Pushover Analysis Procedures

CE 75.
03 – Earthquake Engineering for Tall Buildings

In-class Session
Semester – August 2020
Dr. Fawad A. Najam Prof. Dr. Pennung Warnitchai

Department of Structural Engineering Professor and Department Head
NUST Institute of Civil Engineering (NICE) Department of Civil and Infrastructure Engineering
National University of Sciences and Technology (NUST) School of Engineering and Technology (SET)
H-12 Islamabad, Pakistan Asian Institute of Technology (AIT)
Cell: 92-334-5192533, Email: fawad@nice.nust.edu.pk Bangkok, Thailand
Topics for this Session
Session 1 (1.25 Hours):

Introduction to the Nonlinear Static (Pushover) Analysis Procedures
• The Capacity Spectrum Method (ATC 40, FEMA 356)
• The Displacement Coefficient Method (ATC 40, FEMA 356)
• The FEMA 440 Improved NSPs (FEMA 440, ASCE 41-06/13)
Session 2 (1.25 Hours):
Approximate Multi-mode based Seismic Analysis Procedures
• The Modal Pushover Analysis (MPA) Procedure
• The Uncoupled Modal Response History Analysis (UMRHA) Procedure
2
An Introduction to the Nonlinear Static (Pushover)
Analysis Procedures

Department of Structural Engineering Professor and Department Head
NUST Institute of Civil Engineering (NICE) Department of Civil and Infrastructure Engineering
National University of Sciences and Technology (NUST) School of Engineering and Technology (SET)
H-12 Islamabad, Pakistan Asian Institute of Technology (AIT)
Cell: 92-334-5192533, Email: fawad@nice.nust.edu.pk Bangkok, Thailand
Seismic Analysis Procedures
Structural
Model Linear Nonlinear
E, A, I, L, G etc. = Constant, K= Constant E ≠ Constant, EI ≠ Constant, K≠ Constant
Seismic
Loading
1. Equivalent Lateral Force (ELF) Procedure Several Pushover Analysis Methods or Nonlinear
5. Static Procedures (NSPs)
Static
Response Spectrum Analysis (RSA) Procedure (or
2. Mode Spectral Analysis)
Nonlinear Modal Response History Analysis or Fast
Modal Response History (or Time History) Analysis 6. Nonlinear Analysis (FNA)
3.
Procedure (Modal RHA/THA)
Dynamic Linear Response History (or Time History) Analysis Nonlinear Response History (or Time History)
4. 7. Analysis Procedure (Direct Integration Nonlinear
Procedure [Direct Integration, Modal Superposition]
(Linear RHA /LTHA) RHA/THA)
4
Nonlinear Static Analysis Procedures (NSPs) – Pushover Analysis Procedures
• The nonlinear analysis recommended for the first generation of performance-based seismic design
methodology. Currently, it can be regarded as an alternate method of analysis for carrying out the PBD.
Basic Idea of Pushover Analysis
Full 3D Nonlinear MDF Model Monotonic Pushover Analysis Base Pushover Curve
Shear
𝐹𝑛 𝑥𝑟 𝑉𝑏
.
. Control Node
≈ .
𝐹1 𝑥𝑟
Displacement
𝑉𝑏 of Control Node
Monotonically increasing lateral load Develop the 𝑉𝑏 vs. 𝑥 𝑟 relationship

A ground motion structure in pattern 𝐹1 … … 𝐹𝑛 to push the structure in
a particular X or Y direction that particular X or Y direction
5
Nonlinear Static Analysis Procedures (NSPs) – Pushover Analysis Procedures
Full 3D Nonlinear MDF Model Monotonic Pushover Analysis Base Pushover Curve
Shear
𝑟
𝑥𝑖𝑛 𝐹𝑛 𝑥𝑟 𝑉𝑏
.
. Control Node Structural demands at this
point → Seismic Demands
≈ .
𝐹1 𝑥𝑟
𝑟
𝑉𝑏 𝑥𝑖𝑛 Displacement
of Control Node
Monotonically increasing lateral load Develop the 𝑉𝑏 vs. 𝑥 𝑟 relationship

A ground motion in a pattern 𝐹1 … … 𝐹𝑛 to push the structure in
particular X or Y direction that particular X or Y direction
𝑟 𝑟
Let's set the control node as a roof node and During the push, when 𝑥 𝑟 = 𝑥𝑖𝑛 , the force and How to Determine 𝑥𝑖𝑛 ?
denote the peak inelastic roof displacement displacement demands of the structure are the peak
𝑟 What load pattern to apply?
occurred during the ground motion as 𝑥𝑖𝑛 . seismic demands (produced by the ground motion).
6
Lateral Loads Pattern in Pushover Analysis
• Start with Gravity Loads

• Dead Load
Control Point Displacement Distribution of
• Some Portion of Live Load Pushover Force
• Select Lateral Load Patterns

• Uniform (rectangular)
• Code-specified static lateral load distribution
• Inverted triangle
• First mode shape (assuming that the building
will primarily deform in first mode shape only) 𝑉𝑏
• Combination of mode shapes
What load pattern to apply?
7
Lateral Loads Pattern in Pushover Analysis
• Generally, the deformation pattern corresponding to the “First-mode Inertia Load Pattern” is used as
the basis for analysis.
𝒇𝑖 = 𝑚𝑖 𝝓𝟏
Where 𝑓𝑖 = Lateral force at any 𝑖 𝑡ℎ storey
𝑚𝑖 is the mass of any 𝑖 𝑡ℎ storey
𝝓𝟏 is the first mode shape vector
𝑽𝑏 = ෍ 𝒇𝑖 = ෍ 𝑚𝑖 𝝓𝟏
• This is generally acceptable for structures with time period less than or equal to 1 second.
• For more flexible structures, the contributions of higher vibration modes may become significant and need
to be considered in the pushover analysis (multi-mode pushover analysis).
8
Pushover Analysis for Performance Evaluation of Buildings
Vertical Loads
roof
F4
Pushover Curve
Base Shear, V𝑉𝑏

Performance point
F3
Pushover
Nonlinear Pushover
F2
Analysis
Analysis
Horizontal
Load Pattern F1
𝑟
𝑥𝑖𝑛
Displacement, 𝑥 𝑟
𝑉V𝑏
Pushover Analysis Procedures:
Performance point → A point on the pushover curve
1) Construct the nonlinear structural model and select loads
corresponding to the likely peak inelastic control-node
Performance
2) Apply gravity loads 𝑟
displacement (𝑥𝑖𝑛) during the ground shaking.
Point
3) Apply horizontal load pattern (usually first mode shape)
In other words, the future earthquake is expected to push the
Base Shear, V
4) Incrementally apply the pattern to get the pushover curve building up to this point.
𝑟
5) Determine the performance point Need a separate method to estimate this 𝑥𝑖𝑛
𝑟
6) Evaluate the structural performance at 𝑥𝑖𝑛 (pass or fail criteria evaluated on 𝑟 is also sometimes referred to as the “Target Displacement”.
𝑥𝑖𝑛
component by component or global structural basis)
Spectral
9
Displacement
Collapse Prevention
Operation Level
sta
Re nt
ura
Resta
urant
Life Safety
Loading Severity
Global Response and
Performance
Immediate Occupancy
Resta
urant
Collapsed
Structural Displacement
10
Pushover Curve of an Example 44-storey RC Shear Wall Building Strong (X) Direction
Column’s steel bars begin to yield in tension

0.09
Base Shear/Building Weight (VB/W)

Column’s bars begin to yield in compression
0.08
0.07
Concrete crushing in shear wall

0.06
Shear wall’s steel bars begin to yield in compression

0.05
0.04 Columns begin to Crack

Shear wall’s steel reinforcement bars begin to yield in tension
0.03
0.02
Shear walls begin to crack
0.01 Brick walls begin to crack
0
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045
Roof Drift (Top Displacement/Building Height)

11
Columns begin Shear wall’s steel bars begin
0.04
Normalized Base Shear (𝑉𝑏1 /𝑊) to Crack to yield in compression
0.03
Monotonic Pushover Concrete crushing in

Analysis Results 0.02 Shear wall’s steel shear wall
Shear walls begin reinforcement bars begin to
to crack yield in tension
0.01
20-story Building – First Brick walls begin to crack
Mode in Y (weak) Direction
0
-0.04 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 0.04
Roof Drift (𝑥𝑖𝑟 /𝐻)
-0.01
Monotonic Pushover in Positive Direction

-0.02
Monotonic Pushover in Negative Direction
-0.03
-0.04
MS Thesis Progress
Shear wall’s steel bars begin Column’s bars begin to
to yield in compression yield in compression
0.09
Normalized Base Columns
Shear (𝑉𝑏1 /𝑊) begin to
0.07 Crack
Column’s steel bars
begin to yield in tension
The Reversed-cyclic 0.05 Concrete crushing in
Pushover Analysis shear wall
Shear wall’s steel reinforcement
0.03 bars begin to yield in tension
Shear walls begin to crack
0.01 Brick walls begin to crack
Example of a 44-story case study
building in its Strong Direction -0.04 -0.03 -0.02 -0.01 -0.01 0 0.01 0.02 0.03 0.04
Roof Drift (𝑥𝑖𝑟 /𝐻)
-0.03
Cyclic Pushover
-0.05
-0.07
-0.09
13
Normalized Base Shear ( )
0.06 0.08 0.1
19-story Building
20 33-story Building 44-story Building
0.06 0.08
0.04 Mode 1 Mode 1 Mode 1
0.06
0.04
0.04
0.02
0.02 0.02
0 0 0
-0.02 -0.02
-0.02
-0.04
-0.04
-0.06
-0.04
-0.06 -0.08
-0.06 -0.08 -0.1
-0.035 -0.015 0.005 0.025 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 -0.03 -0.02 -0.01 0 0.01 0.02 0.03
Roof Drift ( ) Roof Drift ( ) Roof Drift ( )
0.12 0.25 0.25
19-story
20 Building 33-story Building 44-story Building
0.2 0.2
0.08
Mode 2 Mode 2 Mode 2
0.15 0.15
The Cyclic Behavior of 0.04 0.1
0.1
0.05
0.05
Some Example Buildings 0

0
-0.05
0
-0.04 -0.05
-0.1
-0.1 -0.15
-0.08
-0.15 -0.2
-0.12 -0.2 -0.25
-0.025 -0.015 -0.005 0.005 0.015 0.025 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 -0.03 -0.02 -0.01 0 0.01 0.02 0.03
0.2 0.3 0.3
19-story
Mode 3 0.2 Mode 3 0.2
Mode 3
0.1
0.1 0.1
0 0 0
-0.1 -0.1
-0.1
-0.2 -0.2
-0.2 -0.3 -0.3

-0.025 -0.015 -0.005 0.005 0.015 0.025 -0.03 -0.01 0.01 0.03 -0.03 -0.02 -0.01 0 0.01 0.02 0.03
Roof Drift ( ) Roof Drift ( ) Roof Drift ( 14 )
The Concept of Equivalent SDF System
Full 3D Nonlinear Monotonic Pushover Determination of Idealization of An Equivalent

MDF Model Analysis Procedure Pushover Curve Pushover Curve SDF System
𝐹𝑛 𝑉𝑏 𝐹
𝑥𝑟
.
.
.
𝐹1
𝑥𝑟 𝐷
𝑉𝑏
Converting a Nonlinear MDF System into an Equivalent NL SDF System
15
Cyclic Pushover Curve of an MDF Converted to Force vs. Displacement

System (Building) Relationship DOF
0.09 0.09
Normalized Base Force
Shear (𝑽𝒃 /𝑾) 0.07
0.07 m
0.05 0.05
0.03 0.03
0.01 0.01
Assumed
𝑘 𝑐
-0.01 0
-0.04 -0.03 -0.02 -0.01 0.01 0.02 0.03 0.04 -0.01 0
-0.04 -0.03 -0.02 -0.01 0.01 0.02 0.03 0.04 Damping
Roof Drift (𝒙𝒓 /𝑯) -0.03
Displacement
-0.03
-0.05 -0.05
-0.07 -0.07
-0.09 -0.09
Normalized Base Shear (𝑽𝒃 /𝑾) Force
Roof Drift (𝒙𝒓 /𝑯) Displacement
16
Assumed Hysteretic Rule
DOF
𝑘 𝑐 Assumed
Damping
Capacity Curve
𝑽𝒃
𝒙𝒓
17
𝑟
How to Determine 𝑥𝑖𝑛 (i.e. Performance Point or Target Displacement)?
Equivalent • Several Equivalent Linearization Procedures (Individual studies)
Linearization • Capacity Spectrum Method (CSM) (ATC 40, FEMA 273, FEMA 356)
• FEMA 440 Improved Equivalent Linearization Procedure
Approach
Performance Point (or
Target Displacement) on
• Several displacement modification procedures (Individual studies)
Pushover Curve Displacement
• Displacement Coefficient Method (ATC 40, FEMA 273, FEMA 356)
Modification
• FEMA 440 Improved Displacement Coefficient Method (FEMA 440,
Approach
ASCE 41-06, ASCE 41-13)
NLRHA of the
Equivalent SDF
System
18
Equivalent Linearization Approach
The Capacity Spectrum Method (ATC 40, FEMA 273, FEMA 356)
19
Analysis Procedures in ATC 40
The Equal-
Displacement
Approximation
estimates that the
inelastic displacement
is the same as that
which would occur if
the structure remained
completely elastic.
20
Conversion of a Nonlinear SDF System into an “Equivalent Linear” SDF System
The basic idea of 𝐹 𝐹

Energy Dissipation by Total Energy Dissipation
equivalent linearization Hysteretic Damping 2
𝑚𝜔𝑒𝑞 by the Equivalent
2
2 (𝜉ℎ ) 1
Viscous Damping (𝜉𝑒𝑞,𝑖 ) 𝑚𝜔𝑒𝑞
approach 𝑚𝜔𝑖 1
1
𝐷𝑚𝑎𝑥 𝐷 𝐷𝑚𝑎𝑥 𝐷
𝜉𝑒𝑞 = κ𝜉𝑖 + 𝜉ℎ 2𝜋
𝑇𝑒𝑞 =
𝜔𝑒𝑞
𝜉𝑒𝑞 and 𝑇𝑒𝑞 are the
“equivalent linear
A nonlinear SDF system with initial An Equivalent Linear System
circular natural frequency 𝜔𝑖 , and with with Elongated Period and properties”
initial inherent viscous damping 𝜉𝑖 Additional Damping
21
Capacity Spectrum Method (ATC 40, 1996)
Conversion of Pushover Curve Determination of Performance
Full 3D Nonlinear MDF Model Monotonic Pushover Analysis
to ADRS Format Point
𝑆𝐴
𝐹𝑛
𝑥𝑟 𝑉𝑏 𝑆𝐴
. Capacity Spectrum
.
Demand Spectrum (modified
. for equivalent damping)
𝑥𝑟 𝑆𝐷
𝐹1
𝑆𝐷
𝑉𝑏 𝑉𝑏 𝑥𝑟 𝛿𝑃
Determine the modal properties 𝑆𝐴 = 𝑆𝐷 =
𝑊 Τ𝐶𝑚 Γ𝜙1𝑟 Performance Point
𝜙1 , 𝑇, Γ and 𝐶𝑚 Monotonically push and get the
𝑇2 The “Demand Spectrum” is the
𝑉𝑏1 vs. 𝑥 1𝑟 relationship 𝑆𝐷 = 𝑆𝐴
4𝜋 2 reduced form of 5% damped spectrum
𝑉𝑏1 Determination of Responses

𝐹𝑛
• Capacity Spectrum → Another form of pushover curve (SA-SD form). . 𝛿𝑃
.
• Demand Spectrum → Another form of response spectrum (ADRS .
Form) reduced based on effective damping (i.e., original inherent 𝐹1
𝑥 1𝑟
damping + additional hysteretic damping) 𝛿𝑃 𝑉𝑏1
Extract the responses from the pushover database at 𝛿𝑃
22
23
24
Reduction of Response Spectrum to get the Demand Spectrum
𝐹
2
Energy Dissipation by 𝑚𝜔𝑠𝑒𝑐
Equal-Energy Assumption Hysteretic Damping (𝜉ℎ ) 1
Energy dissipated by hysteretic damping

in one force-deformation loop of an actual
nonlinear system
𝐸𝐷 (𝐷ሻ
Determination of Hysteretic =
Energy dissipated by an equivalent
Damping – Equal-energy amount of viscous damping in one 𝐸𝑠𝑜 (𝐷ሻ
harmonic cycle of an equivalent linear
Dissipation Rule system
𝐷
𝐷𝑚𝑎𝑥
Energy Dissipation by
Equivalent Viscous
Damping (𝜉𝐸𝐿 )
1 𝐸𝐷 (𝐷𝑖 ሻ
𝜉ℎ =
4𝜋 𝐸𝑠𝑜 (𝐷𝑖 ሻ
For details, see Section 3.9, Dynamics of
Structures by A. K. Chopra (2012), 4 th Edition.
𝐹𝑚𝑎𝑥
25
Reduction of Demand Spectrum
26
(secant period)
𝑇𝑒𝑞 and 𝛽𝑒𝑞 are the

𝑆𝑑
“equivalent linear 𝑇 = 2𝜋
𝑆𝑎
properties”
Taken from FEMA 440 (2005)

27
Reduction of Demand Spectrum
28
Spectral Reduction Factors
29
Intersection of Capacity Spectrum and Demand Spectrum
30
B
Displacement, 
Spectral Displacement
Summary – Capacity Spectrum Method (ATC 40, 1996)
Performance Point
Vertical Loads roof
Pushover Curve ADRS
Base Shear, V
F4
Spectral Acceleration
Base Shear, V
F3
Pushover
Nonlinear Pushover
F2 Analysis
Analysis
Horizontal
Load Pattern F1
Displacement, 
𝑉V𝑏
Performance Point ADRS
Acceleration
Pushover Curve
Earthquake Response Spectra
Shear, V
Base Shear, V
Base
Spectral
𝛿𝑃
Displacement,  Spectral Displacement Spectral Displacement Time Period
ation
Performance Point 31
ADRS Earthquake Response Spectra
Iterative Procedures to Determine Performance Point
Extract all results (structural demands) from the software at that particular
step of pushover analysis (i.e., corresponding to the Performance Point).
32
Equal-displacement Assumption
33
Displacement Modification Approach
The Displacement Coefficient Method (ATC 40, FEMA 273, FEMA 356)
34
Displacement Coefficient Method (ATC 40, FEMA 273, FEMA 356)
• The displacement coefficient method provides a direct numerical process for calculating the displacement
demand. It does not require converting the capacity curve to spectral coordinates.
• The underlying idea is “the peak elastic displacement of the equivalent SDF system of the structure can
be modified with a series of coefficients (each accounting for a specific aspect of nonlinearity) to
approximately obtain the peak inelastic (target) displacement.”
• The coefficient method is based on statistical analysis of the results of time history analysis of single
degree of freedom models of different types.
• The demand displacement in the coefficient method is called the target displacement.
35
• Step 1: Construct a bilinear representation of the capacity curve.
36
• Step 2: Calculate the effective fundamental period (𝑇𝑒 ) as;
37
• Step 3: Calculate the target displacement (𝛿𝑇 ሻas:
For converting elastic to To account for dynamic

inelastic displacement 𝑃 − Δ effects
For converting spectral “Effective” period

to roof displacement
𝑇𝑒2
Target Displacement 𝛿𝑇 = 𝐶0 𝐶1 𝐶2 𝐶3 𝑆𝐴 2 𝑔
4𝜋
To account for the shape Spectral acceleration at 𝑇𝑒

of hysteresis loops
38
39
Extract all results (structural demands) from the
software at that particular step of pushover analysis
(i.e., corresponding to the target displacement 𝛿𝑇 ).
40
FEMA 440 Improved Nonlinear Static Procedures (CSM and DCM)
41
Improved NSPs – FEMA 440 (2005)
Elastic-Perfectly Plastic (EPP) Stiffness-Degrading (SD)
3 3
2 2
1 1
Force
Force
0 0
-4 -3 -2 -1 0 1 2 3 4 -4 -3 -2 -1 0 1 2 3 4
-1 -1
-2 -2
EPP SD
-3 -3
Displacement Displacement
Strength and Stiffness-Degrading (SSD) Nonlinear Elastic (NE)

3 3
2 2
1 1
Force
Force
0 0
-4 -3 -2 -1 0 1 2 3 4 -4 -3 -2 -1 0 1 2 3 4
-1 -1
-2 -2
SSD NE
-3 -3
Displacement Displacement
42
Inelastic-to-elastic Displacement Ratios (IDRs) (FEMA 440, 2005)
𝐶1,𝐸𝑃𝑃 Site Class C – Mean of 20 Ground Motions
4
𝑅𝑦 = 8
3.5 𝑅𝑦 = 6
𝑅𝑦 = 4
3
𝑅𝑦 = 3
2.5 𝑅𝑦 = 2
𝑅𝑦 = 1.5
2 Two Basic Spectral Regions
1.5
𝐶1 is on average larger than 1 𝐶1 is approximately constant with changes in 𝑇

0.5 𝐶1 increases with decreasing 𝑇 𝐶1 doesn’t change much with changes in 𝑅𝑦
𝐶1 increases with increasing 𝑅𝑦 𝐶1 is on average approximately equal to 1
0 𝑇 (sec)
0 0.5 1 1.5 2 2.5 3
43
FEMA 440 (2005) and ASCE/SEI 41-13 (2013) Nonlinear Static Procedure
Full 3D Nonlinear MDF Model Monotonic Pushover Analysis Idealization of Pushover Curve Determination of Target
Displacement
𝐹𝑛 𝑉𝑏1 • Determine the effective time
𝑥1𝑟 period and coefficients
.
𝐾𝑒 𝑅𝑦 − 1
. 𝑇𝑒 = 𝑇𝑖 𝐶1 = 1 +
Idealized 𝐾𝑖 𝑎𝑇𝑒2
. Actual
2
1 𝑅𝑦 − 1
𝐹1 𝑥 1𝑟 𝐶2 = 1 +
800 𝑇𝑒
𝑉𝑏1 • Idealized force-displacement
• Determine the modal • Target displacement
properties 𝜙1 , 𝑇, Γ and 𝐶𝑚 • Monotonically push and get relationship. 𝑇𝑒2
the 𝑉𝑏1 vs. 𝑥 1𝑟 relationship • Determine 𝐾𝑒 , 𝑉𝑦 and 𝑅𝑦 𝛿𝑇 = 𝐶𝑜 𝐶1 𝐶2 𝑆𝐴 2 𝑔
4𝜋
𝑉𝑏1 Determination of Responses

Target displacement:
𝐹𝑛
𝑇𝑒2 . 𝛿𝑇
𝛿𝑇 = 𝐶𝑜 𝐶1𝐶2𝑆𝐴 2 𝑔
4𝜋 .
.
2
𝐾𝑒 𝑅𝑦 − 1 1 𝑅𝑦 − 1 𝐹1
𝑇𝑒 = 𝑇𝑖 , 𝐶1 = 1 + , 𝐶2 = 1 + 𝑥 1𝑟
𝐾𝑖 𝑎𝑇𝑒2 800 𝑇𝑒 𝛿𝑇 𝑉𝑏1
• Extract the responses from the pushover database at 𝛿 𝑇
44
FEMA 440 (2005) Improved Equivalent Linearization Method
• Optimal equivalent linear parameters (i.e., effective period, 𝑇𝑒𝑓𝑓 , and effective damping, 𝛽𝑒𝑓𝑓) are
determined through a statistical analysis that minimizes, in a rigorous manner, the extreme occurrences
of the difference (i.e., error) between the maximum response of an actual inelastic system and its
equivalent linear counterpart.
• New expressions for effective period, 𝑇𝑒𝑓𝑓 , and effective damping, 𝛽𝑒𝑓𝑓 are proposed.
45
46
47
48
49
Modified ADRS to use with Tsec
50
Modified ADRS to use with Tsec
51
Spectral Reduction for Effective Damping
52
The NLRHA of the Equivalent SDF System
53
Determining the Performance Point using NLRHA of the Equivalent SDF System
Full 3D Nonlinear Monotonic Pushover Determination of Idealization of An Equivalent

MDF Model Analysis Procedure Pushover Curve Pushover Curve SDF System
𝐹𝑛 𝑉𝑏 𝐹
𝑥𝑟
.
.
. + Hysteretic Rule
𝐹1
𝑥𝑟 𝐷
𝑉𝑏
𝑉𝑏 Performance Point
𝑥𝑟 𝑟
Extract all results (structural demands) from 𝑥𝑖𝑛
the software at that particular step of
𝑡
𝑟
pushover analysis (i.e., corresponding to 𝑥𝑖𝑛 ). 𝑥𝑟
𝑟
𝑥𝑖𝑛
Structural demands at this
point → Seismic Demands
54
Pushover Analysis (Advantages and Disadvantages)
• Advantages
• It requires less computer time compared to dynamic analysis.
• Pushover curve can provide the lateral capacity of the building and expose the inherent strengths and weaknesses of the structure.
• It can provide useful information on the effects of changing the strength and stiffness.
• Disadvantages
• Consideration of only one-dimensional static loads
• Consideration of only one failure mode
• It does not account directly for the dynamic nature of earthquake loads (Inertia forces not considered).
• It does not account directly for the hysteretic loops – i.e., hysteretic energy dissipation and stiffness degradation are considered only
indirectly.
• It works well if the structure responds in essentially a single mode of vibration (Higher-mode effects are neglected).
• It is less accurate, and possibly inaccurate for tall buildings
55
Approximate Multi-mode based Seismic Analysis
Procedures
In-class Session – Semester - January 2020

Department of Structural Engineering Head, Department of Civil and Infrastructure Engineering
NUST Institute of Civil Engineering (NICE) School of Engineering and Technology (SET)
National University of Sciences and Technology (NUST) Asian Institute of Technology (AIT)
H-12 Islamabad, Pakistan Bangkok, Thailand
Cell: 92-334-5192533, Email: fawad@nice.nust.edu.pk
Pushover Analysis Procedures
Lateral Load Vectors
Simple (Uniform, Triangular etc.) First Modal Inertia Multi-mode
Modal Combination Method (Kalkan and Kunnath, 2004)

Modal Combination of
Upper-bound Pushover Analysis (Jan et al., 2004)
Loading
Multi-mode Pushover Generalized Pushover Analysis (Sucuoglu and Gunay, 2010)
Analysis Procedures Multimode Pushover Analysis (Sasaki et al., 1996)

Modal Combination of
Modal Pushover Analysis (Chopra and Goel, 2002)
Response
Modified Modal Pushover Analysis (Chopra et al., 2004 )
Consecutive Modal Consecutive Modal Pushover Analysis (Poursha et al., 2009)

Pushover (CMP) Modified CMP Analysis (Khoshnoudian and Kiani, 2012)
Adaptive Modal Pushover Analysis (Kalkan and Kunnath, 2006)

Adaptive Modal
Optimal Multimode Pushover Analysis (Attard and Fafitis, 2005)
Pushover (AMP)
Adaptive Pushover Procedure (Antoniou et al., 2002)
57
Why We Still Need Approximate Seismic Analysis Procedures?
Despite the development of fast computing tools, software and other advancements, the detailed nonlinear
RHA is still a difficult task for several reasons.
• Ground motions compatible with the seismic design spectrum for the site must be selected.
• Computationally demanding, inelastic modeling, 3D analysis to account for coupling between lateral
and torsional motions, subjected to 2 horizontal components of ground motions.
• Must be repeated for several excitations.
• Structural model must be sophisticated enough to represent a building realistically, especially
deterioration in its strength at large displacements.
So, approximate methods are still an attractive option as an alternate to the rigorous NLRHA procedure.
58
Effective Earthquake Forces on an MDF System
The governing equation of motion for an elastic MDF system subjected to the earthquake ground
motion is
𝑴𝒖ሷ 𝑡 + 𝑪𝒖ሶ 𝑡 + 𝑲𝒖(𝑡ሻ = −𝑴 𝟏 𝒖ሷ 𝒈 𝒕 = 𝑷𝒆𝒇𝒇 𝒕
Where the effective earthquake forces are
𝑷𝑒𝑓𝑓 𝑡 = −𝑴 𝟏 𝑢ሷ 𝑔(𝑡ሻ
The spatial distribution of these forces over the structure is defined by the vector
𝒔 = 𝑴𝟏
59
The idea of modal expansion of excitation vector 𝑷 𝑡 of the form 𝑷 𝑡 = 𝒔 𝑝(𝑡ሻ
𝑷 𝑡 = 𝒔𝑝 𝑡
The primary idea is to expand the vector 𝒔 as
𝑁 𝑁
𝒔 = ෍ 𝒔𝑟 = ෍ 𝛤𝑟 𝑴 𝜙𝑟
𝑟=1 𝑟=1
This equation may be viewed as an expansion of the distribution 𝒔 of applied forces in terms of inertia force
distributions 𝒔𝑟 associated with natural modes.
Pre-multiplying both sides of above equation by 𝝓𝑇𝑛 and utilizing the orthogonality property of modes gives
𝝓𝑇𝑛 𝒔
𝛤𝑛 =
𝑀𝑛
The contribution of the nth mode to 𝒔 is
𝒔𝑛 = 𝛤𝑛𝑴𝜙𝑛
60
Modal Expansion of the Effective Earthquake Forces
𝑷𝑒𝑓𝑓 𝑡 = −𝑴 𝟏 𝑢ሷ 𝑔 𝑡 = −𝒔 𝑢ሷ 𝑔 𝑡
• This force distribution can be expanded as a summation of modal inertia force distributions
𝑁 𝑁
𝒔 = 𝑴 𝟏 = ෍ 𝒔𝑛 = ෍ Γ𝑛 𝑴 𝝓𝑛
𝑛=1 𝑛=1
𝐿𝑛
Where Γ𝑛=
𝑀𝑛
𝐿𝑛 = 𝜙𝑛𝑇 𝑴 𝟏
𝑀𝑛 = 𝜙𝑛𝑇 𝑴 𝜙𝑛
61
• The effective earthquake forces can then be expressed as
𝑁 𝑁
𝑷𝑒𝑓𝑓 𝑡 = ෍ 𝑷𝑒𝑓𝑓 , 𝑛 𝑡 = ෍ −𝒔𝑛 𝑢ሷ 𝑔(𝑡ሻ

𝑛=1 𝑛=1
• The contributions of the 𝑛𝑡ℎ mode to 𝑷𝑒𝑓𝑓 𝑡 and 𝒔 are
𝑷𝑒𝑓𝑓 , 𝑛 𝑡 = −𝒔𝑛 𝑢ሷ 𝑔(𝑡ሻ
𝒔𝑛 = 𝛤𝑛 𝑴 𝝓𝑛
62
Modal expansion of the distribution 𝒔 = 𝑴 𝟏 of effective earthquake forces
The direction of force 𝑠𝑗𝑛 at the 𝑗𝑡ℎ floor level is controlled by the algebraic sign of
𝜙𝑗𝑛 , the 𝑗𝑡ℎ-floor displacement in mode 𝜙𝑛
63
• Utilizing the modal expansion of 𝑷𝑒𝑓𝑓 𝑡 and 𝒔, two procedures for approximate analysis of inelastic
buildings are proposed by Chopra and Goel (2002).
• The uncoupled modal response history analysis (UMRHA), and
• The modal pushover analysis (MPA)
• Not intended for practical application, the UMRHA procedure is developed only to provide a rationale for
the MPA procedure.
• In the UMRHA procedure, the response history of the building to 𝑷𝑒𝑓𝑓 , 𝑛 𝑡 , the nth-mode component of
the excitation, is determined by nonlinear RHA of an inelastic SDF system, and superposition of these
“modal” responses gives the total response.
• In the MPA procedure, the peak response to 𝑷𝑒𝑓𝑓 , 𝑛 𝑡 is determined by a nonlinear static, or pushover,
analysis, and the peak modal responses are combined by modal combination rules to estimate the total
response.
64
The Uncoupled Modal Response History Analysis (UMRHA)
Procedure (Chopra and Goel, 2002)
65
The UMRHA Procedure
Mode 1
Mode 2
Mode 3
The Classical Modal ≅

Analysis Procedure
+ + +…
A Detailed 3D Elastic
Structural Model
Mode 1
Mode 2
Mode 3
The Uncoupled Modal Response ≅
History Analysis (UMRHA) F
F F
NL
Procedure D
+ D NL + D
NL +…
A Detailed 3D Inelastic
Structural Model
66
The Uncoupled Modal Response History Analysis (UMRHA) Procedure
Linearly Elastic Systems

• The classical modal analysis procedure for linearly elastic systems is equivalent to finding the response of the structure
to 𝑷𝑒𝑓𝑓, 𝑛 𝑡 for each 𝑛 and superposing the responses for all 𝑛.
• The response of the system to 𝑷𝑒𝑓𝑓, 𝑛 𝑡 is entirely in the 𝑛𝑡ℎ mode, with no contribution from other modes, which implies
that the modes are uncoupled.
• The equations governing the response of the linearly elastic MDF system to 𝑷𝑒𝑓𝑓, 𝑛 𝑡 ,
ሷ
𝑴𝒖(𝑡ሻ ሶ
+ 𝑪𝒖(𝑡ሻ + 𝑲𝒖(𝑡ሻ = −𝒔𝑛 𝑢ሷ 𝑔 𝑡
and the resulting floor displacements (using the idea of modal expansion from modal analysis) are
𝒖𝑛 𝑡 = 𝝓𝑛 𝑞𝑛 𝑡
Substituting this 𝒖𝑛 𝑡 in governing equation and pre-multiplying by 𝝓𝑇𝒏 leads to the equation governing the modal
coordinate 𝑞𝑛 𝑡 :
𝑞ሷ 𝑛 𝑡 + 2𝜉𝑛 𝜔𝑛 𝑞ሶ 𝑛 𝑡 + 𝜔𝑛2 𝑞𝑛 𝑡 = −𝛤𝑛 𝑢ሷ 𝑔 𝑡

Where 𝜔𝑛 is the natural frequency, 𝜉𝑛 is the damping ratio, and 𝛤𝑛 is the modal participation factor for the nth mode.
67
• As demonstrated in classical modal analysis, the solution of nth-mode equation of motion is
𝑞𝑛 𝑡 = 𝛤𝑛𝐷𝑛 𝑡
Where 𝐷𝑛 𝑡 is deformation response of the nth mode linearly elastic SDF system governed by
𝐷ሷ 𝑛 + 2𝜉𝑛 𝜔𝑛 𝐷ሶ 𝑛 + 𝜔𝑛2 𝐷𝑛 = −𝑢ሷ 𝑔 𝑡
Therefore,
𝒖𝑛 𝑡 = 𝛤𝑛𝝓𝑛 𝐷𝑛 𝑡
𝛥𝑗𝑛 𝑡 = 𝛤𝑛 𝜙𝑗𝑛 − 𝜙𝑗−1,𝑛 𝐷𝑛 𝑡
• The above equations represent the response of the MDF system to 𝑷𝑒𝑓𝑓 , 𝑛 𝑡 , and superposing the
responses for all 𝑛 gives the response of the system due to total excitation 𝑷𝑒𝑓𝑓 𝑡 :
𝑁
𝑟 𝑡 = ෍ 𝑟𝑛 𝑡
𝑛=1
• The UMRHA procedure for exact analysis of linearly elastic systems is identical to classical modal RHA.
But to derive these equations, now we have used the modal expansion of spatial distribution of effective
earthquake forces.
68
Inelastic Systems
• Although modal analysis is not valid for an inelastic system, its dynamic response can usefully be
discussed in terms of the natural vibration modes of the corresponding linear system.
• Each structural element of this linear system is defined to have the same stiffness as its initial stiffness in
the inelastic system; both systems have the same mass and damping. Therefore, the natural vibration
periods and modes of the corresponding linear system are the same as the vibration properties of the
inelastic system undergoing small oscillation, which are referred to as “periods” and “modes” of the inelastic
system.
• Thus, the modal expansion of effective earthquake forces is also valid for inelastic systems, where 𝜙𝑛 now
represents the modes of the corresponding linear system.
69
Inelastic Systems
• The equations governing the response of the inelastic MDF system to 𝑷𝑒𝑓𝑓, 𝑛 𝑡 are
𝑴𝒖(𝑡ሻ
ሷ + 𝑪𝒖(𝑡ሻ
ሶ + 𝒇𝒔 (𝒖ሻ = −𝒔𝑛𝑢ሷ 𝑔 𝑡
• The solution of this quation will no longer be described by 𝒖𝑛 𝑡 = 𝝓𝑛𝑞𝑛 𝑡 because modes other than the nth mode will
also contribute to the system response, implying that the vibration modes are now coupled.
• Thus, the floor displacements are given by:
𝑁
𝒖𝑛 𝑡 = ෍ 𝝓𝑟 𝑞𝑟 𝑡
𝑟=1
• However, because for linear systems 𝑞𝑟 𝑡 = 0 for all modes other than the nth mode, it is reasonable to expect that 𝑞𝑟 𝑡
may be small for inelastic systems, implying that the elastic modes are, at most, weakly coupled. Therefore,
𝑁
𝒖𝑛 𝑡 = ෍ 𝝓𝑟 𝑞𝑟 𝑡 ≃ 𝝓𝑛𝑞𝑛 𝑡
𝑟=1
70
Figure shows that the roof
displacement due to the force
vector 𝑷𝑒𝑓𝑓 , 𝑛 𝑡 is due primarily to
the nth mode but that other modes
contribute to the response. The
second, third, and fourth modes
start responding to excitation
𝑷𝑒𝑓𝑓 , 1 𝑡 the instant the structure
first yields.
Although the natural vibration
modes are no longer uncoupled if
the system responds
in the inelastic range, modal
coupling is weak.
71
• This weak coupling of modes implies that the structural response due to excitation 𝑷𝑒𝑓𝑓, 𝑛 𝑡 may be approximated b
𝒖𝑛 𝑡 ≃ 𝝓𝑛𝑞𝑛 𝑡
• Substituting this approximation into governing equation and pre-multiplying by 𝝓𝑻𝒏 gives
𝐹𝑠𝑛
𝑞ሷ 𝑛 (𝑡ሻ + 2𝜉𝑛 𝜔𝑛 𝑞ሶ 𝑛(𝑡ሻ + = −𝛤𝑛 𝑢ሷ 𝑔 𝑡
𝑀𝑛
where 𝐹𝑠𝑛 is a nonlinear hysteretic function of the nth modal coordinate 𝑞𝑛 :
𝐹𝑠𝑛 = 𝐹𝑠𝑛 𝑞𝑛 = 𝜙𝑛𝑇 𝒇𝑠 𝑞𝑛

• If the smaller contributions of other modes had not been neglected, 𝐹𝑠𝑛 would depend on all modal coordinates, and
the set of equations would be coupled because of yielding of the structure.
• By comparing the nth-mode governing equation of motion with linear counterpart, the solution can be expressed as
𝑞𝑛 𝑡 = 𝛤𝑛 𝐷𝑛 𝑡 , where 𝐷𝑛 𝑡 is now governed by
𝐹𝑠𝑛
𝐷ሷ 𝑛 (𝑡ሻ + 2𝜉𝑛 𝜔𝑛 𝐷ሶ 𝑛(𝑡ሻ + = −𝑢ሷ 𝑔 𝑡
𝐿𝑛
72
• 𝐷𝑛 may be interpreted as the deformation response of the nth-mode inelastic SDF system, an SDF defined by
1) small-oscillation vibration properties—natural frequency 𝜔𝑛 (natural period 𝑇𝑛 ) and damping ratio 𝜉𝑛 —of the nth
mode of the MDF system; and
2) The force–deformation (𝐹𝑠𝑛 /𝐿𝑛 − 𝐷𝑛 ) relation. Introducing the nth-mode inelastic SDF system permitted the
extension to inelastic systems of the well-established concepts for elastic systems.
• The solution of the nonlinear modal equation provides 𝐷𝑛 𝑡 , which can be substituted into following (same) equations
to obtain floor displacements and story drifts.
𝒖𝑛 𝑡 = 𝛤𝑛 𝝓𝑛 𝐷𝑛 𝑡
𝛥𝑗𝑛 𝑡 = 𝛤𝑛 𝜙𝑗𝑛 − 𝜙𝑗 −1,𝑛 𝐷𝑛 𝑡

• These equations approximate the response of the inelastic MDF system to 𝑷𝑒𝑓𝑓, 𝑛 𝑡 , the nth-mode contribution to
𝑷𝑒𝑓𝑓 𝑡 .
• Superposition of responses to 𝑷𝑒𝑓𝑓, 𝑛 𝑡 —n = 1 , 2 , . . . , N—provides the total response to p eff ( t ) .
𝑁
𝑟 𝑡 = ෍ 𝑟𝑛 𝑡
𝑛=1
• This is the UMRHA procedure for approximate analysis of inelastic systems.
73
To test the modal
uncoupling approximation
in UMRHA, the response
of the a 9-story building to
𝑷𝑒𝑓𝑓 , 𝑛 𝑡 = −𝒔𝑛 𝑢ሷ 𝑔 𝑡 is
determined by two
methods and compared.
Left: Roof Displacement

Right: Roof Drift
74
The UMRHA Procedure
Base shear
Vb
xr
Roof Displacement
Cyclic Modal Load
Cyclic Modal Pushover

MDOF
Displacement-related responses
m
ξi Fsi/Li
SDOF
Di
Idealized Hysteretic Model

Force-related responses
75
Shear wall’s steel bars begin Column’s bars begin to
to yield in compression yield in compression
0.09
Identification of force– Normalized Base Columns
Shear (𝑉𝑏1 /𝑊) begin to
deformation (𝐹𝑠𝑛/𝐿𝑛 − 𝐷𝑛 ) 0.07 Crack
Column’s steel bars
begin to yield in tension
relationship for each
0.05 Concrete crushing in
shear wall
significant mode
Shear wall’s steel reinforcement
0.03 bars begin to yield in tension
The Cyclic Pushover Analysis of an Shear walls begin to crack
Example Building under First-mode 0.01 Brick walls begin to crack
Inertia Load Pattern (𝑠𝑛 )
-0.04 -0.03 -0.02 -0.01 -0.01 0 0.01 0.02 0.03 0.04
44-story case study building in Strong Roof Drift (𝑥𝑖𝑟 /𝐻)

Direction -0.03
Cyclic Pushover
-0.05
-0.07
-0.09
76
0.06 0.08 0.1
19-story Building
20 33-story Building 44-story Building
0.06 0.08
0.04 Mode 1 Mode 1 Mode 1
0.06
0.04
0.04
0.02
0.02 0.02
0 0 0
-0.02 -0.02
-0.02
-0.04
-0.04
-0.06
-0.04
-0.06 -0.08
-0.06 -0.08 -0.1
-0.035 -0.015 0.005 0.025 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 -0.03 -0.02 -0.01 0 0.01 0.02 0.03
0.12 0.25 0.25
19-story
0.2 0.2
0.08
Mode 2 Mode 2 Mode 2
0.15 0.15
0.1
The Cyclic Behavior of 0.04 0.1
0.05
0.05
0 0
0
Some Example Buildings -0.04 -0.05
-0.05
-0.1
-0.1 -0.15
-0.08
-0.15 -0.2
-0.12 -0.2 -0.25
-0.025 -0.015 -0.005 0.005 0.015 0.025 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 -0.03 -0.02 -0.01 0 0.01 0.02 0.03
0.2 0.3 0.3
19-story
Mode 3 0.2 Mode 3 0.2
Mode 3
0.1
0.1 0.1
0 0 0
-0.1 -0.1
-0.1
-0.2 -0.2
-0.2 -0.3 -0.3

-0.025 -0.015 -0.005 0.005 0.015 0.025 -0.03 -0.01 0.01 0.03 -0.03 -0.02 -0.01 0 0.01 0.02 0.03
Roof Drift ( ) Roof Drift ( ) Roof Drift ( 77 )
Idealization of Cyclic Pushover Curves and Modeling of Nonlinear Modal SDF Systems
Normalized Base Shear (𝑉𝑏1 /𝑊)
𝐹𝑠𝑖 Τ𝐿𝑖 0.05
20-story Building
Unloading
starts
Mode 1
𝐹𝑠𝑢𝑖 /𝐿𝑖
Yield 2
𝐹𝑠𝑦𝑖 /𝐿𝑖 𝛽 𝐹𝑠𝑦𝑖
1 0.03
3
Compression
4 yielding of steel
Cracks bars
close
𝜔𝑖2
1
0.01
𝐷𝑖
𝐷𝑦𝑖 𝐷𝑢𝑖
Energy dissipated by
-0.0075 -0.005 -0.0025 0 0.0025 0.005 0.0075
hysteretic damping -0.01 Roof Drift (𝑥𝑖𝑟 /𝐻)
Cyclic Pushover Curve
-0.03 Response of an SDF system

under a ground motion
Convert 𝑉𝑏𝑛 − 𝑢𝑟𝑛 curves for each mode into 𝐹𝑠𝑛 /𝐿𝑛 − 𝐷𝑛
form. Assign 𝐹𝑠𝑛 /𝐿𝑛 − 𝐷𝑛 relations the nonlinear modal
SDF systems. -0.05
78
Idealization of Cyclic Pushover Curves for few Example Buildings
Normalized Base Shear ( ) Normalized Base Shear ( ) Normalized Base Shear ( )

0.06 0.08 0.1
Mode 1 (B1) Mode 1 (B2) Mode 1 (B3)
0.06 0.08
0.04
0.06
0.04
0.04
0.02
0.02
0.02
0 0 0
Cyclic pushover -0.02 -0.02
-0.02 curve
Idealized SDF -0.04
system -0.04
-0.06
-0.04
-0.06
-0.08
-0.06 -0.08 -0.1
-0.02 -0.01 0 0.01 0.02 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 -0.015 -0.005 0.005 0.015
Roof Drift Ratio ( ) Roof Drift Ratio ( ) Roof Drift Ratio ( )
Response of Idealized Modal SDF systems (red dotted lines) under an example ground motion is laid over the actual modal
cyclic pushover curves of the building. A good match indicates that the First-mode SDF systems can reasonably represent the
First-mode responses of the buildings.
79
Idealization of Cyclic Pushover Curves
0.12 0.2
0.07 Mode 2 (B1) Mode 2 (B2) Mode 2 (B3)
0.09 0.15
0.04 0.06 0.1
0.03 0.05
0.01
0 0
-0.02 -0.03 -0.05

Cyclic pushover
curve -0.06 -0.1
Idealized SDF
-0.05
system -0.15
-0.09
-0.08 -0.12 -0.2

-0.003 -0.002 -0.001 0 0.001 0.002 0.003 -0.003 -0.002 -0.001 0 0.001 0.002 0.003 -0.004 -0.002 0 0.002 0.004
Mode 2 - For an example Ground Motion Set
80
Idealization of Cyclic Pushover Curves
0.15 0.12 0.4

0.09 0.3
0.1
0.06 0.2
0.05
0.03 0.1
0 0 0
-0.03 -0.1
-0.05
Cyclic pushover
curve -0.06 -0.2
Idealized SDF
-0.1 system -0.09 -0.3
-0.15 -0.12 -0.4

-0.003 -0.001 0.001 0.003 -0.003 -0.001 0.001 0.003 -0.006 -0.004 -0.002 0 0.002 0.004 0.006
Mode 3 - For an example Ground Motion Set
81
0.05 0.06 0.15
0.04
0.04 0.1
0.03
0.02
0.02 0.05
0.01
0 0 0
-0.01 Cyclic pushover
curve -0.02 -0.05
-0.02 Idealized SDF
system
-0.03
-0.04 -0.1
-0.04
-0.05 -0.06 -0.15
-0.01 -0.005 0 0.005 0.01 -0.002 -0.001 0 0.001 0.002 -0.002 -0.001 0 0.001 0.002
The Idealization of Cyclic Roof Drift Ratio ( ) Roof Drift Ratio ( ) Roof Drift Ratio ( )
Pushover Curves Normalized Base Shear (

0.04
)
0.08 0.2
0.03 0.06 0.15
For an Example Ground 0.02 0.04 0.1
0.01 0.02 0.05

Motion Set 0 0 0
Cyclic pushover
-0.01 curve -0.02 -0.05
Idealized SDF
system
-0.02 -0.04 -0.1
-0.03 -0.06 -0.15
-0.04 -0.08 -0.2

-0.01 -0.005 0 0.005 0.01 -0.002 -0.001 0 0.001 0.002 -0.002 -0.001 0 0.001 0.002

0.06 0.15 0.3
0.04 0.1 0.2
0.02 0.05 0.1
0 0 0
Cyclic pushover
curve
-0.02 -0.05 -0.1
Idealized SDF
system
-0.04 -0.1 -0.2
-0.06 -0.15 -0.3

-0.005 -0.0025 0 0.0025 0.005 -0.004 -0.002 0 0.002 0.004 -0.006 -0.004 -0.002 0 0.002 0.004 0.006
Roof Drift Ratio ( ) Roof Drift Ratio ( ) 82
Roof Drift Ratio ( )
Modal Decomposition of 45 45
Mode 1
Nonlinear Responses 40
Mode 3 40
using the UMRHA 35 Mode 3
35
Procedure
30 Mode 2 30
Level No.
25 25 Mode 1
44-story case study building in 20 20

Strong Direction Envelope of Mode 2
Envelope of
15 Combined History 15
Combined History
Example Ground Motion Set
10 10
5 5
0 0
0 100 200 300 400 500 600 0 0.2 0.4 0.6 0.8
Peak Displacement (mm) Peak Inter-story Drift Ratio (%)
83
Modal Decomposition of 45 45
Nonlinear Responses 40 40
Mode 3
using the UMRHA 35 35 Envelope of
Combined History
Procedure 30 30
Mode 2
Level No.
Mode 2
25 25
Mode 1
44-story case study building in 20 20 Mode 1
Strong Direction Envelope of
15 Combined History 15
Example Ground Motion Set
10 10
Mode 3
5 5
0 0
0 15 30 45 0 0.5 1 1.5 2 2.5
Peak Story Shear (x 10 6 N) Peak Moment (x 10 6 KN m)
84
UMRHA vs. NLRHA
Displacement Envelope Inter-story Drift Ratio Story Shear Envelope Overturning Moment
45 45 45 45
40 40 40 40
35 35 35 35
30 30 30 30
No. of Stories
25 25 25 25
20 20 20 20
15 15 15 15
UMRHA
UMRHA(Combined 10
10 3(Combined
Modes) 3 Modes) 10 10
5 NLRHA
NLRHA 5 5 5
0 0 0 0
0 200 400 600 0 0.25 0.5 0.75 1 0 10 20 30 40 50 0 0.5 1 1.5 2
Displacement (mm) IDR (%) Story Shear (x106 N) Moment (x106 KN m)
Exampple 44-story case study building in Strong Direction - Ground Motion Set 4
85
Displacement (mm) Inter-story Drift Ratio (%) Story Shear (x 106 N) Moment (x 106 KN m)
20 20 20 20
15 15 15 15
Modal Decomposition of
Level No.
10 10 10 10
Nonlinear Responses
5 5 5 5
using the UMRHA B1
0 0 0 0
Procedure 0 500 1000 1500 0 0.5 1 1.5 2 2.5 0 10 20 30 40 50 0 0.2 0.4 0.6 0.8
35 35 35 35
30 30 30 30
25 25 25 25
Level No.
20 20 20 20
44-story case study building in
15 15 15 15
Strong Direction
10 10 10 10
5 5 5 5
Example Ground Motion Set B2
0 0 0 0
0 1000 2000 3000 0 0.5 1 1.5 2 2.5 3 0 15 30 45 60 0 0.5 1 1.5 2
45 45 45 45
40 40 40 40
NLRHA 35
35 35 35
UMRHA (Combined)
Level No.
30 30 30 30
Individual Modes 25 25 25 25
20 20 20 20
15 15 15 15
10 10 10 10
5 5 5 5
B3 0 0 0 0
0 500 1000 1500 2000 0 0.5 1 1.5 0 25 50 75 100 0 1 2 86
3
0.15
0.1 Ground Acceleration,
0.05
0
-0.05
Modal Decomposition of -0.1 Time (sec)
-0.15
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
Nonlinear Responses 600
400 Roof Displacement, Mode 1
using the UMRHA 200

0
-200
Procedure -400
Time (sec)
-600
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
150
100
Roof Displacement, Mode 2
50
44-story case study building in 0

-50
Strong Direction -100
Time (sec)
-150
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
Example Ground Motion Set 15
10
Roof Displacement, Mode 3
5
0
-5
-10
Time (sec)
-15
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
800
600 Roof Displacement, UMRHA Combined
400 NLRHA
200
0
-200
-400 Time (sec)
-600
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160
87
170
• For elastic systems → UMRHA = Classical Modal RHA (an exact analysis procedure).
• For inelastic systems → UMRHA = Approximate analysis procedure.
• The UMRHA for inelastic systems is based on two approximations
1) Superposition of responses (Strictly valid for only elastic systems. Approximately valid for inelastic systems)
2) Neglecting the coupling of modal coordinates, which permitted computing the response of inelastic MDF
system to 𝑷𝑒𝑓𝑓 ,𝑛 𝑡 from that of an SDF system. This approximation is reasonable only because the excitation
is 𝑷𝑒𝑓𝑓 ,𝑛 𝑡 , the nth-mode contribution to the total excitation 𝑷𝑒𝑓𝑓 𝑡 . It would not be valid for an excitation with
lateral force distribution different than 𝒔𝒏 [e.g., the total excitation 𝑷𝑒𝑓𝑓 𝑡 ], pointing out that the modal
expansion of effective earthquake forces is a key concept underlying the UMRHA.
88
The Modal Pushover Analysis (MPA) Procedure (Chopra and Goel, 2002)
89
The Modal Pushover Analysis (MPA) Procedure
• The response spectrum analysis (RSA) procedure, which is a dynamic analysis procedure, can be interpreted in two
ways: as static analysis or as pushover analysis.
a) Static analysis of the building subjected to lateral forces 𝒇𝑛 = 𝒔𝑛 𝐴𝑛 = 𝛤𝑛 𝑴𝝓𝑛 𝐴𝑛 will provide the same value
of 𝑟𝑛 , the peak value of the 𝑛𝑡ℎ -mode response 𝑟𝑛 (𝑡ሻ, as obtained from the RSA procedure. (Where 𝐴𝑛 =
𝐴 𝑇𝑛 , 𝜉𝑛 , the pseudo-acceleration spectrum ordinate corresponding to the natural vibration period 𝑇𝑛 and
damping ratio 𝜉𝑛 of the nth mode).
b) Alternatively, this peak modal response can be obtained by linear static analysis of the structure subjected to
monotonically increasing lateral forces with an invariant height-wise distribution: 𝒔∗𝑛 = 𝑴 𝝓𝑛 , pushing the
structure up to the roof displacement, 𝑢𝑟𝑛 .
𝑢𝑟𝑛 is the peak value of the roof displacement due to the nth mode, and is given be
𝑢𝑟𝑛 = 𝛤𝑛𝜙𝑟𝑛 𝐷𝑛
where 𝐷𝑛 ≡ 𝐷(𝑇𝑛 , 𝜉𝑛 ሻ is the ordinate of the deformation response spectrum corresponding to the period 𝑇𝑛 and
damping ratio 𝜉𝑛 of the nth mode.
90

• The peak modal responses, 𝒓𝒏 , each determined by one pushover analysis, can be combined according to the modal
combination rules (as used in the Response Spectrum Analysis, RSA [SRSS or CQC]) to obtain an estimate of the
peak value 𝒓 of the total response.
• Being equivalent to the standard RSA procedure, the MPA procedure offers no advantage for linearly elastic systems,
but this interpretation of RSA permits extension of MPA to inelastic systems.
• Note that 𝒓𝒏 determined by pushover analysis can also be interpreted as the peak response of the linearly elastic
system to 𝑷𝑒𝑓𝑓, 𝑛 𝑡 , the nth-mode component of the effective earthquake forces. This interpretation is valid because,
the system responds only in its nth mode when subjected to this excitation.
91
Inelastic Systems
• The peak response 𝑟𝑛 of the inelastic system to 𝑷𝑒𝑓𝑓 , 𝑛 𝑡 is also determined by a pushover analysis, which is now a
nonlinear static analysis instead of a linear static analysis, of the structure subjected to lateral forces distributed over
the building height according to 𝒔∗𝑛 with the forces increased to push the structure up to roof displacement 𝑢𝑟𝑛.
𝑢𝑟𝑛 = 𝛤𝑛 𝜙𝑟𝑛 𝐷𝑛
𝐷𝑛 is now the peak deformation of the nth-mode inelastic SDF system (instead of the nth-mode elastic SDF system).
𝐹𝑠𝑛
𝐷ሷ 𝑛 (𝑡ሻ + 2𝜉𝑛 𝜔𝑛 𝐷ሶ 𝑛 (𝑡ሻ + = −𝑢ሷ 𝑔 𝑡
𝐿𝑛
• At 𝑢𝑟𝑛 , the results of nonlinear static analysis provide an estimate of the peak value 𝑟𝑛 of the response quantity 𝑟𝑛 (𝑡ሻ:
floor displacements, story drifts, and other deformation quantities.
92
Inelastic Systems
• Nonlinear static analysis using force distribution 𝒔∗𝒏 leads to the nth-mode pushover curve, a plot of base shear
𝑉𝑏𝑛 versus roof displacement 𝑢𝑟𝑛 .
• From the nth-mode pushover curve is obtained the force–deformation (𝐹𝑠𝑛 /𝐿𝑛 − 𝐷𝑛 ) curve for the nth-mode inelastic
SDF system, which is required to determine 𝐷𝑛 from the following equation.
𝐹𝑠𝑛
𝐷ሷ 𝑛 (𝑡ሻ + 2𝜉𝑛 𝜔𝑛 𝐷ሶ 𝑛 (𝑡ሻ + = −𝑢ሷ 𝑔 𝑡
𝐿𝑛
• The forces and displacements in the two sets of curves are related as follows:
𝑢𝑟𝑛 𝐹𝑠𝑛 𝑉𝑏𝑛
𝐷𝑛 = = ∗
𝛤𝑛 𝜙𝑟𝑛 𝐿𝑛 𝑀𝑛
Where, 𝑀𝑛∗ = 𝐿𝑛 𝛤𝑛 is the effective modal mass.
93
At yield point,
𝐹𝑠𝑛𝑦
= 𝜔𝑛2 𝐷𝑛𝑦
𝐿𝑛
94
The Modal Pushover Analysis Procedure (MPA)
History Analysis (NLRHA)

Nonlinear Response
=
𝑥ሷ 𝑔 (𝑡ሻ
𝑷 𝑡 = −𝑴𝒍𝑥ሷ 𝑔 (𝑡ሻ
Multi-mode Pushover
𝑴𝝓𝟐 𝑴𝝓𝟑
Analysis
+ + +…
𝑴𝝓𝟏
95
UMRHA vs. MPA
• The response value 𝑟𝑛 determined by pushover analysis is an estimate of the peak value of the response 𝑟𝑛 (𝑡ሻ of the
inelastic structure to 𝑷𝑒𝑓𝑓 , 𝑛 𝑡 , but it is not identical to another estimate determined by UMRHA.
• For elastic systems UMRHA = Modal RHA, and MPA = RSA.
• For inelastic systems the two—UMRHA and MPA—estimates of the peak modal response are both approximate and
different from each other; the only exception is the roof displacement because it is deliberately matched in the two
analyses.
96
UMRHA vs. MPA Procedures
• The two estimates differ because the underlying analyses involve different assumptions.
• The UMRHA is based on the approximation contained in 𝒖𝑛 𝑡 ≃ 𝝓𝑛 𝑞𝑛 𝑡 , which is avoided in MPA because the
floor displacements, story drifts, and other deformation quantities are determined by nonlinear static analysis using
force distribution 𝒔∗𝒏 . As a result, the floor displacements of the inelastic system are no longer proportional to the nth-
mode shape, in contrast to 𝒖𝑛 𝑡 ≃ 𝝓𝑛 𝑞𝑛 𝑡 . In this sense, the MPA procedure represents the nonlinear behavior of
the structure better than UMRHA.
• However, the MPA procedure contains a different source of approximation, which does not exist in UMRHA. The
peak modal responses 𝑟𝑛 , each determined by one nonlinear static analysis, are combined by a modal combination
rule, just as in RSA of linearly elastic systems. This application of modal combination rules to inelastic systems lacks
a rigorous theoretical basis, but seems reasonable because the modes are only weakly coupled.
97
Thank you

Pushover Analysis Procedures

Uploaded by

Pushover Analysis Procedures

Uploaded by

CE 75.

03 – Earthquake Engineering for Tall Buildings

Dr. Fawad A. Najam Prof. Dr. Pennung Warnitchai

Session 1 (1.25 Hours):

Dr. Fawad A. Najam Prof. Dr. Pennung Warnitchai

Basic Idea of Pushover Analysis

Monotonically increasing lateral load Develop the 𝑉𝑏 vs. 𝑥 𝑟 relationship

Monotonically increasing lateral load Develop the 𝑉𝑏 vs. 𝑥 𝑟 relationship

• Start with Gravity Loads

• Select Lateral Load Patterns

What load pattern to apply?

Base Shear, V𝑉𝑏

Column’s steel bars begin to yield in tension

Base Shear/Building Weight (VB/W)

Concrete crushing in shear wall

Shear wall’s steel bars begin to yield in compression

0.04 Columns begin to Crack

Roof Drift (Top Displacement/Building Height)

Monotonic Pushover Concrete crushing in

Monotonic Pushover in Positive Direction

Some Example Buildings 0

-0.2 -0.3 -0.3

Full 3D Nonlinear Monotonic Pushover Determination of Idealization of An Equivalent

Converting a Nonlinear MDF System into an Equivalent NL SDF System

Cyclic Pushover Curve of an MDF Converted to Force vs. Displacement

Normalized Base Shear (𝑽𝒃 /𝑾) Force

Roof Drift (𝒙𝒓 /𝑯) Displacement

Assumed Hysteretic Rule

Equivalent • Several Equivalent Linearization Procedures (Individual studies)

The basic idea of 𝐹 𝐹

𝑉𝑏1 Determination of Responses

Energy dissipated by hysteretic damping

𝑇𝑒𝑞 and 𝛽𝑒𝑞 are the

Taken from FEMA 440 (2005)

• Step 1: Construct a bilinear representation of the capacity curve.

• Step 2: Calculate the effective fundamental period (𝑇𝑒 ) as;

• Step 3: Calculate the target displacement (𝛿𝑇 ሻas:

For converting elastic to To account for dynamic

For converting spectral “Effective” period

To account for the shape Spectral acceleration at 𝑇𝑒

Strength and Stiffness-Degrading (SSD) Nonlinear Elastic (NE)

𝐶1 is on average larger than 1 𝐶1 is approximately constant with changes in 𝑇

𝑉𝑏1 Determination of Responses

Full 3D Nonlinear Monotonic Pushover Determination of Idealization of An Equivalent

Dr. Fawad A. Najam Prof. Dr. Pennung Warnitchai

Lateral Load Vectors

Simple (Uniform, Triangular etc.) First Modal Inertia Multi-mode

Modal Combination Method (Kalkan and Kunnath, 2004)

Analysis Procedures Multimode Pushover Analysis (Sasaki et al., 1996)

Consecutive Modal Consecutive Modal Pushover Analysis (Poursha et al., 2009)

Adaptive Modal Pushover Analysis (Kalkan and Kunnath, 2006)

𝑴𝒖ሷ 𝑡 + 𝑪𝒖ሶ 𝑡 + 𝑲𝒖(𝑡ሻ = −𝑴 𝟏 𝒖ሷ 𝒈 𝒕 = 𝑷𝒆𝒇𝒇 𝒕

Where the effective earthquake forces are

• The effective earthquake forces can then be expressed as

𝑷𝑒𝑓𝑓 𝑡 = ෍ 𝑷𝑒𝑓𝑓 , 𝑛 𝑡 = ෍ −𝒔𝑛 𝑢ሷ 𝑔(𝑡ሻ

• The contributions of the 𝑛𝑡ℎ mode to 𝑷𝑒𝑓𝑓 𝑡 and 𝒔 are

𝑷𝑒𝑓𝑓 , 𝑛 𝑡 = −𝒔𝑛 𝑢ሷ 𝑔(𝑡ሻ

The Classical Modal ≅

Linearly Elastic Systems

𝑞ሷ 𝑛 𝑡 + 2𝜉𝑛 𝜔𝑛 𝑞ሶ 𝑛 𝑡 + 𝜔𝑛2 𝑞𝑛 𝑡 = −𝛤𝑛 𝑢ሷ 𝑔 𝑡

𝐹𝑠𝑛 = 𝐹𝑠𝑛 𝑞𝑛 = 𝜙𝑛𝑇 𝒇𝑠 𝑞𝑛

𝛥𝑗𝑛 𝑡 = 𝛤𝑛 𝜙𝑗𝑛 − 𝜙𝑗 −1,𝑛 𝐷𝑛 𝑡

Left: Roof Displacement

Cyclic Modal Pushover

Idealized Hysteretic Model

44-story case study building in Strong Roof Drift (𝑥𝑖𝑟 /𝐻)

-0.2 -0.3 -0.3

Cyclic Pushover Curve

-0.03 Response of an SDF system

Normalized Base Shear ( ) Normalized Base Shear ( ) Normalized Base Shear ( )

0.04 0.06 0.1

-0.02 -0.03 -0.05

-0.08 -0.12 -0.2

Mode 2 - For an example Ground Motion Set

0.15 0.12 0.4

-0.15 -0.12 -0.4