Comparison Between Pushover Analysis Techniques and Validation of The Simplified Modal Pushover Analysis

World Academy of Science, Engineering and Technology
International Journal of Urban and Civil Engineering

Vol:11, No:5, 2017
Comparison between Pushover Analysis Techniques

and Validation of the Simplified Modal Pushover
Analysis
N. F. Hanna, A. M. Haridy
 are based on the mode shapes of vibration of the building

Abstract—One of the main drawbacks of the Modal Pushover frame. The main difficulty of the MPA is that it requires non-
Analysis (MPA) is the need to perform nonlinear time-history linear time-history analysis of an equivalent single-degree-of-
analysis, which complicates the analysis method and time. A freedom system. This may not be practical in a design office,
simplified version of the MPA has been proposed based on the
as typical engineers are not familiar with this advanced type of
concept of the inelastic deformation ratio. Furthermore, the effect of
the higher modes of vibration is considered by assuming linearly- analysis. So, a simplified MPA (SMPA) method has been
elastic responses, which enables the use of standard elastic response developed to overcome this difficulty. This simplified method
International Science Index, Urban and Civil Engineering Vol:11, No:5, 2017 waset.org/Publication/10007610
spectrum analysis. In this thesis, the simplified MPA (SMPA) utilizes the concept of the deformation ratio which relates the
method is applied to determine the target global drift and the inter- inelastic deformations of SDF system to the corresponding
story drifts of steel frame building. The effect of the higher vibration elastic deformations that can be directly estimated using the
modes is considered within the framework of the SMPA. A
regular response spectrum analysis. Recent expressions for the
comprehensive survey about the inelastic deformation ratio is
presented. After that, a suitable expression from literature is selected deformation ratio will be presented and applied in the
for the inelastic deformation ratio and then implemented in the simplified POA.
SMPA. The estimated seismic demands using the SMPA, such as POA can be either force-controlled or displacement-
target drift, base shear, and the inter-story drifts, are compared with controlled. In force-controlled POA, the entire lateral load up
the seismic responses determined by applying the standard MPA. The to failure is known and hence applied along the height of the
accuracy of the estimated seismic demands is validated by comparing
frame building. In force-controlled POA, some numerical
with the results obtained by the nonlinear time-history analysis using
real earthquake records. problems that affect the accuracy of results may occur since
the target drifts may be associated with a very small positive
Keywords—Modal analysis, pushover analysis, seismic or even a negative lateral stiffness. On the other side,
performance, target displacement. displacement-controlled POA is based on using a known
lateral load pattern, but not the actual values of the loads till
I. INTRODUCTION failure as mentioned earlier. This pre-selected lateral load
S TANDARD pushover analysis (POA) is a simple pattern is using to push the building until a pre-defined failure
technique that has been effectively used to estimate the state or target drift state has been reached. At this state, the
seismic demands of frame buildings. It is a static non-linear actual lateral load can be determined from the static nonlinear
analysis method based on applying a selected lateral load analysis.
pattern, which represents the equivalent lateral forces due to One of the main steps in the POA is the construction of the
earthquake, along the height of the building frame, and the global capacity curve, which represents the nonlinear relation
building is pushed laterally until a pre-defined failure state is between the base shear and the global drift at the top of the
reached. The expected damage in the main structural members building. After applying the SMPA method, the performance
is modeled using plastic hinges placed at strategic locations state of the building due to the given earthquake can be
along the member. The main drawbacks of the standard POA identified, and the associated seismic responses are
are its inability to consider the effect of the higher modes of determined: target drift, base shear, interstory drifts, and
vibration and its inability to consider the changes in the lateral distribution of the plastic hinges. A good estimation of the
stiffness properties of the building as damage progresses building drifts is desirable for better judgment of the expected
during the pushover. The first drawback related to the higher damage in the building due to the earthquake. So, the drift
vibration modes effect is deemed more important in practice. results obtained from the SMPA are compared with both the
Several advanced methods have been proposed in literature results determined by the MPA and by the non-linear time-
to account for the effect of the higher modes of vibration. The history analysis using a set of real and artificial earthquake
most popular method is known by the Modal Pushover records. POA is becoming the preferred tool for seismic
Analysis (MPA). The lateral load patterns used in the MPA performance and evaluation of frame buildings, and several
modern seismic design codes, e.g. Eurocode 8, have included
provisions for performing static nonlinear analysis.
Professor Hanna, N.F, Faculty of Engineering and Material Sciences, German
University in Cairo, Cairo, Egypt Haridy, A.M, Faculty of Engineering and
Material Sciences, German University in Cairo, Cairo, Egypt
International Scholarly and Scientific Research & Innovation 11(5) 2017 687 scholar.waset.org/1307-6892/10007610
Vol:11, No:5, 2017
II. STANDARD PUSHOVER ANALYSIS the first three modes of vibration.

Aspects of Pushover Analysis—To appraise the seismic
performance of existing building and newly designed
buildings the static POA method is applied, which is a non-
linear static analysis method. For performance based seismic
design (PBSD) buildings are designed to perform according to
a specific performance objective based on its function and
type.
The POA method is considered to be a powerful tool, even
Fig. 1 First, second, and third modes of vibration
though it still has no strict theoretical basis. As for the POA,
there are three main assumptions that it relies on; the first Limitations—Taking in consideration that POA is a
mode of vibration is what controls the seismic response of the nonlinear static analysis, it needs numerous features of its
building, secondly through the elastic and inelastic response of dynamic counterpart which might be basic in certain analysis
the building the lateral load pattern remains constant, the third cases. Nonetheless, it delivers the engineers with a practical
assumption is the relativity between the single degree of option to the Non-Linear Time History Analysis (NLTHA)
freedom system to the multi degree of freedom system. and demonstrates standard seismic codes of practice with
What recent studies on POA have shown is that in case basic tools for enhancing the seismic design of structures.
where the first mode of vibration dominates the responses, In the following, the major limitations of the standard pushover
these assumptions can produce good evaluation of the seismic techniques are listed:
responses of building. In the other words, an approximate  It is simulated that the structural harm is a function of just
analysis method in which the building is subjected to the lateral deformation of the building. By this way, the
monotonically expanding later forces with an invariant earthquake duration impact neglecting duration impacts
distribution over the height of the building till a pre- and the cumulative energy dissipation demand is
determined or target displacement is come, is what is known neglected. Harm is a duty of both deformation and energy,
as the standard POA. For approximating the nonlinear relation and this was generally acknowledged. So, particularly for
between the base shear and the roof displacement of the non-ductile structures which display pinched hysteretic
building a nonlinear static analysis is required. A model for behavior, the applicability of POA is slightly simplistic.
the building is built in the POA. First, gravity loads are  Many researches recognized that the independent
implemented, afterwards it is applied the selected later load reactions between the structural capacity and the
pattern onward the building height. To approximately perform earthquake demand are suggested to be interconnected. In
the relative inertia actions developed throughout an addition, the division between the loading input and
earthquake, excitation is the main purpose and objective of structural reaction is not regularly sufficient because the
this lateral load pattern. Till the displacement at the top of the nonlinear structural behavior is dependent on the load
building achieves a certain value or a breakdown mechanism path.
evolves for the building, the later strengths are raised.  It was not accounted that progressive determination of the
Subsequently, the capacity of the building can be tested before stiffness of the building because of the yield and harm
and after yielding. evolving in the building prompts to period elongation.
Elastic Vibration Properties—For demonstrating the This is because of the invariant lateral load pattern
reinforced concrete (RC) building frames, SAP200 is utilized utilized, which disregards the redistribution of the inertia
as a standard structural analysis program. In order to represent forces as yielding rules the inelastic behavior of the
the beams and columns of the RC frame, 2-D frame building.
components are used and where two nodes situated at the ends  The strain energy of the structures throughout a
of the element portray each frame elements. Every node has monotonic static loading is what is only concentrated on.
three degrees of freedom; one rotational degree of freedom for Different sources of energy that are connected with the
in-plane bending, and two translational degrees of freedom dynamic parts of forces are dismissed for example as the
through the vertical and horizontal directions. For deciding the viscous damping energy and the kinetic energy.
vibration properties of the RC building frames, these 2D  The impacts of the higher-modes on the seismic reactions
frames are utilized like the time of vibration, the modal and demands of the structures are not represented. If there
participation mass ratios, the modal participating factors, and should be an occurrence of mid-to high structures then the
the mode shapes of vibration. contribution of the higher-modes might be of great
The satisfactory number of vibration modes required for the impact.
examinations of the frames must guarantee that no less than When accomplishing a POA, it was tried by [1] to
90% of the total seismic weight is taking in consideration in recognize some potential pitfalls. Therefore, ten significant
the analysis; this was suggested by numerous latest seismic conditions that are supposed to be taken into consideration
regulations. In general, to acquire the dynamic qualities of before the POA were outlined; the following are the most vital
seismic analysis which mode shapes, participation factor, mass ones:
participation ratio and time period are involved. Fig. 1 shows
Vol:11, No:5, 2017
 The gravity loads and the shear failure mechanisms are as follow:
not supposed to be neglected.
1. Vibration Analysis
 Before the building is pushed, the performance purpose is
supposed to be identified  Compute vibration properties of building using Sap2000
by choosing the number of modes required for the
 The underestimation of the loading shape function is not
effective modal weight to vibrate, which is given to be
supposed to happen
more than 90% according to [7].
 The P-Delta impacts should be included, because if not
 Get natural frequencies ωn or natural periods Tn
then the outcomes gained by the pushover could be non-
conservative.
 As three-dimensional structures may require more than a ω= (1)
one planar POA, then the pushover loading should not be
mistaken for the actual earthquake.  Get modes Φn, for linearly elastic vibration of the
A few progressed pushover techniques have been created to building.
manage the limitations of the standard POA, due to the 2. Application of Gravity Load
complexity of the nonlinear dynamic analysis contrasted with
the simplicity and practically of the standard POA. The theory  Assign Gravity load (Dead and Live). Using load
that the seismic reaction of the building is monopolized by the combinations:
primary mode of vibration is one of these limitations. In the

*U=1.4 D.L + 1.6 L.L (2)
case of low-rise structures, this presumption is real and
rational, and great results are given by the standard pushover
*U=1.12 D.L + ΩL.L + E (3)
procedures. While modes of vibration higher than the
fundamental mode have a significant impact that cannot be *U= 0.8 (1.4 D.L + 1.6 L.L + 1.6 W.L) (4)
ignored in the case of mid-rise and high-rise structure.
The adequate number of modes that is supposed to be 3. Selection of Lateral Load Pattern
incorporated to accomplish acceptable exactness to be all  Get mass (seismic weight/gravity) matrix using one of
modes with cumulative seismic mass more than 90% of the these methods:
total seismic mass of the building is characterized by I. Modal load in Sap2000
numerous updated seismic regulations like [2], [3]. Few OR
pushover procedures have been created for the objective of II. Hand Calculations as follow for the rigid diaphragm
incorporation the impact of the higher modes of vibrations.
0 0 0
III. ADVANCED PUSHOVER TECHNIQUES 0 0 0
[M] where m1=W1/G, n=floor
Modal Pushover Analysis— There are some limitations for 0 0 ⋱ 0
the standards POA, which gave reasons for a more advanced 0 0 0
pushover technique. This technique is considered to be more
 Get Mode shape vector using Sap2000
conservative when compared to the standard techniques either
in its concept of the idea or in its application. What [4], [5] did Φ
is that, without losing the simplicity of the standard POA, the Φ
limitation of the unique vibration mode that commanded the {Φj}=
⋮
behavior of the structure was considered. In order to reach this Φ
objective, the known and familiar modal analysis approach
was joined with the POA for evolving the MPA which can  Lateral Load Pattern {fj}=[M] {Φj} (5)
generally be described as a more progressed pushover
procedure. According to the limitations and shortcomings of m Φ
the mentioned Standard POA, a MPA developed by [6]. MPA m Φ
{fj}=
main upgrades are: ⋮
1. Taking the effect of higher modes into consideration. m Φ
2. Account progressive drop in stiffness as a result of
4. Apply Selected Lateral Load Pattern in the Building
yielding response of structure leading to period
elongation.  Assign the values for the concentrated lateral load pattern
The involvement of the higher modes will affect the seismic forces per floor using the previous vector {fj}.
response especially in the mid-rise and high-rise buildings.  Develop the pushover curve for the whole building, Base
Using nonlinear dynamic analysis is a procedure in the MPA Shear-Roof Displacement Vby-∆ using nonlinear static
to consider the effects of higher modes. analysis in Sap2000.
A. Steps of MPA 5. Transform the Capacity Curve of SDOF System to

Bilinear Using Standards by FEMA356
After the model is constructed the procedures of the MPA
Vol:11, No:5, 2017
 Calculate the Area of the Capacity Curve “Actual Area” ᴦn= ({Φn}T [M] {1}) / ({Φn}T [M] {Φn}) (8)
using Simpson’s rule Aa.
1
Aa= ≈
∆
0 4 1 2 2 1
[1] is a vector of unit values , T= 2 /
⋮
4 1 2 2 … (6) 1
Assuming yielding force “Fby” that does not exceed the 7. Compute the Peak Deformation Dn for each mode
ultimate force.  Using the stiffness of spring “Ki” from the bilinear curve.
 The slope of the first linear segment is taken as the same  Apply Earthquake using Sap2000.
slope of the line connecting the origin point and 0.6 from  Get the peak inelastic displacements or deformation of the
the yielding point (0.6Fby). equivalent SDOF “Dn” using nonlinear dynamic analysis.
8. Transform the Equivalent SDOF System Back
 This displacement of the SDOF system is transformed
back to its corresponding target displacement of the
building
*∆=DrΦr (9)
* Vby = Fsy M* (10)

9. Get the Desired Responses
 From the modal pushover steps (step 3), extract values of
desired responses (rn) such as floor drifts, story drifts and
plastic hinge rotations due to the combined effects of
Fig. 2 Conversion of capacity curve to bilinear curve gravity and lateral loads at roof displacement equal to ∆
 The slope of the second segment is determined by Summation of all modes r = ∑√ (11)
connecting the yielding force “Fy” with the ultimate force
“Fu” from the actual curve.  Repeat steps for as many modes as required for sufficient
 Calculate the area under the bilinear curve “Bilinear accuracy.
Area” Advantages and Disadvantages— The advantages of the
MPA can be summarized in the following points:
Abi = ⁄2 Dy Fy + ⁄2. (Du-Dy) (7)  The technique is based on a well-known structure
dynamic theory giving it a reliable basis.
 It requires only one or two modes to be in a good
 Checking that (Abi) =Area of actual curve (Aa), if not
accuracy which is a small number compared to all of the
change assumed Fy. vibration modes of a building.
6. Capacity Curve “F-D”  Taking into consideration the effect of higher modes.
Transform the pushover curve to the force–deformation  Elastic-strain hardening instead of elastic-perfectly plastic
curve “capacity curve” of an equivalent SDOF system: providing better accuracy.
 The method can be used in both new as well as existing
* D=∆/ᴦΦr * Fsy/Ln= Vby/M*= ω2Dy buildings.
For the MPA, there are also disadvantages to be mentioned,
∆= the roof drift of the building, Φr= the ordinate of the first such as:
mode shape at roof of the building Fsy=the yield force of the  Complexity of determining SDOF inelastic peak response
equivalent SDOF system. using nonlinear dynamic analysis.
 Requiring of the NLDA for SDOF system which can be
Ln=Φr*m1 done without using the MPA from the beginning
consuming more time.
Vby= the base shear of the building at yield, M*= the  Neglecting the coupling happening between modes during
effective modal mass, ω= the natural vibration frequency for inelastic response of structures.
the equivalent, The change in vibration shapes and stiffness is ignored by
SDOF system; Dy= the yield displacement of the equivalent using invariant load patterns.
SDOF system, ᴦn= the modal participation factor and is  The MPA requires as much standard POA as many the
calculated as follow: modes are involved.
Vol:11, No:5, 2017
 The MPA uses approximate rule to get the peak responses L 1 (16)
(the-square-root-sum-of-squares “SRSS”)
SMPA—After the development of the MPA, some Steps of SMPA—The SMPA is introduced as an upgrade or
limitations were found in the method. However, the main improved version for the MPA which is based on NLDA.
disadvantage of the MPA was the time consumption using the These are the steps for the SMPA using SAP2000 as a
NLDA in the steps. In order to minimize the computational modeling tool. From step 1 to step 6, it is the same steps as the
effort and time consumption, SMPA was firstly introduced by steps of MPA discussed previously. Then, by applying
[8] assuming that the higher modes cause only elastic earthquake response spectrum for the SDOF system, the peak
responses to the building. elastic displacement Uep is computed. For the inelastic
Concept of Deformation Ratio—Deformation ratio concept response of the SDOF system, the empirical relation of the
has been under considerable investigations; in order to deformation ratio provided is used for this point, in (9) and
perform the NLDA for the determination of peak elastic (10.) For the MDOF system, the target drift at the roof of the
displacement of the equivalent SDOF system. The concept building ∆p for each vibration mode can be calculated from
depends on relating the inelastic displacement up with the the peak response of the equivalent SDOF system for each
elastic displacement uep with a deformation ratio factor as mode separately as:
given by (12).
∆ ɼ∅ u (17)
C (12)
Moreover, all quantities of interest such as story drifts,
plastic hinge rotations, and floor drifts can be extracted from
where uP= the peak inelastic displacement; uep = the peak the model at roof displacement (∆). Therefore, the total peak
elastic displacement and C =the deformation ratio. response (r) can be computed using SRSS rule for each j-
Therefore, to find the value of the deformation ratio “C”, an vibration mode as shown in (18)
NLDA should be performed calculating both the elastic and
inelastic peak displacements. The value of C is determined by
applying earthquake acceleration-time history, in which the ∑ (18)
value can be affected by the earthquake characteristics. As a
result, [9] studied the effect of some parameters such as Advantages and Disadvantages—As the SMPA is though-
rapture distance, site classes, earthquake magnitude, and near out as an advancement of the MPA (MPA) so the SMPA has
fault condition on the computed value of C. many advantages compared to MPA, these advantages are:
Reference [9] introduced two empirical equations to  SMPA does not require performing any sort of pushover
compute the value of the deformation ratio based on SDOF dynamic analysis, as it exposes the complication of
system represented by elastic-strain hardening. The two determining the SDOF inelastic peak reaction by
equations depend on different factors, the first for systems presenting the proximate idea of the inelastic deformation
with known ductility Cµ as shown in (13), while the second ratio.
for systems with known yield-strength reduction factor CR as  This estimated deformation ratio is resolved such that its
shown in (13). The median values of C are for any ground lowest value is equivalent to unity; for long period
motion ignoring values for building frames, this is constant with the conservative
approximation of the equivalent displacement rule.
Cμ 1 Lμ 1 c (13) Therefore, it is exacted that the drift outcomes are going
μ
to be on the conservative side, which for practical
a, b, c, and d are constants determined from regression purposes is useful
analysis = 105, 2.3, 1.9 and 1.7. T = the elastic natural  SMPA corresponds with the way of utilizing the design
vibration period, while TC = the period separating acceleration response spectrum suggested by the late worldwide and
sensitive and velocity sensitive region. Lµ = the deformation local seismic regulations.
ratio Cµ but for zero period systems and is given by  SMPA, utilizing its actual response spectrum and the
design response spectrum also suggested by seismic
μ regulations, can be utilized for each individual
Lμ (14)
μ α
earthquake.
Nonetheless SMPA gains some limitations of the MPA
where α is the ratio between post-yield stiffness to the elastic
some of these are:
stiffness of the system, while the second relation is used for
 SMPA join distinctive modes contributions relying on the
systems with known strength reduction factor as:
square root of sum squares (SRSS) combination rule,
C 1 L 1 c (15) which rejects the algebraic sings.
 The plastic hinge rotations or other localized demands
a, b, c, and d = 61, 2.4, 1.5, and 2.4 while LR equals: may not be assessed precisely. Therefore, it is suggested
to utilize the proposed process in the estimation of the
Vol:11, No:5, 2017
probable plastic hinge area just lowering their numerical Building structure— The elevation of the structure consists
qualities. of moment-resisting frames as shown in Fig. 3. The plan of the
 Because SMPA utilizes invariant lateral load pattern, evaluation model building is a symmetric square, 45.73 m
accordingly it discards the collective harm happening in each side as shown in Fig. 4; it is divided into five steel beams
the structure. It is hard to fix this limitation without losing in each direction. The elevation of the structure consists of
the common sense of the proposed technique, therefore moment-resisting frames as shown in Fig. 3, whereas the
this limitation stays unfixed. dimensions of the building are described as the following:
basement level height is 3.65 m, ground level height is 5.49 m,
IV. MODELING AND COMPARISON BETWEEN TECHNIQUES and from the 1st to the 8th floors are 3.96 m each. The
Model Description and Characteristics—In order to connections of the columns are indicated using this sign in
compare between the previous different techniques a model is Fig. 4 and located at 1.83 m from the beam-column joint.
setup. This model is a 9-story steel structure; this building was Concrete foundation walls and surrounding soil are assumed
designed by Brandow and Johnston Associates as a typical to restrain the structure at the ground level from horizontal
medium-rise building in Los Angeles, California. displacement [10].
Fig. 3 Model building elevation
 3rd to 6th: W36x135

 7th level: W30x99
While the maximum stress for all of the beams is 248 MPa.
For the columns, the connections and the sections location are
shown in Fig. 4. While the stress for all of the columns are the
same, which is 345 MPa.
Analysis Parameters—The seismic masses of the building
including all of its components were given to be 9.65x105 kg
for the ground level, 1.01x106kg for the 1st level, 9.89x105 kg
from the 2nd to the 8th, 1.07x106 kg for the 9th level, and for the
entire structure above ground is 9.00x106 kg. It is mentioned
that each frame resists only half of the seismic mass associated
with the entire structure. EL Centro, Loma Prieta, and
Northridge earthquakes were selected for the evaluation of the
structure.
Fig. 4 Model building plan Modeling using SPA—To use the standard POA, the model
Sections Properties—The section properties of the beams is to be constructed on SAP2000 [11], then choose a load
are as follow: pattern to work on. According to the literature review, the
 From ground to the 2nd level: W36x160
Vol:11, No:5, 2017
inverted triangular load pattern is the most accurate. Fig. 5

shows the chosen load pattern in the analysis. Then, the load
pattern is used to push the building till reaching the maximum
displacement observed from the nonlinear dynamic analysis.
Fig. 5 Inverted triangle load pattern Fig. 7 Load pattern distributions for each mode
Modeling using MPA—Firstly, start applying the steps

mentioned above previously, after constructing the model on
SAP2000. To get the load pattern for the analysis, a modal
analysis is required as in [12]. While the force at each floor

equals the mode shape component multiplied by the modal
mass. Fig. 5 shows the mode shape components for the first
three modes; however, the maximum component is unity for
the multiplication simplicity. The distribution of the load
patterns on the floors for each mode is according to the
previous mode shapes and the modal masses which are given
to be the seismic mass x 0.5 as it is said to be one half and
converted from kilograms to Kips.sec2 /inch as shown in Fig.
6. Then, the assigning of the plastic hinges for the beams as
shown in Fig. 3 is required for the analysis. The next step from
Fig. 8 Pushover curve for mode 1
[13] is running the model for the three modes, while SAP2000
generates the pushover curves automatically as shown in Figs.
8-10.
The following idealized bilinear pushover curves are done

to calculate the stiffness (K) of the structure which is the slope
of the line. For mode 1, the first segment slope is 18554.6
Fig. 6 Mode shape component per floor for each mode shape
kN/m, and for the second segment is 5044.7 kN/m. For mode
2, the slope is equal to 41457.4 kN/m and 9755.7 kN/m, while
mode 3 stiffnesses are 116642.9 kN/m and 27839 kN/m. After
the idealization of the curve, comes step 6 of the MPA steps,
which is converting from the MDOF system to the equivalent
SDOF system by using the capacity curve method. Figs.11-13
demonstrate the capacity curves for the equivalent SDOF
systems for the first three modes.
Vol:11, No:5, 2017
Fig. 13 Capacity curve for mode 3

Fig. 11 Capacity curve for mode 1
Fig. 14 Equivalent SDOF system
After applying the ground motion for nonlinear dynamic

analysis, D which is the peak displacement of the equivalent
SDOF system is then transformed back to its MDOF system,
thus seismic responses can be obtained for the MPA.
Modeling using SMPA—The Simplified POA is introduced
as a modification technique for the MPA as when coming to
step 7 of the MPA steps, there is no need for the NLDA as the
analysis to be linear, and by the concept of deformation ratio,
the inelastic displacement is to be calculated. In this model,
the deformation ratio equation that will be used is the one for
the known strength reduction factor CR equation 4.17, while
Fig. 12 Capacity curve for mode 2 the four constants a, b, c, and d are equal to 61, 2.4, 1.5, and
2.4. Then, after calculating the deformation ratio for each
Then, the stiffness can be calculated from the capacity
mode, the inelastic displacement can be easily calculated from
curves which is the slope of the line segment K=V/D, and this
the elastic displacement for the SDOF system
stiffness is to be used in the SDOF using Sap2000 as a model
Comparison of Results—For the comparison between the
of a vertical cantilever with unit length having the lumped
methods, each technique is done separately, while the
mass attached to its vertical tip as shown in Fig. 14 and
validation of any of the method is due to its relativity to the
adjusted to provide the same natural period of free vibration
NLTHA. The following tables show the results of the analysis
with the stiffness calculated [14].
using ElCentro ground motion scaled to 0.25 and 1.5, Loma
Prieta and Northridge earthquakes exported from university of
Berkeley database.
Vol:11, No:5, 2017
TABLE I
PEAK FLOORS DISPLACEMENT FOR EL CENTRO X 0.25
Displacement /building Height (%)
Floor SPA MPA SMPA NLDA Error %
Mode 1 1 Mode 2 Modes 3 Modes All Modes All Modes SPA MPA SMPA
1st 0.04 0.04 0.04 0.04 0.04 0.05 -29.69 -17.41 -16.50
2nd 0.06 0.07 0.07 0.07 0.07 0.09 -29.54 -18.30 -18.67
3rd 0.09 0.09 0.10 0.10 0.10 0.12 -26.98 -18.01 -17.52
4th 0.11 0.12 0.12 0.12 0.12 0.15 -25.18 -15.89 -15.54
5th 0.14 0.15 0.15 0.15 0.15 0.17 -19.69 -13.63 -12.69
6th 0.16 0.17 0.17 0.17 0.17 0.19 -14.31 -9.15 -8.44
7th 0.19 0.19 0.19 0.19 0.20 0.20 -6.76 -2.66 -1.41
8th 0.21 0.21 0.22 0.22 0.22 0.21 -0.26 2.81 5.08
9th 0.23 0.23 0.24 0.24 0.24 0.23 0.00 2.06 4.65
TABLE II
MAXIMUM INTER-STORY DRIFTS FOR EL CENTRO X 0.25
Inter-story drift /Floor Height (%)
1st 0.25 0.28 0.29 0.33 0.29 0.22 12.50 49.12 33.61
2nd 0.24 0.27 0.28 0.29 0.26 0.22 6.82 28.35 18.19
3rd 0.23 0.26 0.26 0.23 0.25 0.22 8.14 8.20 15.56
4th 0.22 0.26 0.26 0.24 0.25 0.23 -4.35 2.33 7.74
5th 0.25 0.25 0.24 0.25 0.23 0.23 8.70 7.60 -0.48
6th 0.22 0.24 0.22 0.25 0.22 0.23 -4.30 5.59 -7.92
7th 0.23 0.23 0.22 0.22 0.23 0.25 -8.08 -13.18 -9.54
8th 0.22 0.21 0.22 0.20 0.23 0.26 -12.87 -22.37 -10.87
9th 0.20 0.17 0.19 0.18 0.19 0.21 -8.24 -15.51 -10.37
Fig. 15 Displacement/Building height % comparison for El Centro x Fig. 16 Inter-story drift/Floor height % comparison for El Centro x
0.25 0.25
TABLE III
PEAK FLOORS DISPLACEMENT FOR EL CENTRO X 1.5
1st 0.18 0.24 0.26 0.26 0.26 0.32 -42.37 -16.66 -18.61
2nd 0.33 0.42 0.44 0.45 0.45 0.52 -36.47 -13.21 -14.25
3rd 0.49 0.61 0.63 0.63 0.63 0.71 -31.38 -10.58 -10.95
4th 0.64 0.79 0.81 0.81 0.81 0.88 -26.80 -8.32 -8.24
5th 0.79 0.94 0.95 0.96 0.95 1.02 -22.35 -6.51 -6.60
6th 0.93 1.07 1.08 1.08 1.08 1.12 -17.20 -3.58 -3.94
7th 1.08 1.21 1.21 1.21 1.21 1.19 -9.71 1.54 1.23
8th 1.23 1.33 1.34 1.34 1.34 1.28 -3.93 4.29 4.31
9th 1.37 1.43 1.45 1.45 1.45 1.37 0.00 5.41 5.39
Vol:11, No:5, 2017
TABLE IV
MAXIMUM INTER-STORY DRIFTS FOR EL CENTRO X 1.5
1st 1.24 1.64 1.74 1.79 1.75 1.35 -8.36 32.53 29.43
2nd 1.39 1.71 1.76 1.76 1.77 1.28 8.70 37.73 38.31
3rd 1.45 1.73 1.74 1.70 1.73 1.31 11.20 29.98 31.95
4th 1.49 1.69 1.66 1.63 1.66 1.38 7.48 17.92 20.17
5th 1.39 1.43 1.37 1.38 1.37 1.40 -0.36 -1.09 -2.09
6th 1.29 1.24 1.17 1.21 1.18 1.41 -8.93 -14.65 -16.79
7th 1.37 1.23 1.20 1.19 1.19 1.49 -8.45 -20.39 -20.14
8th 1.45 1.17 1.22 1.18 1.22 1.53 -5.28 -22.94 -20.50
9th 1.33 0.91 1.03 1.04 1.03 1.29 3.13 -19.39 -19.88
Fig. 18 Inter-story drift/Floor height % comparison for El Centro x

Fig. 17 Displacement/Building height % comparison for El Centro x
1.5
1.5
TABLE V
PEAK FLOORS DISPLACEMENT FOR LOMA PRIETA
1 Mode 1 Mode 2 Modes 3 Modes All Modes All Modes SPA MPA SMPA
1st 0.18 0.13 0.13 0.14 0.13 0.10 75.74 36.30 34.14
2nd 0.29 0.21 0.22 0.22 0.22 0.17 72.76 35.71 34.48
3rd 0.38 0.30 0.31 0.31 0.31 0.23 68.79 35.02 34.69
4th 0.48 0.38 0.39 0.39 0.39 0.32 51.19 23.29 23.38
5th 0.56 0.46 0.47 0.47 0.47 0.42 35.34 12.82 12.71
6th 0.64 0.54 0.54 0.54 0.54 0.53 21.52 3.64 3.40
7th 0.71 0.62 0.62 0.62 0.62 0.64 9.86 -3.75 -3.92
8th 0.77 0.69 0.70 0.70 0.70 0.75 2.84 -7.30 -7.28
9th 0.82 0.75 0.76 0.76 0.76 0.82 -0.65 -7.82 -7.82
TABLE VI
MAXIMUM INTER-STORY DRIFTS FOR LOMA PRIETA
1st 1.19 0.86 0.90 0.92 0.91 0.68 75.74 36.30 34.14
2nd 1.03 0.80 0.82 0.82 0.83 0.61 68.18 34.81 35.00
3rd 0.93 0.79 0.80 0.78 0.80 0.59 58.37 33.22 35.26
4th 0.89 0.81 0.79 0.78 0.79 0.84 6.33 -6.61 -5.43
5th 0.79 0.77 0.74 0.74 0.74 0.93 -15.14 -20.51 -21.29
6th 0.70 0.72 0.69 0.70 0.69 1.02 -31.44 -31.52 -32.25
7th 0.65 0.73 0.71 0.71 0.71 1.12 -41.63 -36.39 -36.26
8th 0.59 0.69 0.72 0.70 0.71 0.99 -40.05 -28.98 -27.81
9th 0.43 0.54 0.59 0.59 0.59 0.68 -36.53 -13.18 -13.33
Vol:11, No:5, 2017
Fig. 20 Inter-story drift/Floor height % comparison for Loma Prieta

Fig. 19 Displacement/Building height % comparison for Loma Prieta
TABLE VII
PEAK FLOORS DISPLACEMENT FOR NORTHRIDGE
1st 0.25 0.18 0.19 0.20 0.20 0.22 14.94 -8.19 -7.98
2nd 0.42 0.31 0.32 0.33 0.33 0.34 22.95 -3.46 -3.23
3rd 0.56 0.43 0.44 0.45 0.45 0.43 29.91 4.06 3.89
4th 0.69 0.56 0.57 0.57 0.57 0.56 22.53 0.89 0.95
5th 0.80 0.67 0.68 0.68 0.68 0.70 14.24 -2.32 -2.26
6th 0.89 0.78 0.79 0.79 0.79 0.82 8.74 -3.59 -3.52
7th 0.98 0.90 0.90 0.90 0.90 0.92 6.52 -2.38 -2.33
8th 1.07 1.00 1.01 1.01 1.01 1.04 2.27 -3.48 -3.45
9th 1.13 1.09 1.09 1.10 1.10 1.13 0.00 -2.71 -2.67
TABLE VIII
MAXIMUM INTER-STORY DRIFTS FOR NORTHRIDGE
1st 1.70 1.25 1.30 1.35 1.36 1.48 14.94 -8.19 -7.98
2nd 1.56 1.16 1.19 1.19 1.19 1.13 37.42 5.06 5.35
3rd 1.32 1.16 1.17 1.12 1.11 0.85 55.95 32.27 30.54
4th 1.21 1.16 1.15 1.11 1.12 1.23 -1.64 -9.50 -8.65
5th 1.04 1.10 1.07 1.09 1.09 1.29 -19.61 -15.42 -15.38
6th 0.91 1.04 1.00 1.04 1.04 1.17 -22.08 -10.76 -10.61
7th 0.85 1.05 1.03 1.03 1.03 0.96 -11.35 7.37 7.32
8th 0.77 1.00 1.02 0.98 0.98 1.11 -30.84 -12.03 -12.22
9th 0.56 0.78 0.83 0.84 0.84 0.79 -28.30 6.91 7.05
Fig. 21 Displacement/Building height% comparison for Northridge Fig. 22 Inter-story drift/Floor Height % comparison for Loma Prieta
Vol:11, No:5, 2017
V. CONCLUSION REFERENCES
There are new techniques developed often in the [1] Mwafy A.M., Elnashai A.S. (2000) “Static Pushover versus Dynamic
Collapse Analysis of RC Buildings”, Journal of Engineering Structures,
seismology of structures, but the nonlinear dynamic analysis 23, pp. 407-424, 2001.
remains the most accurate and the used one. However, the [2] FEMA-273 (1997), Prestandard and Commentary for the Seismic
NLDA cannot be used for everyday analysis because it is time Rehabilitation of Buildings, Federal Emergency Management Agency,
Washington, DC, U.S.A.
consuming. Accordingly, the nonlinear static analysis is [3] ECP 201 (2012), Egyptian Code for Calculating Loads and Forces in
introduced as the meaning of POA. Nowadays, the POA has Structural Work and Masonry, National Research Center for Housing
been important issue for its effectiveness in the seismic and Building. Egypt.
analysis. Due to the limitations of the POA, the method passed [4] Paret T.F., Sasaki K.K., Eilbeck D.H., Freeman S.A. (1996).
“Approximate Inelastic Procedures to Identify Failure Mechanisms from
through many upgrades till reaching the MPA technique, Higher Mode Effects”, Proceedings of the 11th World Conference on
which gains a great attention. Earthquake Engineering, No. 966, Acapulco, Mexico.
The MPA which is the improved version of the Standard [5] Chopra A.K. and Goel R.K. (2001). “A modal pushover analysis
procedure for estimating seismic demands for buildings: Theory and
POA had the same problem as the NLDA, for that reason the preliminary evaluation.” PEER Report 2001/2003, Pacific Earthquake
SMPA was created to overcome the shortcomings of the MPA Engineering Research Center, College of Engineering, University of
which is the use of NLDA. In this thesis, the SMPA is California, Berkeley.
[6] Chopra A.K., and Goel R.K. (2003) “A Modal Pushover 15 Analysis
presented and used in the analysis. To validate the SMPA Procedure to Estimate Seismic Demands for Buildings: Summary and
method, a comparison between the techniques has been done Evaluation.” 5th National Conference on Earthquake Engineering, May,
and related to the most accurate technique which is the NLDA. Istanbul, Turkey.
[7] Chopra, A.K., and Goel, R.K., and Chintanapakdee, C. (2004),
The methods are introduced and used to compute the values of “Evaluation of a Modified MPA Procedure Assuming Higher Modes as
the floor displacements, base shear, and inter-story drifts. Elastic to Estimate Seismic Demands”, Earthquake Spectra, Vol. 20 (3),
Case Study Conclusion—The Results show the pp. 757-778Ali, S. (2013). “Energy based pushover analysis of a
building considering the effects of higher modes” (1. Aufl. Ed.).
effectiveness of the proposed SMPA in estimating the seismic Saarbrücken: LAP LAMBERT Academic Publishing.
demand of the steel frame building. The results differ from the [8] Chopra, A.K., and Goel, R.K. (2011), “A Modal Pushover Analysis
SPA, MPA, and SMPA. SPA values are far away from those Procedure to Estimate Seismic Demands for Buildings: Theory and
computed using NLDA as SPA relying on the load pattern Preliminary Evaluation”, Peer Report 2001/03, Pacific Earthquake
Engineering Research Center, University of California, Berkeley, U.S.A.
without any concern about the ground motion and don’t [9] Abdel-Rahman, A.M. (2010), “Development of Modal Pushover
consider any of the higher modes. Nevertheless, MPA and Procedure Based on Response Spectrum for Estimating seismic
SMPA values are very efficient compared to the NLDA. Demands of RC Building Frames”, M.Sc. Thesis, Structural Engineering
Department, Faculty of Engineering, Cairo University, EgyptBaik S.-W.,
According to the three ground motions [15] used in the Lee D.-G., Krawinkler H. (1988) “A simplified model for seismic
analysis, the outcomes illustrate that resonance may occur in a response prediction of steel frame structures”, Proc. 9th world
few stories, only if frequency at which ground shakes is steady conference earthquake engineering., Tokyo, Kyoto, Vol. 5, , pp. 375-
380.
at or near any of the natural frequencies of building and [10] Chintanapakdee, C. (2002). “Evaluation of the Modal Pushover Analysis
applied over an extended period of time. However, most of the Procedure using vertically "regular" and irregular Generic Frames”,
results show that there is an underestimation (does not exceed Berkeley, California, USA.
[11] SAP2000 (2013), Integrated Software for Structural Analysis and
20%) in some floors according to the peak floor displacements Design, Computers and Structures, Inc., Berkeley, CA, U.S.A.
and inter-story drifts in the weak ground motion, while the [12] Musa, A.M. (2011), “Comparison between Advanced Multimode
opposite happens in the strong ground motion. However, the Pushover Techniques for Evaluation of Seismic Demands and
Performance of Code-Designed Steel Frames”, M.Sc. Thesis, Structural
number of modes used in the MPA shows that the higher Engineering Department, Faculty of Engineering, Cairo University,
modes effect is a must to be considered. As a result, the MPA Egypt
for the three modes and the SMPA for all modes appear to be [13] Chopra, A.K., & Chintanapakdee, C. (2004), “Inelastic Deformation
equal in most of the values, but the MPA took more analysis Ratios for Design and Evaluation of Structures: Single-Degree-of-
Freedom Bilinear Systems”, Journal of Structural Engineering, ASCE,
time. Vol. 130 (9), pp. 1309-1319.
Thus, SMPA simplifies the use of POA for everyday [14] El-Esnawy, N.A. (2007), “Evaluation of Seismic Demands for RC
seismic design. As the SMPA does not need the nonlinear Building Frames Using Modal Pushover Analysis Method”, Journal of
Engineering and Applied Science, Faculty of Engineering, Cairo
dynamic analysis, the method became an effective and time University, Vol. 54 (3), pp. 339-358.
saving technique. In general, the SMPA results are [15] Jingjiang S., Ono T., Yangang Z., Wei W., (2003), “Lateral load pattern
conservative and better than those from the MPA. in pushover analysis.” Earthquake Engineering and Engineering
Vibration, Vol. 2(1), 99-107.
Future Work—According to the presented study, some
investigations are listed as follow for the future researches and
study:
1. Application of the proposed SMPA for irregular frame
buildings and shear walls.
2. Investigation of the SMPA under more real and artificial
earthquakes time-history.
3. Study the accuracy of the following SMPA for the
buildings designed using different codes of design
including the Egyptian code of Practice.

Comparison Between Pushover Analysis Techniques and Validation of The Simplified Modal Pushover Analysis

Uploaded by

Copyright:

Available Formats

Comparison Between Pushover Analysis Techniques and Validation of The Simplified Modal Pushover Analysis

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Comparison Between Pushover Analysis Techniques and Validation of The Simplified Modal Pushover Analysis

Uploaded by

Copyright:

Available Formats

World Academy of Science, Engineering and Technology

International Journal of Urban and Civil Engineering

Comparison between Pushover Analysis Techniques

 are based on the mode shapes of vibration of the building

II. STANDARD PUSHOVER ANALYSIS the first three modes of vibration.

primary mode of vibration is one of these limitations. In the

A. Steps of MPA 5. Transform the Capacity Curve of SDOF System to

* Vby = Fsy M* (10)

Fig. 3 Model building elevation

 3rd to 6th: W36x135

inverted triangular load pattern is the most accurate. Fig. 5

Modeling using MPA—Firstly, start applying the steps

analysis is required as in [12]. While the force at each floor

Fig. 9 Pushover curve for mode 2

The following idealized bilinear pushover curves are done

Fig. 13 Capacity curve for mode 3

Fig. 10 Pushover curve for mode 3

Fig. 11 Capacity curve for mode 1

Fig. 14 Equivalent SDOF system

After applying the ground motion for nonlinear dynamic

Fig. 18 Inter-story drift/Floor height % comparison for El Centro x

Fig. 20 Inter-story drift/Floor height % comparison for Loma Prieta

You might also like