other ones like
the locations of cracked and yielded bending moments in beam elements subjected to gravity
loads, cannot be omitted by existing models in which the gravity load effects are ignored.

References

Al-Haddad, M.S. and Wight, J.K. (1986), “Feasibility and consequences of moving beam plastic hinging
zones for earthquake resistant design of R/C buildings”, Report No. UMCE 86-1, The University of
Michigan, Ann Arbor, Michigan.
Alva, G.M.S. and de Cresce El, A.L.H. (2010), “Application of lumped dissipation model in nonlinear
analysis of reinforced concrete structures”, Eng. Struct., 32(4), 974-81.
Aoyama, H. and Sugano, T. (1968), “A generalized inelastic analysis of reinforced concrete structures based
on the tests of members”, Recent researches of structural mechanics, Contribution in Honour of the 60th
Birthday of Professor Y. Tsuboi, Uno-Shoten, Tokyo, 15-30.
Birely, A.C., Lowes, L.N. and Lehman, D.E. (2012), “A model for the practical nonlinear analysis of
reinforced-concrete frames including joint flexibility”, Eng. Struct., 34(1), 455-65.
Clough, R.W. and Johnston, S.B. (1966), “Effect of stiffness degradation on earthquake ductility
requirements”, Transactions of Japan Earthquake Engineering Symposium, Tokyo.
FEMA273 (1997), “NEHRP Guideline for the seismic rehabilitation of buildings”, Federal Emergency
Management Agency, Washington, DC.
Giberson, M.F. (1967), “The response of nonlinear multi-story structures subjected to earthquake
excitation”, Earthquake Engineering Research Laboratory, California Institute of Technology, Pasadena,
CA, EERL Report.
Giberson, M.F. (1969), “Two nonlinear beams with definitions of ductility”, J. Struct. Div., 95(ST2), 137-57.
Habibi, A.R, (2007), “Optimal seismic performance-based design of 2D RC frameworks”, Ph.D. Thesis,
Tarbiat Modares University, Tehran, Iran.
Habibi, A.R. and Moharrami, H. (2010), “Nonlinear sensitivity analysis of reinforced concrete frames”,
 188 Mehdi Izadpanah and AliReza Habibi

Finite Elem. Anal. Des., 46, 571-584.

Hajjar, J.F., Molodan, A. and Schiller, P.H. (1998), “A distributed plasticity model for cyclic analysis of
concrete-filled steel tube beam-columns and composite frames”, Eng. Struct., 20(4-6), 398-412.
Hajjar, J.F., Schiller, P.H. and Molodan, A. (1998), “A distributed plasticity model for concrete filled steel
tube beam-columns with interlayer slip”, Eng. Struct., 20(8), 663-76.
He, R. and Zhong, H. (2012), “Large deflection elasto-plastic analysis of frames using the weak-form
quadrature element method”, Finite Elem. Anal. Des., 50, 125-133.
Kim, S.P. and Kurama, Y.C. (2008), “An alternative pushover analysis procedure to estimate seismic
displacement demands”, Eng. Struct., 30, 3793-3807.
Kucukler, M., Gardner, L. and Macorini, L. (2014), “A stiffness reduction method for the in-plane design of
structural steel elements”, Eng. Struct., 73, 72-84.
Kunnath, S.K. and Reinhorn, A.M. (1989), “Inelastic three-dimensional response analysis of reinforced
concrete structures subjected to seismic loads”, Technical Report No. NCEER-88-0041, University at
Buffalo, The State University of New York.
Nguyen, P.C. and Kim, S.E. (2014), “Distributed plasticity approach for time-history analysis of steel frames
including nonlinear connections”, J. Constr. Steel Res., 100, 36-49.
Otani S. (1980), “Nonlinear dynamic analysis of reinforced concrete building structures”, Can. J. Civil Eng.,
7(2), 333-44.
Park, Y.J., Reinhorn, A.M. and Kunnath, S.K. (1987), “IDARC: Inelastic damage analysis of reinforced
concrete frame-shear wall structures”, Technical Report No. NCEER- 87-0008, University at Buffalo, The
State University of New York.
Powell, G.H. (l975), “Supplement to computer program DRAIN-2D”, Supplement to report, DRAIN-2D
user's guide, University of California, Berkeley, CA.
Reinhorn, A.M., Roh, H., Sivaselvan, M., Kunnath, S.K., Valles, R.E., Madan, A., Li, C., Lobo, R. and Park,
Y.J. (2009), “IDARC 2D Version 7.0: A Program for the Inelastic Damage Analysis of Structures”,
MCEER Technical Report, MCEER-09-0006, University at Buffalo-the State University of New York.
Roh, H., Reinhorn, A.M. and Lee, J.S. (2012), “Powerspread plasticity model for inelastic analysis of
reinforced concrete structures”, Eng. Struct., 39, 148-161.
Takizawa, H. (1973), “Strong motion response analysis of reinforced concrete buildings”, J. Jpn. Nat.
Council Concrete, II(2), 10-21.
Valles, R.E., Reinhorn, A.M., Kunnath, S.K., Li, C. and Madan, A. (1996), “IDARC2D Version 4.0 – a
computer program for the inelastic drainage analysis of buildings”, Technical Report No. NCEER-96-
0010, University at Buffalo, The State University of New York.
Wen, R.K. and Janssen, J.G. (1965), “Dynamic analysis of elasto-inelastic frames”, Proceedings of 3rd
World Conference on Earthquake Engineering, 713-29.

CC

View publication stats