Lateral Load Distribution in Nonlinear Static Procedures For Seismic Design

Lateral load distribution in nonlinear static procedures for seismic design
E. Kalkan1 and S. K. Kunnath2

1
University of California Davis, Department of Civil and Environmental Engineering,

One Shields Avenue, 2001 Engineering III, Davis, CA 95616; PH (530) 754-4958;
FAX (530) 752-7872; email: ekalkan@ucdavis.edu
2
University of California Davis, Department of Civil and Environmental Engineering,
One Shields Avenue, 2001 Engineering III, Davis, CA 95616; PH (530) 754-6428
FAX (530) 752-7872; email: skkunnath@ucdavis.edu
Abstract
Nonlinear static analysis using pushover procedures are becoming increasingly
common in engineering practice for seismic evaluation of building structures. Various
invariant distributions of lateral forces are recommended in FEMA-356 (2000) to
perform a pushover analysis. However, the use of these invariant force distributions
does not adequately represent the effects of varying dynamic characteristics during
the inelastic response or the influence of higher modes. More recently, new
approaches to combining lateral load distributions have been proposed to overcome
some of the drawbacks in FEMA procedures. In this paper the validity and
applicability of several lateral load configurations are assessed by comparison of the
pushover response of eight and twelve story steel moment frame buildings with
benchmark solutions based on nonlinear time history analyses. The study reveals the
suitability of using unique modal combinations to determine lateral load
configurations that best approximate the inter-story demands in multistory frame
buildings subjected to seismic loads.
Introduction
Although current seismic design practice is still governed by force-based design
principles, a common trend in structural earthquake engineering practice is to use
performance-based seismic evaluation methods for the estimation of inelastic
deformation demands in structural members. A widely used and popular approach to
establish these demands is a pushover analysis in which a model of the building
structure is subjected to an inverted triangular distribution of lateral forces. While
such a load distribution may be adequate for regular and low-rise structures whose
response is primarily in their fundamental mode, it can produce misleading results for
structures with significant higher mode contributions. This accentuates the need for
improved procedures that addresses current drawbacks in the lateral load patterns
used in pushover analyses. New lateral load configurations using modal combinations
originally proposed by Kunnath (2004) and some variations of the approach are
investigated in this paper. In all cases, the computed peak response is compared to
FEMA-based patterns and to results from nonlinear time-history analyses.
1
Copyright ASCE 2004
Structures 2004
Downloaded 14 Nov 2005 to 169.237.215.179. Redistribution subject to ASCE license or copyright, see http://www.ascelibrary.org/
Description of Buildings Used in Evaluation

Two special moment resisting steel frame buildings were selected as representative
case studies to carry out the evaluation of different later load distributions. The
building designs are based on a configuration presented in the SEAOC Seismic
Design Manual (SEAOC, 2000). The original design presented in the manual pertains
to a four-story building. The buildings lateral force resisting system is composed of
steel perimeter moment resisting frames (MRF).
1
6
12th Floor
11th Floor
10th Floor
9th Floor
11@ 4.1 (13.5')
8th Floor
7th Floor
6th Floor
5th Floor
4th Floor
3rdFloor
2ndFloor
1st Floor
4.5 (15')
3@8.53 (28')
5.94 (19.5')
5.94 (19.5')
(a) Elevation along line A

A
1
8-Story
2
3
12-Story
4
5
6
20.4 (67')
15 (49')
Levels
7 8
56
34
12
Beam Section
W21x73
W24x94
W27x102
W27x114
Column Section
W14x132
W14x159
W14x211
W14x283
11 12
9 10
78
5 6
34
12
W27x94
W27x102
W27x114
W30x124
W30x132
W30x148
W14x132
W14x193
W14x257
W14x311
W14x370
W14x426
(c) Member sizes
Moment resisting connection

Simple hinge connection
(b) Plan view

Figure 1. Structural details of 8 and 12 story buildings (units in meters)
2
Copyright ASCE 2004
Structures 2004
In this study, the same floor plan was extended to eight and twelve stories. These
buildings are 37.5 m (123) by 56 m (200) in plan and 33.4 m (109.5) and 50 m
(163.5) in elevation for eight and twelve story cases, respectively. In the north-south
(N-S) direction the interior bays are 8.5 m (28) and exterior bays are 6m (19.5) with
a total of five bays. The floor-plan and elevation view of the building are illustrated in
Figure 1. Also shown in this figure are the member sizes for the 8 and 12 story
buildings.
The design roof dead load is 939 tons (2066 kips) and the dead load of each of the
remaining floors is 1016 tons (2235 kips). The yield strength of steel is assumed to
be fy = 345 Mpa (50 ksi) for all structural members. Since the plan of structures is
essentially symmetric, only a single frame in the transverse direction (line A in Figure
1) is analyzed. Composite action of floor slabs is not taken into consideration. Since
the intent of the SEAOC design manual is to simply illustrate the design process, the
final design presented in the manual is not optimized. In the present study, the
SEAOC design was modified to result in optimized member sizes that conform to the
requirements of the UBC (1997) provisions. The design base shears for the eight and
twelve story structures are approximately 4.8% of their respective building weights.
Ground Motions
In order to establish a benchmark response to examine the validity of the different
pushover procedures based on invariant load distributions, nonlinear time-history
analyses were performed on the same set of buildings. The seismic excitation used for
nonlinear time history evaluations is defined by a set of seven strong ground motions.
These ground motion records are recommended by ATC-40 (1996). All ground
motions were recorded from California earthquakes having a magnitude range of 6.6
to 7.5 at soil sites and at distances of 4.5 to 31 km. Details of these records are
presented in Table 1, and their five-percent damped elastic acceleration and
displacement response spectra along with their median spectra are presented in Figure
2.
Table 1. Details of ground motion recordings
Eq. No Magnitude Year Earthquake
6.6
1971 San Fernando
1
2
6.6
1971 San Fernando
3
7.1
1989 Loma Prieta
4
7.1
1989 Loma Prieta
5
7.5
1992 Landers
6
7.5
1992 Landers
7
6.7
1994 Northridge
Recording Station
Station 241
Station 458
Hollister, South & Pine
Gilroy #2
Yermo
Joshua Park
Century City LACC North
PGA (g)
0.26
0.12
0.18
0.32
0.15
0.28
0.26
Distance (km) *
16.5
18.3
17.2
4.5
31.0
10.0
23.7
* Closest distance to fault
3
Copyright ASCE 2004
Structures 2004
Spectral Acceleration (g)
1.2
Median Spectrum
Median +1 Sigma
8St Mode-1
8St Mode-2
16St Mode-1
16St Mode-2
1.0
0.8
0.6
0.4
0.2
0.0
0.0
0.5
1.0
(a)
2.0
2.5
3.0
3.5
4.0
3.0
3.5
4.0
Median Spectrum
Median +1 Sigma
8St. 1-Mode
8St. 2-Mode
16St. 1-Mode
16St. 2-Mode
20
Spectral Displacement (in)
1.5
Period, T n (sec)
15
10
0
0.0
(b)
0.5
1.0
1.5
2.0
2.5
Period, T n (sec)
Figure 2. (a) Pseudo spectral acceleration spectra, and (b) Spectral displacement
spectra
Lateral Load Configurations
Several lateral load cases were evaluated in this study. The first set of three patterns
is derived from FEMA-356. The following notations are used to describe these
patterns:
NSP-1: The buildings are subjected to a lateral load distributed across the height of
the building based on the following formula specified in FEMA-356:
Wx * hx k
Fx =
V
Wi * hi k
(1)
In the above expression, Fi is the applied lateral force at level x, W is the story
weight, h is the story height and V is the design base shear. This results in an inverted
4
Copyright ASCE 2004
Structures 2004
triangular distribution of the lateral load when the period-dependent power k is set
equal to unity.
NSP-2: A uniform lateral load distribution consisting of forces that are proportional
to the story masses at each story level.
NSP-3: A lateral load distribution that is proportional to the story shear distribution
determined by combining modal responses from a response spectrum analysis of the
building using the appropriate ground motion spectrum. Herein, we utilized the BSE2 hazard level design spectrum (Figure 3) for calculation of story shear forces.
The next set of lateral load configurations involved a combination of modes as
proposed in Kunnath (2004). The modal combination procedures involve identifying
appropriate modes to include in the analysis and the manner in which the combination
will be carried out. Two different methods of combination were used in this study.
These methods require an eigenvalue analysis of the structure to be carried out at the
initial elastic state and possibly again in their inelastic states.
Modal Combination Procedure 1, (MCP-1): In this procedure the spatial variation of
applied forces can be determined from Equation-1.
F j = n n m n S a ( n , Tn )
(1)
where n is a modification factor that can assume positive or negative values; n is

the mode shape vector corresponding to mode n; S a is the spectral acceleration at the
period corresponding to mode n; and
= ([ ] [m]{}) / M n in which M n = [ ]T [m][ ]
T
(2)
If only the first two modes were combined, then Equation 1 would have the following
form:
Fj = 11m1 Sa ( 1 , T1 ) 2 2 m 2 Sa ( 2 , T2 )
(3)
Therefore the procedure requires multiple pushover analyses wherein a range of

modal load patterns are applied. In order to arrive at estimates of deformation and
force demands, it is necessary to consider peak demands at each story level and then
establish an envelope of demand values for use in performance based-evaluation.
Modal Combination Procedure 2, (MCP-2): An alternative approach to the above
combination scheme is also investigated. Here, the lateral forces are determined in a
similar manner to the previous technique for each independent mode and then
combined using an appropriate combination rule such as SRSS. The individual
invariant load pattern is computed from the following expression:
5
Copyright ASCE 2004
Structures 2004
Fij = j ijWi S a ( j )
(4)
where i is the floor number and j is the mode number. The key aspect of this
procedure is the varying target displacement for each lateral loading case. Since the
first mode behavior of a structure is generally more dominant than the higher modes,
the calculated target displacement is kept constant for the first mode and scaled for
the other modes based on their corresponding spectral displacement values obtained
from response spectrum analysis (Figure 2b). Then the pushover procedure is applied
considering each individual invariant load distribution based on mode shapes. Further
details on this procedure are given elsewhere (Kalkan et al., 2004) and only
representative results are presented in this paper.
Details of Evaluation Procedure
For the pushover analyses, two-dimensional computer models of each building were
developed for use in OpenSees (2003). The program utilizes the layered fiber
approach for inelastic frame analysis. It has also the feature of representing the spread
of inelasticity along the member length as well as section level. In the finite element
domain, all members were simulated as nonlinear beam-column elements that
accounts for axial-moment interaction. The target displacement for each building
model was computed using the provisions in FEMA-356. Accordingly the sitespecific spectrum corresponding to BSE-2 hazard level was used (Figure 3).
2.5
BSE-2 Hazard Level Design Spectrum
Spectral Acc. (g)
2.0
1.5
1.0
0.5
0.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Period (sec)
Figure 3. Response spectrum for BSE-2 hazard level

The target displacements of 1.07m (42.5) and 1.30m (51.0) were computed using
Equation (3-15) in FEMA-356 for the 8 and 12-story buildings, respectively. Each
building model was subjected to the different lateral load configurations outlined
above. These lateral loads were incrementally applied to structures till the roof node
reached the specified target displacement.
6
Copyright ASCE 2004
Structures 2004
Summary of Results
Figure 4a-4b and 7a-7b show the peak displacement and peak inter-story drift profiles
obtained from nonlinear time history analyses of the seven ground motions for 8 and
12 storey structures, respectively. For transient analyses, records were scaled so that
their peak displacements are comparable to target displacements used in pushover
analyses. The median and 84 percentile curves of peak displacement and inter-story
drift profiles are presented in Figures 5a-5b and 8a-8b for two of the buildings. These
results are next used as benchmark solutions for comparing pushover analyses results.
8
7
7
6
Story Level
Story Level
4
3
4
3
0
0.00
(a)
Eq-1
Eq-2
Eq-3
Eq-4
Eq-5
Eq-6
Eq-7
Median
Eq-1
Eq-2
Eq-3
Eq-4
Eq-5
Eq-6
Eq-7
Median
0.01
0.02
0.03
0.04
0.00
(b)
Roof Drift Ratio
0.02
0.04
0.06
0.08
0.10
Interstory Drift Ratio
Figure 4. Nonlinear time history analysis results for 8-story building; (a) Roof
drift ratio; (b) Inter-story drift ratio
8
Median
Me dian
+/- 1 Sigm a
Story Level
Story Level
+ /- 1 Sigm a
4
3
4
3
1
0
0
0.00
(a)
0.01
0.02
0.03
0.00
0.04
(b)
Roof Drift Ratio
0.02
0.04
0.06
0.08
0.10
Figure 5. Nonlinear time history analysis results for 8-story building; Median
and 84 percentile curves of (a) Roof drift ratio; (b) Inter-story drift ratio
7
Copyright ASCE 2004
Structures 2004
Comparison of results (Figures 6a and 9a) reveal that peak displacements are
generally well represented by FEMA NSP-1 procedure that is closely analogous to
the elastic first mode loading pattern. The remaining two patterns (NSP-2 and 3)
overestimated the peak displacement at almost all levels. The plot of peak inter-story
drift, on the other hand, clearly highlights the inability of FEMA load patterns to
predict this critical design parameter (Figures 6b and 9b). An important consideration
in evaluating a pushover method, therefore, is its ability to predict inter-story drifts
rather than roof displacements. Consequently, the concept of a roof drift ductility
factor is not meaningful in the design or assessment of structures since the controlling
parameter may be local story failure mechanism.
8
NSP-1
NSP-1
NSP-2
NSP_3
NTH
NSP-2
NSP-3
4
3
4
3
4
3
0.00
(a)
Story Level
Story Level
Story Le vel
NTH
0.02
0.04
0.06
Roof Drift Ratio
0.00
(b)
0.03
0.05
0.08
0.10
Inter Story Drift Ratio
MCP-1
MCP-2
NTH
+/- 1 Sigm a
0.00
(c)
0.03
0.05
0.08
0.10
Figure 6. Nonlinear static pushover analysis results for 8-story building; (a-b)
FEMA procedures; (c) Modal combination procedures
12
11
10
9
11
10
9
8
7
6
5
6
5
4
(a)
0.000
Eq-1
Eq-2
Eq-3
Eq-4
Eq-5
Eq-6
Eq-7
Median
12
Story Level
Story Level
Eq-1
Eq-2
Eq-3
Eq-4
Eq-5
Eq-6
Eq-7
Median
0.010
0.020
0.030
0.000
(b)
Roof Drift Ratio
0.020
0.040
0.060
Figure 7. Nonlinear time history analysis results for 12-story building; (a) Roof
drift ratio; (b) Inter-story drift ratio
8
Copyright ASCE 2004
Structures 2004
It is clear from the findings reported here that the proposed modal-combination
procedures for capturing the inter-story drifts are significantly better than those of
single mode nonlinear static procedures (Figures 6c and 8c).
12
Median
12
11
+/- 1 Sigm a
11
Story Level
10
Story Level
10
6
5
Median
+/- 1 Sigm a
6
5
1
0
0
0.00
0.01
(a)
0.02
0.00
0.03
(b)
Roof Drift Ratio
0.02
0.04
0.06
Figure 8. Nonlinear time history analysis results for 12-story building; Median
and 84 percentile curves of (a) Roof drift ratio; (b) Inter-story drift ratio
12
NSP-1
NSP-2
NSP-3
NTH
11
11
10
11
10
9
5
4
6
5
4
(a)
Story Level
0.02
0.04
Roof Drift Ratio
0.06
(b)
0 .00
7
6
5
4
3
2
1
0
0.00
MCP-1
MCP-2
NTH
+/- 1 Sigm a
12
Story Level
S tor y Level
10
NSP-1
NSP-2
NSP_3
NTH
12
0.02
0.04
0.06
0.08
0.00
(c)
0.02
0.04
0.06
0.08
Figure 9. Nonlinear static pushover analysis results for 12-story building; (a-b)
FEMA procedures; (c) Modal combination procedures
Conclusions
Given the increasing use of nonlinear static pushover analysis in engineering practice,
the aim of the present paper is to develop alternative multi-mode pushover analysis
procedures in an attempt to better estimate the critical inelastic response quantities
such as inter-story drift and plastic hinge rotations. Various practically applicable
9
Copyright ASCE 2004
Structures 2004
modal combination techniques proposed for that purpose, and these procedures
intentionally avoid the complexity of adaptive methods by using invariant load
patterns.
This study also indicates that simple pushover methods such as those recommended
in FEMA-356 are incapable of predicting the story level at which the critical demands
occur. On the other hand, the results of modal combination procedures based on our
ongoing research appear to be promising in terms of better estimating the peak values
of the major inelastic response quantities such as lateral displacement, inter-story
drifts, and plastic hinge rotations (not reported here). The validity of the proposed
procedures should be evaluated through statistical studies considering various
structural systems and ground motions.
Acknowledgement
Funding for this study provided by the National Science Foundation under Grant
CMS-0296210, as part of the US-Japan Cooperative Program on Urban Earthquake
Disaster Mitigation, is gratefully acknowledged. Any opinions, findings, and
conclusions or recommendations expressed in this material are those of the authors
and do not necessarily reflect the views of the National Science Foundation.
References
Applied Technology Council (ATC-40). (1996). Seismic evaluation and retrofit of

concrete buildings, Redwood, CA.
FEMA-356. (2000). Prestandard and Commentary for the seismic rehabilitation of
buildings, American Society of Civil Engineers (ASCE), Reston, VA.
Kalkan, E., and Kunnath, S. K. (2004). Method of modal combinations for pushover
analysis of buildings. Proc. of the 13th World Conference on Earthquake
Engineering, August 1-6, 2004, Vancouver, BC, Canada.
Kunnath, S. K. (2004). Identification of modal combinations for nonlinear static
analysis of building structures. Computer-Aided Civil and Infrastructure
Engineering. (in press)
OpenSees. (2003). Open system
http://opensees.berkeley.edu.
for
earthquake
engineering
simulation,
SEAOC. (2000). Seismic design manual, Volume III, Building design examples:
Steel concrete and cladding, Structural Engineers Association of California,
Sacramento, CA.
International Conference of Building Officials (ICBO). (1997). Uniform building
code, Whittier, CA.
10
Copyright ASCE 2004
Structures 2004

Lateral Load Distribution in Nonlinear Static Procedures For Seismic Design

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Lateral Load Distribution in Nonlinear Static Procedures For Seismic Design

Hochgeladen von

Copyright:

Verfügbare Formate

Lateral load distribution in nonlinear static procedures for seismic design

E. Kalkan1 and S. K. Kunnath2

University of California Davis, Department of Civil and Environmental Engineering,

Description of Buildings Used in Evaluation

11@ 4.1 (13.5')

(a) Elevation along line A

(c) Member sizes

Moment resisting connection

(b) Plan view

* Closest distance to fault

Spectral Acceleration (g)

Spectral Displacement (in)

where n is a modification factor that can assume positive or negative values; n is

Therefore the procedure requires multiple pushover analyses wherein a range of

BSE-2 Hazard Level Design Spectrum

Spectral Acc. (g)

Figure 3. Response spectrum for BSE-2 hazard level

Roof Drift Ratio

Interstory Drift Ratio

Roof Drift Ratio

Interstory Drift Ratio

Roof Drift Ratio

Inter Story Drift Ratio

Interstory Drift Ratio

Roof Drift Ratio

Interstory Drift Ratio

Roof Drift Ratio

Interstory Drift Ratio

Roof Drift Ratio

Interstory Drift Ratio

Interstory Drift Ratio

Applied Technology Council (ATC-40). (1996). Seismic evaluation and retrofit of

Das könnte Ihnen auch gefallen