1.

Question:

In seismic design provisions in the U.S. and Canada, response modification factors (R in ASCE
7 and RdRo in NBCC 2005) are used when base shear force is calculated. Explain in detail what is the
intent of applying response modification factors and how the factors are affected by ductility of a
lateral load resisting system.

Answer:

Figure: Inelastic force deformation curve

Figure shows the inelastic force deformation curve for a building. V e is the elastic base shear for the
building. Vs and Vy is the design base shear and inelastic base shear capacity of the building respectively.
Response modification factor R is the ratio elastic bas shear to design base shear.

The Response modification factor "R” accounts for the ductility of the building and adjusts the design
lateral loads accordingly. Studies of existing buildings during earthquakes have demonstrated that ductile
buildings (like flexible moment frames) perform much better in seismic events than rigid buildings (like
masonry shear wall buildings) because of the inherent ability of flexible systems to dissipate the energy of
the ground motion.

The main intent to use response modification factor, R is

 To produce ductile failure instead of brittle failure. The ductile failure is assured by increasing
energy dissipation capacity of the building. The energy dissipation capacity can be increased by
detailing of the beam column joint.
 The ductile building will provide more warning before collapse.
 It makes a balance between cost and benefit.

The response modification factor, R largely affected by the ductility of the lateral load resisting system.

Ductility is the ratio of the maximum roof displacement of the building to the yield displacement of the
building. The system has more ductility that means it has more inelastic energy dissipation capacity. The
basic concept equivalent static analysis is that the total energy for an elastic system and an inelastic
system will be constant. So if the ductility decrease the response modification factor will decrease and if
the ductility increase the response modification factor will be increased.

Page 1 of 17
 2. Question:

Explain the concept of capacity design and how it is achieved using the over-strength factor in
ASCE-7.

Answer:

To explain the capacity design approach we can consider a chain made of glass rings and hence brittle,
and one ring is made of steel and hence ductile. Suppose the chain is tauted by a force P.

In order to get a ductile chain the glass ring needs to be more resistant than the steel one. Hence the
design force for the steel ring will be equal to P, but for the glass ring, that has to be in the elastic range,
the design force will be equal to the resistance of the steel ring, amplified by an appropriate safety factor:
the over strength factor Ro .

Psteel = P

Pglass = Ro · Psteel, Ro> 1

If the strength of the steel ring is lower than the glass ring one, the behaviour of chain will be ductile. In
fact the steel ring is able to stretch a lot before breaking. If the strength of the steel ring is higher than the
glass ring one, the behaviour of chain will be brittle. In fact the glass ring breaks immediately after
reaching its strength force.

The over strength factor is used to ensure that the resistance of the brittle element is always greater than
the ductile one, in order to achieve a global ductile failure of the structure.

To ensure the ductile failure ASCE-7 states that for seismic design category (SDC) D-F buildings having
irregularities, the design forces determined from the equivalent lateral force procedure must be increased
25 percent for connections of diaphragms to vertical elements and to collectors, and for connections of
collectors to the vertical elements of the seismic force-resisting system. The listed irregularities are:
torsional irregularity, extreme torsional irregularity, re-entrant corners, diaphragm discontinuity, out-of-
plane offsets of the vertical lateral force-resisting elements and in-plane discontinuity in vertical lateral
force-resisting elements. This increase of lateral force is well known as over strength factor.

Page 2 of 17
 3. Question:

Through nonlinear analysis, we try to replicate the real behaviour of structural elements and
system. In this process, it is critical to understand how numerical elements are formulated and how the
input parameters are calibrated if calibration is needed. Clearly explain what it means by ‘verification’
and ‘validation’ processes which are key steps in numerical element development.

Answer:

Verification is the process of determining that a model implementation accurately represents the
developer’s conceptual description of the model and the solution to the model.

Validation is concerned with quantifying the accuracy of the model by comparing numerical solutions to
experimental data

Verification Validation

Making sure it works Making sure it does what it’s supposed to do.

Getting the math right. Getting the physics right.

Providing an accurate FE analysis. Checking the FEA against test or experimental data.

4. Question:

Fiber-section elements are widely used for nonlinear performance assessment of frame structures.
Discuss underlying assumptions in fiber-section elements and limitations.

Answer:

Assumption:

 The plane section remains plane before and after bending.

 Plasticity is uniformly distributed throughout the element.

Limitations:

It can capture behaviours such as initiation of concrete cracking and steel yielding but it has the following
limitations:

 Fiber section fails to capture strength degradation such as reinforcing bar buckling, bond slip,
and shear failure, dowel action, and spalling in reinforced concrete,
 It fails to account the connection behaviour, local instabilities, and post-buckling behaviour in
steel and
 Cyclic deterioration of the model.

Page 3 of 17
 5. Question:

When we build a numerical model of a building structure we often represent a three-dimensional structure
with two-dimensional frame if torsional response can be negligible. For perimeter moment resisting
frames, for example, we can just model one of the moment resisting frames with Diaphragm constraint in
each floor and leaning (gravity) columns with gravity loads. Explain why we need to use diaphragm
constraints and leaning (gravity) columns.

Answer:

Diaphragm constraints:

A diaphragm constraint

 Creates links between joints located within a plane such that they move together as a planar
diaphragm,
 Rigid against membrane (in-plane) deformation, but susceptible to plate (out-of-plane)
deformation and associated effects.
 Diaphragm constraints relieve numerical accuracy problems which result when floor diaphragms
are modeled with very high in-plane stiffness.
 They also enhance the computational efficiency of dynamic lateral analysis by reducing the size
of the eigenvalue formulation.
 To distribute the lateral load

Leaning (gravity) columns:

A leaning column carrying gravity loads is linked to the 2D frame to simulate P-Delta effects. These
columns with pinned ends are commonly referred to as “leaning columns.” When modeling the frame,
leaning-column effects can be captured either by developing a complete 3D model of the frame or by
assigning a single equivalent leaning column carrying the summation of all of the gravity loads on all of
the leaning columns in the structure, as a pin-connected part of a 2D frame. An example of how this might
be modeled in a 2D analysis is shown in Figure. The leaning columns are modeled as Elastic Beam-
Column Elements. These columns have moments of inertia and areas about two orders of magnitude
larger than the frame columns in order to represent aggregate effect of all the gravity columns. Truss
Elements are used to link the frame and leaning columns and transfer the P-Delta effect. The trusses
have areas about two orders of magnitude larger than the frame beams in order to represent aggregate
effect of all the gravity beams and can be assumed to be axially rigid.

Figure: 2D frame elements that can capture leaning column effects

Page 4 of 17
 6. Question:

Several analysis methods have been proposed and adopted for seismic performance assessment.
Since structures are designed to behave in inelastic regime, the analysis for performance assessment is
much more complicated than the linear elastic analysis which is most widely used in the design process.
Inelastic analysis methods can be largely grouped as nonlinear static procedure (pushover analysis
combined with capacity spectrum method) and nonlinear dynamic procedure (nonlinear time history
analysis). Explain the 1) assumptions in the capacity spectrum method (nonlinear static
procedure), 2) the limitations of the method due to the assumptions, and 3) other proposed
methods available in literature to overcome the limitations.

Answer:

Assumptions in the capacity spectrum method:

The nonlinear pushover analysis has no rigorous theoretical foundation.

 It is based on the assumption that the response of the structure can be related to the response of
an equivalent of single degree of freedom system (SDOF)
 The shape of this mode remains constant throughout the time history response.
 It assume a bilinear relationship to form an acceleration displacement relation of the structure

Limitations in the capacity spectrum method:

 Force distribution
o Dynamic property of a structure changes once a structure yields. Hence, the distribution
of inertial force is also changed depending on the inelasticity of the system.
o In the conventional pushover analysis, higher mode effect is not considered.
 Conversion from capacity curve to capacity spectrum
o Roof displacement is arbitrarily chosen as a measure of displacement.
o The conversion of roof displacement to spectral displacement is not theoretically correct
once a structure yields.
 Dynamic loading history
o Nonlinear structural behaviour is load path dependent. In CSM, the effects of number of
inelastic loading cycles on strength or stiffness degradation are not considered. (Pinching
effect, etc. are not fully considered).

 For taller structures where higher modes effects are more important it provides incorrect results.
 The formation of bilinear relation for the structure a sound and practical knowledge is required.

Other proposed methods available in literature to overcome the limitations:

 Methods to improve the accuracy

 Consideration of higher mode effects

o Summary
 Combination of inelastic response of higher modes using modal combination rule.
 To find the performance point of each mode, nonlinear response history analysis is
carried out based on the pushover curve.
 Practically accurate but theoretically not rigorous.
 Higher-mode pushover curve may be unrealistic.

Page 5 of 17
  Energy based pushover analysis
o Summary:
 Realistic representation of inelastic pushover curve of higher modes.
 Eliminate arbitrary selection of roof drift in pushover analysis.
 Can be combined with modal pushover analysis.
 Defines only pushover curve, not seismic demand. The accuracy in performance
assessment will also depend on how the performance points are calculated.
 Pushover results combination
o Moghadam and Tso [2002]

 Consideration of variation of the modal properties during the inelastic process –

Adaptive pushover
o Summary:
 Variation of mode shapes during failure process is considered in load pattern.
 The load pattern includes higher mode effect.
 Defines only pushover curve, not seismic demand. The accuracy in performance
assessment will also depend on how the performance points are calculated.
 Consider higher mode effects in the load pattern (Update the load pattern at every time
step)
o Papanikolaou [2000], Antoniou [2003], and Elnashai [2002]

7. Question:
The energy dissipation in a structural system is typically modelled with the equivalent viscous
damping (i.e. Rayleigh damping). Discuss the potential unintended consequences of using
Rayleigh damping for the analysis of an inelastic system, and recently proposed methods
to overcome the limitation.

Answer:

It is observed from the recently published articles, that the vibration response of structures is always
related with the phenomenon of damping. However, appropriate quantification of damping ratio is very
important for proper assessment and design of buildings to minimize damages. The equivalent viscous
damping is most widely used to model the inherent energy dissipation in structures. It includes a lot of
uncertainties even when a structure behaves in the linear elastic range. Once a structure behaves in the
inelastic range, the stiffness proportional damping may lead to excess energy dissipation because of
energy dissipation involves with the inelastic hysteretic behaviour.

To overcome this limitation currently researcher proposed layout of dashpot dampers, non-viscous
damping, and viscous damping that have the ability to incorporate nonlinearity of the structures.

Page 6 of 17
 8. Question:

Difference between frequency domain and time domain

Answer:

The time domain is the domain in which all the signals are represented. Time domain signal can be tested
or verified with the use of oscilloscope. In time domain signals are represented by amplitude on Y axis
and time on X axis.

The frequency domain is useful to do deeper analysis of the time domain signal. Frequency domain helps
study frequency contents of the discrete time domain signals as well as continuous time domain signal.
The frequency domain signal can be analyzed with the use of spectrum analyzer.

In frequency domain signals are represented by power (amplitude2) on Y axis and frequency on X axis.
Time domain signal can be converted to Frequency domain signal with the use of Discrete Fourier
Transform or Fast Fourier Transform (FFT).

Time domain signal is useful to perform nonlinear analysis

Frequency domain signal is useful to perform linear analysis.

Time domain shows the every point for the time interval considered.

Frequency domain shows the peak point only.

9. Question:

Difference between implicit method and explicit method

Answer:

The explicit method and implicit method are numerical analysis methods used to solve a time-dependent
differential equation.

The explicit method calculates the system status at a future time from the currently known system status.
The implicit method calculates the system status at a future time from the system statuses at present and
future times.

For example, when there is a differential equation , the explicit method expresses it as

That is, if you know the state at n, you can calculate the state at n+1.

On the contrary, the implicit method has the state at n+1 on the right-hand side as in
.

The explicit method is easier to program and can be calculated within a shorter time. But its stability is so
low that you need to use a step size small enough to prevent divergence.

Page 7 of 17
On the contrary, the implicit method has high stability and converges if you set proper parameters. But, as
you need to solve an equation at every step, it takes a long time to calculate.

As the implicit method can use a sufficiently large step size, it is suitable for solving equations that involve
a long time. Also, in non-linear equations such as contact, it is difficult to predict a future from the past
state. So, in these cases, it is recommended that you use the implicit method rather than the explicit
method.

10. Question:
Limitation of Linear Static Procedure?
Answer:

 Force distribution is based on approximate 1st mode.

 Higher mode effects cannot be considered.
 Components should behave close to linear elastic range.

It is not applicable for the following conditions:

 The fundamental period of the building, T, is greater than or equal to 3.5 times T s.
 The ratio of the horizontal dimension at any story to the corresponding dimension at an
adjacent story exceeds 1.4 (excluding penthouses).
 The building has a torsional stiffness irregularity in any story. [1.5 times]
 The building has a vertical stiffness irregularity. [1.5 times]
 The building has a non-orthogonal seismic-force-resisting system.

11. Question:

Numerical modelling of nonlinear structural system

Answer:

The nonlinearity comes from the following

 Geometric non linearity,

 Material non linearity and
 Nonlinearity from the boundary condition.
Geometric nonlinearity takes in to consideration in numerical modelling by considering P-∆ effect.
The element level nonlinearity also take into consideration by assigning plastic hinge or elements
that can consider the plastic formation. Beam column connection plays a vital source for non
linear behaviour of the structure. This needs to assign proper way so that it can take in to
consideration of the exact behaviour for the structure.
Material nonlinearity takes in to consideration by assigning hysteretic behaviour of the material
under cyclic load.
Nonlinearity arises from boundary condition also take into consideration during numerical
modelling.

Page 8 of 17
 12. Question:

Advantages and disadvantages of hybrid simulation.

Answer:

Advantages:

To get the exact response of the structure experimental testing is mandatory. In reality civil engineering
structure is very large and it carries huge forces. It is difficult to do experiment for these large structures. It
needs large reaction frame but if we do hybrid simulation (experimental +analytical) it is possible to
understand the real behaviour of the structure under dynamic loads. It hybrid simulation substructure is
tested in the laboratory and the full structure is modeled in analytically. The response of the substructure
in the experiment connects with network to the analytical model. The response of the analytical model
observed under the experimental result.

Disadvantages:

High speed computer network is required. Identify the critical element is challenging. Accuracy of the
experimental equipments largely affects the performance of the structure.

13. Question:

In the seismic design of building structures, the maximum base shear is reduced by the strength
reduction factor, R. The value of the strength reduction factor, R, is related to the inherent ductility of the
system. Similarly, in the coefficient method, the maximum seismic demand is evaluated based on the
relationship between ductility, strength reduction factor, and the period of the structural system (i.e. R-mu-
T relationship). Explain in detail what the R-mu-T relationship is by defining each term. Explain how the
relationship can be derived, general shape of the curves, how it is used for the ‘approximate’ demand
assessment of inelastic system, and inherent uncertainty of the relationship.

Answer:

14. Question:

Several analysis methods have been proposed and adopted for seismic performance assessment. Since
structures are designed to behave in inelastic regime, the analysis for performance assessment is much
more complicated than the linear elastic analysis which is most widely used in the design process. The
inelastic analysis methods can be largely grouped as nonlinear static procedure (pushover analysis
combined with capacity spectrum method) and nonlinear dynamic procedure (nonlinear time history
analysis). Explain the advantages and limitations of nonlinear dynamic procedure.

Answer:

Page 9 of 17
 15. Question:

Answer one of the following questions. (You can choose only one.) a. What does it mean by the
convergence criteria in an inelastic analysis? If your analysis shows diverging solution, what would you do
to obtain a converging solution? Will it be possible to obtain converging solution always? Why (or why
not)?

Answer:

a) In the inelastic analysis explicit and implicit method are used. In both methods time integration
method is used to reach the desired solution. In every time steps the change of force,
displacement or energy change shall be larger than a limiting value. If the change of the
displacement, force or energy is less than the limiting value that’s mean it is satisfying the
convergence criteria otherwise it is diverging.
Displacement, unbalance force and energy or a combination of them is used as convergence
criteria.
Unbalance forces are robust criteria that used as a convergence criteria.
If the analysis shows diverging solution than to get e convergence solution need to check the
following:
Time increment, mesh of the model, stability of the model, stiffness of the model, convergence
criteria.
It is not possible to obtain converging solution always:
 Bad starting points
 Iteration point is stationary
 Starting point enters a cycle
 Derivative issues
 Derivative does not exist at root
 Discontinuous derivative
 Zero derivatives
b) By modelling structural components with nonlinear elements, you are trying to capture realistic
behaviour of the element. How would you ascertain that the modelling parameters for the
nonlinear element are correct?
Answer:
To determine the nonlinear behaviour it is mandatory to use the nonlinear element in the
modelling. The nonlinear element needs to verify and validate to ascertain that the parameters
are correct.

Verification is the process of determining that a model implementation accurately represents the
developer’s conceptual description of the model and the solution to the model.

Validation is concerned with quantifying the accuracy of the model by comparing numerical solutions to
experimental data

Verification Validation

Making sure it works Making sure it does what it’s supposed to do.

Page 10 of 17
Getting the math right. Getting the physics right.

Providing an accurate FE analysis. Checking the FEA against test or experimental data.

c) Time history analysis of a linear elastic system may result in unstable response. What could be
the source of the stability issue? How would you resolve the stability issue?
Answer:

16. Question:

If one wants to evaluate seismic performance of a structure using nonlinear response history analyses, it
is necessary to select a set of ground motions which represent seismic hazard of a site. What criteria
should be used when ground motions are selected and scaled? And what are the practical limitations
when using recorded ground motions? (Use following keywords in your explanation: seismic hazard
desegregation, structural periods, uniform hazard response spectra, spectral acceleration, inelasticity of
structures)

Answer:

The parameters that need to be considered in identifying the scenario conditions are those that have the
most influence on ground motion spectral shape:

 Site specific hazard spectrum

 Earthquake magnitude
 Distance from the source of energy release
 Local geology and Site-condition (site-geology generally described by average shear-wave
velocity within 30 m)
 Source mechanism and characteristics
 Wave travel path
 Rupture directivity
 Structure with short period of time affected by acceleration
 Structure with intermediate time period affected by velocity and
 Structure with long period affected by displacement

Practical limitations when using recorded ground motions

 It is difficult to get site specific ground motion records.
 Small variation of ground motion records may produce abnormal results.
 Matching with target response spectrum is not all about ground records.
 There is no specific guideline about duration of records

17. Question:

Discuss pros and cons of nonlinear static analysis methods (i.e. capacity spectrum method or coefficient
method) in comparison with nonlinear response history analysis method.

Page 11 of 17
Answer:

Pros in the nonlinear static analysis method:

 The pushover analysis is an effective tool for the performance evaluation of a structural
system, by estimating its strength and deformation demands induced during a seismic event,
by means of a static nonlinear analysis.
 The demands are then compared to available capacities at the performance levels of interest.
 The evaluation is based on assessment of important performance parameters such as global
drift, inter-story drift and inelastic element deformations.

 The pushover analysis can be viewed as a methodology for predicting seismic force and
deformation demands, which can account for, in an approximate manner, the redistribution of
internal forces occurring when the structure is subjected to inertia forces that no longer can
be resisted in the elastic range.
 The design is achieved by controlling the deformations in the structure;

 Consideration of the non-linear behaviour, which avoids the use of behavior coefficients
(reduction factors), that can not be rigorously assessed;

 Allows tracing the sequence of yielding and failure on the member and the structure levels,
as well as the progress of the overall capacity curve of the structure;

 Its applicability to performance-based seismic design approaches as it permits different

design levels to verify the performance targets.

Cons in the nonlinear static analysis method:

 Force distribution
o Dynamic property of a structure changes once a structure yields. Hence, the distribution
of inertial force is also changed depending on the inelasticity of the system.
o In the conventional pushover analysis, higher mode effect is not considered.
 Conversion from capacity curve to capacity spectrum
o Roof displacement is arbitrarily chosen as a measure of displacement.
o The conversion of roof displacement to spectral displacement is not theoretically correct
once a structure yields.
 Dynamic loading history
o Nonlinear structural behaviour is load path dependent. In CSM, the effects of number of
inelastic loading cycles on strength or stiffness degradation are not considered. (Pinching
effect, etc. are not fully considered).

 For taller structures where higher modes effects are more important it provides incorrect results.
 The formation of bilinear relation for the structure a sound and practical knowledge is required.

Page 12 of 17
 18. Question:

In one-dimensional site response analysis of soil, equivalent linear analysis has been
predominantly used. Explain what it means by modulus reduction and damping curves, and the
procedure to run equivalent linear analysis.

Answer:

Page 13 of 17
Page 14 of 17
 19. Question:

In the site response analysis of a soil domain (i.e. soil column), frequency domain analysis has
been typically used. If the soil behaves largely in inelastic range, it is necessary to run time
domain analysis because the governing differential equation (i.e. equation of motion) needs to be
solved with a step by step procedure. The time domain analysis method, however, may result in
different response from the frequency domain analysis even if the soil behaves in the elastic
range. Discuss why the responses are different, and a method to improve the accuracy of the
time domain analyses in the linear elastic range.

Answer:

Page 15 of 17
 20. Question:

In one-dimensional site response analysis of soil, equivalent linear analysis has been
predominantly used. Explain what it means by modulus reduction and damping curves, and the
procedure to run equivalent linear analysis.

Answer:

1. Dynamic behavior of soils is quite complex and requires models which characterize the important
aspects of cyclic behavior, but need to be simple, rational models.
2. Three classes of dynamic soil models: a) equivalent linear (SHAKE and DEEPSOIL) b) cyclic
nonlinear (DEEPSOIL) c) advanced constitutive (DEEPSOIL and FLAC) vertically 1-D
propagation of shear waves in a multi-layered system is assumed in EQL method. a. EQL method
produces an approximation to the nonlinear response of soils under earthquake loading, but is
very efficient computationally. b. In the EQL method, the nonlinear stress strain loop is
approximated by a single equivalent linear secant shear modulus that is a function of the amount
of shear strain. c. Iteration is required to determine the appropriate equivalent secant shear
modulus Geq that is compatible with the amount of strain that develops during the modeling
process. d. The equivalent damping is determined from strain-controlled laboratory tests and is
defined as a function of the shear strain level and such damping is used in the modeling process.
e. Because the EQL method is fundamentally a damped linear elastic method using strain
compatible secant shear modulus and the associated damping, it cannot be used directly to solve
problems involving permanent shear deformation because it does not calculate permanent strain.
Because the EQL model does not follow the actual hysteresis loops, the final shear strain is zero
after cycling has stopped with no residual permanent shear strain. f. Also, because it is a linear
elastic model, there is no limiting value for the shear strength of the soil (no failure criterion
required), so failure, or yielding, is not allowed in the model.

Page 16 of 17
Page 17 of 17