This action might not be possible to undo. Are you sure you want to continue?

BooksAudiobooksComicsSheet Music### Categories

### Categories

Scribd Selects Books

Hand-picked favorites from

our editors

our editors

Scribd Selects Audiobooks

Hand-picked favorites from

our editors

our editors

Scribd Selects Comics

Hand-picked favorites from

our editors

our editors

Scribd Selects Sheet Music

Hand-picked favorites from

our editors

our editors

Top Books

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Audiobooks

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Comics

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Sheet Music

What's trending, bestsellers,

award-winners & more

award-winners & more

P. 1

a modal pushover analysis procedure - chopra - goel |Views: 120|Likes: 7

Published by gogus79

a research report modal pushover analysis

a research report modal pushover analysis

See more

See less

https://www.scribd.com/doc/118924333/a-modal-pushover-analysis-procedure-chopra-goel

02/25/2015

text

original

- 1 Introduction
- 2 One-Story Systems
- 2.1 EQUATION OF MOTION
- 2.2 SYSTEM AND EXCITATION CONSIDERED
- 2.3 RESPONSE HISTORY ANALYSIS
- 2.4 PUSHOVER ANALYSIS
- 3. Elastic Multistory Buildings
- 3.1 MODAL RESPONSE HISTORY ANALYSIS
- 3.2 MODAL RESPONSE SPECTRUM ANALYSIS
- 3.3 MODAL PUSHOVER ANALYSIS
- 3.4 COMPARATIVE EVALUATION OF ANALYSIS PROCEDURES
- 3.4.1 System and Excitation Considered
- 3.4.3 Modal Pushover Analysis
- 4 Inelastic Multistory Buildings
- 4.1 RESPONSE HISTORY ANALYSIS
- 4.2 UNCOUPLED MODAL RESPONSE HISTORY ANALYSIS
- 4.2.1 Underlying Assumptions and Accuracy
- 4.2.2 Properties of the nth-“mode” Inelastic SDF System
- 4.2.3 Summary
- 4.3 MODAL PUSHOVER ANALYSIS
- 4.3.1 Summary
- 4.4 COMPARATIVE EVALUATION OF ANALYSIS PROCEDURES
- 4.4.1 Uncoupled Modal Response History Analysis
- 4.4.2 Modal Pushover Analysis
- 5. Comparison of Modal and FEMA Pushover Analyses
- 5.1 FEMA-273 PUSHOVER ANALYSIS
- 5.2 COMPARATIVE EVALUATION
- 6 Conclusions
- 7 References
- Appendix A Uncoupled Modal Response History Analysis

CHOPRA • RAKESH K. GOEL PEER 2001/03 JAN. 2001

Pacific Earthquake Engineering Research Center

**A Modal Pushover Analysis Procedure to Estimate Seismic Demands for Buildings:
**

Theory and Preliminary Evaluation

Anil K. Chopra University of California Berkeley Rakesh K. Goel California Polytechnic State University San Luis Obispo

A report on research conducted under grant no. CMS-9812531 from the National Science Foundation: U.S.-Japan Cooperative Research in Urban Earthquake Disaster Mitigation

PEER 2001/03

A Modal Pushover Analysis Procedure to Estimate Seismic Demands for Buildings: Theory and Preliminary Evaluation

Anil K. Chopra University of California Berkeley

Rakesh K. Goel California Polytechnic State University San Luis Obispo

A report on research conducted under Grant No. CMS-9812531 from the National Science Foundation: U.S. Japan Cooperative Research in Urban Earthquake Disaster Mitigation

PEER Report 2001/03 Pacific Earthquake Engineering Research Center College of Engineering University of California Berkeley January 2001 i

ii .

Instead. The standard response spectrum analysis (RSA) for elastic buildings is reformulated as a Modal Pushover Analysis (MPA). The results presented for El Centro ground motion scaled by factors varying from 0. it is demonstrated that the FEMA force distributions greatly underestimate the story drift demands. the total demand. is determined by combining the rno n = 1. and nonlinear RHA. The base shear-roof displacement Vbn . ro . p eff t = . and the MPA procedure is more accurate than all the FEMA force distributions methods in estimating seismic demands.mι ug t . Second. show that MPA estimates the response of buildings responding well into the inelastic range to a similar degree of accuracy when compared to standard RSA for estimating peak response of elastic systems. all pushover analysis procedures considered do not seem to compute to acceptable accuracy local response quantities. Comparing the earthquake-induced demands for the selected 9-story building determined by pushover analysis using three force distributions in FEMA-273.25 to 3. Combining these peak modal responses by modal combination rule leads to the MPA procedure. such as hinge plastic rotations. The peak response of the elastic structure due to its nth vibration mode can be exactly determined by pushover analysis of the structure subjected to lateral forces distributed over the height of the building according to s* = mφ n . rno . However. This pushover curve is idealized as bilinear n and converted to the force-deformation relation for the nth-“mode” inelastic SDF system. where m is the n mass matrix and φ n its nth-mode. MPA. and the structure is pushed to the roof displacement determined from the peak deformation Dn of the nth-mode elastic SDF system. The MPA procedure is extended to estimate the seismic demands for inelastic systems: First. is determined by pushover analysis.ABSTRACT The principal objective of this investigation is to develop a pushover analysis procedure based on structural dynamics theory.n t = . but provides superior accuracy in estimating seismic demands on buildings.urn curve is developed from a pushover analysis for force distribution s* . bg bg bg bg b g b g Comparing the peak inelastic response of a 9-story SAC building determined by the approximate MPA procedure with rigorous nonlinear response history analysis (RHA) demonstrates that the approximate procedure provides good estimates of floor displacements and story drifts. at which the seismic response. Thus the present trend of comparing computed hinge plastic rotations against rotation limits established in FEMA-273 to judge structural performance does not seem prudent.0. structural performance evaluation should be based on story drifts known to be closely related to damage and can be estimated to a higher degree of accuracy by pushover analyses.… according to an appropriate modal combination rule. The peak deformation of this SDF system is used to determine the roof displacement. plastic hinge rotations are less accurate. p eff . iii . 2. in the modal expansion of the effective earthquake forces. Thus the MPA procedure is accurate enough for practical application in building evaluation and design.sn ug t . a pushover analysis determines the peak response rno of the inelastic MDF system to individual modal terms. and identifies locations of most plastic hinges. which retains the conceptual simplicity and computational attractiveness of current procedures with invariant force distribution.

iv .

a part of the U.S.ACKNOWLEDGMENT This research investigation is funded by the National Science Foundation under Grant CMS9812531. v . This financial support is gratefully acknowledged. Japan Cooperative Research in Urban Earthquake Disaster Mitigation.

vi .

.34 4............2......2...................................................................................................................................................................................................................1 Underlying Assumptions and Accuracy ...........4...................................................2 Response History Analysis ...................................2 2.........................2 Modal Pushover Analysis ..........................38 4.....4 3....12 Modal Pushover Analysis ......1 2...................................................................4.........2 Response History Analysis ..........................................55 Comparative Evaluation...................................................................................... Inelastic Multistory Buildings ............ vii 1..........................3 System and Excitation Considered .............................4......................1 One-Story Systems ......................3 4....................3 Modal Pushover Analysis ...................................................................3 2.......4 5.37 Comparative Evaluation of Analysis Procedures ................................................................................47 4...................................................6 Elastic Multistory Buildings ................................................1 Summary...........................................................................................................................................................................1 5...............4....................................13 3....................3 2......13 3.............................2....................................................................... iii Acknowledgment ............................................................28 4...............................................................................................................................................................................................1 3............................................................3 3................................2 3...................................... 2......................................2 Properties of nth-“mode” Inelastic SDF System ................6 Pushover Analysis ........................................................................36 4..........................................4......................................................27 Uncoupled Modal Response History Analysis .........................................9 Modal Response Spectrum Analysis .............. Comparison of Modal and FEMA Pushover Analyses............................9 3.................................................27 4............................1 4........................................CONTENTS Abstract .......41 4...........17 3............................................................................................................................2 FEMA-273 Pushover Analysis .........................4.............5 Response History Analysis ...4 Modal Response History Analysis...........................3........................v Table of Contents................................22 4...............33 4........................1 System and Excitation Considered .................3 Modal Pushover Analysis with Gravity Loads ....38 4.....55 vii ................................55 5.........12 Comparative Evaluation of Analysis Procedures .........................................36 Modal Pushover Analysis .................. Introduction............................................................ Equation of Motion ...................................................................................................................................3 Summary..................................1 Uncoupled Modal Response History Analysis ..........

..........6.................................................71 B: Modal Pushover Analysis .......................................................................................69 Appendices A: Uncoupled Modal Response History Analysis .85 viii .................................................65 References......... 7...............................................................................81 C: FEMA Force Distribution Calculation................................................................................. Conclusions.......................................................................................................................................................

Miranda. 1999. or for a redistribution of inertia forces because of structural yielding and the associated changes in the vibration properties of the structure. 1999]. Lawson et al. 1997. Kim and D’Amore. Maison and Bonowitz. 1981. Both the force distribution and target displacement are based on the assumption that the response is controlled by the fundamental mode and that the mode shape remains unchanged after the structure yields.and medium-rise structures in which inelastic action is distributed throughout the height of the structure [Krawinkler and Seneviratna. Obviously. Gupta and Krawinkler. Bracci et al. after the structure yields both assumptions are approximate. Skokan and Hart. 1991. requires explicit consideration of inelastic behavior of the structure. 2000]. Gupta and Krawinkler. 2000] have led to good estimates of seismic demands.. 1999. but investigations [Saiidi and Sozen. While nonlinear response history analysis (RHA) is the most rigorous procedure to compute seismic demands.1 Introduction Estimating seismic demands at low performance levels. Krawinkler and Seneviratna. several researchers have proposed adaptive force distributions that attempt to follow more closely the time-variant distributions of inertia forces [Fajfar and Fischinger. Fajfar and Fischinger. such satisfactory predictions of seismic demands are mostly restricted to low.. 1999. None of the invariant force distributions can account for the contributions of higher modes to response. To overcome these limitations. 1988. such as life safety and collapse prevention. Attempts 1 . 1998. 1994. However. 1997]. 2000]. While these adaptive force distributions may provide better estimates of seismic demands [Gupta and Kunnath. current structural engineering practice uses the nonlinear static procedure (NSP) or pushover analysis in FEMA-273 [Building Seismic Safety Council. Gupta and Kunnath. they are conceptually complicated and computationally demanding for routine application in structural engineering practice. 1998. 1988. The seismic demands are computed by nonlinear static analysis of the structure subjected to monotonically increasing lateral forces with an invariant height-wise distribution until a predetermined target displacement is reached. 2000.

. the seismic demands determined by pushover analysis using three force distributions in FEMA-273 are compared against the MPA and nonlinear RHA procedures. Gupta and Kunnath. 2000. The MPA procedure is then extended to inelastic buildings. Next we develop a modal pushover analysis (MPA) procedure for linearly elastic buildings and demonstrate that it is equivalent to the well-known response spectrum analysis (RSA) procedure. The principal objective of this investigation is to develop an improved pushover analysis procedure based on structural dynamics theory that retains the conceptual simplicity and computational attractiveness of the procedure with invariant force distribution. Sasaki et al. Kunnath and Gupta.. we show that pushover analysis of a one-story system predicts perfectly peak seismic demands. Finally. 2000]. 2000. 1998. but provides superior accuracy in estimating seismic demands on buildings.have also been made to consider more than the fundamental vibration mode in pushover analysis [Paret et al. First. Matsumori et al. and the errors in the procedure relative to a rigorous nonlinear RHA are documented. 2 .. the underlying assumptions and approximations are identified. 1996.

the strength required for the structure to remain elastic during the ground motion. the yield deformation.mug t bg The governing equation for this inelastic system subjected to horizontal ground acceleration b g bg 3 (2. It is the familiar bilinear hysteretic relationship. through the yield strength reduction factor. this system is linearly elastic with stiffness k as long as the force does not exceed f y . defined by f Ry = o fy (2. and a linear viscous damper with damping coefficient c. sign u . It consists of a mass m (or weight w) concentrated at the roof level. reloading from a minimum deformation takes place along a path parallel to the initial elastic branch.1) b g ug t is mu + cu + f s u. R y .1 One-Story Systems EQUATION OF MOTION Consider the idealized one-story structure shown in Fig. Unloading from a maximum deformation takes place along a path parallel to the initial elastic branch. On initial loading. The yield strength is related to f o . sign u = . the yield strength. The yield strength is the same in the two directions of deformation.1a.2) . The hysteretic relation between the lateral force f s and lateral displacement u of the mass relative to the base of the frame is denoted by f s u. During yielding the stiffness of the frame is a k . 2. This lateral-force displacement relation is idealized as shown in Fig. Similarly. a massless frame that provides stiffness to the system.2 2. Yielding begins when the force reaches f y and the deformation reaches u y . where 0 < a << 1. 2.1b.

deformation is denoted by um .e. 2. and z is the damping ratio of the system vibrating within its linear elastic range (i.3) ωn = k m ζ = c 2mω n f fs = s fy (2. Tn = 2p w n is the natural vibration period. The peak.fS m fs c ut u fy 1 αk k 1 uy ug -fy k k um u 1 1 (a) (b) Fig. or absolute (without regard to algebraic sign) maximum.4) and w n is the natural vibration frequency. This becomes evident if Eq. (2. (a) Idealized one-story structure. z . and (b) bilinear hysteretic forcedeformation relation For a given excitation u g t . z y . m depends on three system parameters: w n .ug t n where bg bg b g bg (2. bg 4 . and R y (Chopra . and u y . u £ u y ). and the ductility factor is u m= m uy (2.3).1) is divided by m to obtain ~ u + 2zw n u + w 2 u y f s u. 2001. the deformation u t depends on three systems parameters: w n .5) For a given u g t . sign u = .1. in addition to the form of the force-deformation relation. Section 7..

2.32 m • Fig. 2.2. 2.2 kips) .826 kips) for w = 169. one-bay frame 50 uy = 1. f o w = 184 . Pushover curve for structure shown in Fig.2. u (cm) 8 Fig.376 cm.9 kN (38.2 5 . m h = 3. 2001) and scaled up by a factor of 2. α = 0. One-story. 2. is f y / w = ( f o / w ) ÷ 8 = 0.2 SYSTEM AND EXCITATION CONSIDERED Consider the one-story system in Fig.4 in Chopra.3.04 40 Base Shear. based on Ry = 8 .26 kN (8. For . Vby = 39.2: the dimensions and flexural rigidity of the structural elements are noted.2311 .5 sec and z = 5% subjected to the north-south component of the El Centro (1940) ground motion (Fig. and f y = 39.26 kN. with Tn = 0. The yield strength of the inelastic system.1. 6. this system and excitation.66 m • EIb EIc EIc • • L = 7. Vb (kN) 30 20 10 0 0 2 4 6 Roof Displacement.

and (e) shows the force-deformation relation. wherein the hysteretic force-deformation history of Fig.36 cm.3 RESPONSE HISTORY ANALYSIS Figure 2.1 kip-in.. 2. e. pushover analysis cannot provide any cumulative measure of response.36 cm . Observe that the pushover curve matches the initial loading path of the hysteretic system. 2.65 kN-m (191. (d) shows the rotation q p bt g of the plastic hinges at the bg bg beam ends. The yielding stiffness of each structural element is defined as 3% of its initial stiffness. and q pm = 0. The system is excited well beyond the yield deformation.0217 rad . (b) shows the lateral force f s t or base shear Vb t normalized relative to the bg weight.g.) for the beam and columns. as apparent in Fig. 2. 2.The yield moments in the beam and columns are defined as the bending moments due to the lateral force f y . 2.35.4f. The peak values of the various response quantities are as follows: um = 7. Implementing this analysis with I c = 6.134 × 107 mm 4 (75. Determined from pushover analysis at the exact peak deformation. (c) shows the joint rotation q bt g . the peak value determined from RHA.18 kN-m (444.6 kip-in. The resulting pushover curve is shown in Fig.4 ). respectively.017 rad .4 PUSHOVER ANALYSIS Static analysis of the one-story nonlinear system subjected to lateral force that increases in small increments is implemented until the lateral displacement reaches um = 7.4e.4edetermined from RHAis superimposed. It is organized in five parts: (a) shows the deformation u t . the energy dissipated in 6 . However. This pushover curve turns out to be bilinear because the beam and the columns are designed to yield simultaneously when f s reaches f y . and E = 2 × 108 kPa (29×103 ksi) gives yield moments of 21.4 shows the earthquake response of the system described in the preceding section determined by response history analysis (RHA). the ductility factor m = 5.) and 50. A static pushover analysis of this one-story system leads to the force-displacement relationship shown in Fig.3. q m = 0. 2.3 in. I b = 3.4 ).077 × 107 mm 4 (146 in. the joint rotation and beam hinge rotation are identical to values q m and q pm determined from RHA.

2311 y (b) −0.4 −10 −5 0 u (cm) 5 10 −0. 15 u (cm) 0 • 7.4.36 −15 0. (b) base shear. or the cumulative rotation at a plastic hinge.4 0.04 0 5 10 15 Time (sec) 20 25 30 0. 2.04 0.2 −0.2311 (a) Vb / w 0 f / w = −0.017 • 0 (d) θp (rad) −0. and (f) pushover curve 7 .yielding during the ground motion.04 0. (c) joint rotation. Response of one-story system to El Centro ground motion: (a) deformation.2 0 (e) 0 (f) −0.2 Vb / w 0.4 0. (d) plastic hinge rotation.4 −10 −5 0 u (cm) 5 10 Fig.5 fy / w = 0. This represents an inherent limitation of pushover analyses. (e) force-deformation relation.0217 • θ (rad) 0 (c) −0.04 0.2 −0.5 0.

8 .

classical damping. The right side of Eq. 2001: Section 13. m.2) The spatial distribution of these “forces” over the height of the building is defined by the vector s = m ι and their time variation by u g t . each element of the influence vector ι is equal to unity.3) where φ n is the nth natural vibration mode of the structure. and lateral stiffness matrices of the system.1 MODAL RESPONSE HISTORY ANALYSIS The differential equations governing the response of a multistory building to horizontal earthquake ground motion u g t are as follows: mu + cu + ku = .n t = Â .mι ug t bg bg bg (3.4) 9 . This force distribution can be expanded as a summation of modal inertia force distributions s n (Chopra. Elastic Multistory Buildings 3. c. (3.snug t n= n =1 The effective earthquake forces can then be expressed as af N af N af (3.1) can be interpreted as effective earthquake forces: p eff t = .1) where u is the vector of N lateral floor displacements relative to the ground.m ι u g t bg bg (3. and k are the mass. and p eff t = Â peff.12): mι = n =1 Â sn = N n =1 Â Γ nmφ n N (3.3.

The equations governing the response of the system are mu + cu + ku = . (3.sn ug t bg (3. (3. subjected to u g t : Dn + 2z nw n Dn + w 2 Dn = . with no contribution from other modes.u g t n Comparing Eqs. an SDF system with vibration propertiesnatural frequency w n and damping ratio z n of the nth-mode of the MDF system. The solution qn t can readily be obtained by comparing Eq.Gn ug t (3.9) in which w n is the natural vibration frequency and z n is the damping ratio for the nth mode.n t = . we will outline that the response of the MDF system to p eff .9) to the equation of motion for the nth-mode elastic SDF system.peff (t ) = n =1 ∑ peff .n (t ) = N n =1 ∑ −sn u g (t ) N (3.7) By utilizing the orthogonality property of modes.10) gives qn t = Gn Dn t bg bg bg (3. Then the floor displacements are u n t = φ n qn t bg bg bg bg (3. p eff .11) and substituting in Eq.9) and (3.10) bg bg bg 10 (3.8) where the modal coordinate qn t is governed by 2 qn + 2z nw n qn + w n qn = .8) gives the floor displacements u n t = Gnφ n Dn t bg (3.5) The contribution of the nth mode to s and to p eff t are: sn = Gn mφ n respectively. it can be demonstrated that none of the modes other than the nth mode contribute to the response. (3.12) .6) Next.n t is entirely in the nth-mode.sn ug t bg bg bg bg (3.

9) is the standard modal equation governing qn t .12) and (3.13) define the contribution of the nth-mode to the response. (3.14) is the pseudo-acceleration response of the nth-mode SDF system (Chopra.15) rt = bg n =1 Â rn t = bg n =1 Â rnst An bt g N (3. Eqs. ζn ¨ g(t ) u (a) Static Analysis of Structure (b) Dynamic Analysis of SDF System Fig.13) represent the response of the MDF system to p eff.1. and An t = w 2 Dn t n bg (3. 3.can be expressed by st rn t = rn An t bg bg bg bg bg (3. 2001. However.12) and (3. and Eqs.Forces sn An(t ) st rn ωn. (3. Equations (3.1).1. Conceptual explanation of modal response history analysis of elastic MDF systems Any response quantity r t story drifts. st The two analyses that lead to rn and An t are shown schematically in Fig. etc. these standard equations have been derived in an unconventional way. Therefore. the response of the system to the total excitation p eff t is ut = bg bg bg n =1 N Â un bt g = Â Gnf n Dn bt g n =1 N N (3.16) reflect combining the response contributions of all modes.15) and (3. 3. the static value of r due to external forces sn . internal element forces.16) This is the classical modal RHA procedure wherein Eq.n t .13) st where rn denotes the modal static response. In contrast to the classical 11 bg . (3. Section 13.

18) To develop a pushover analysis procedure consistent with RSA.8.16).(3. (3. In such a response spectrum analysis (RSA). which is valid for structures with well-separated natural frequencies such as multistory buildings with symmetric plan. The SRSS rule. Sections 12. Alternatively. 2001.9) . The peak modal responses are combined according to the Square-Root-of-Sum-ofSquares (SRSS) or the Complete Quadratic Combination (CQC) rules.17) (Chopra. 3.3 MODAL PUSHOVER ANALYSIS (3. the peak nth-mode response as in Eq. the peak value rno of the nth-mode contribution rn t to response r t is determined from st rno = rn An bg bg bg (3.1). we used the modal expansion of the spatial distribution of the effective earthquake forces.g.3). provides an estimate of the peak value of the total response: b g F N r 2 I 1/ 2 ro ª G Â no J H n =1 K 3. (3. we note that static analysis of the structure subjected to lateral forces f no = Γ n mφ n An (3. This concept provides a rational basis for the modal pushover analysis procedure developed later.1.17) where An is the ordinate A Tn . Chopra. Section 13.derivation found in textbooks (e.2 MODAL RESPONSE SPECTRUM ANALYSIS The peak value ro of the total response r t can be estimated directly from the response spectrum for the ground motion without carrying out the response history analysis (RHA) implied in Eqs.19) will provide the same value of rno . this response value can be obtained by static analysis of the structure subjected to lateral forces distributed over the building height according to 12 . 2001. and Tn = 2p w n is the natural vibration period of the nth-mode of the MDF system..4 and 13. z n of the pseudo-acceleration response (or design) spectrum for the nth-mode SDF system.

4. California. Typical floor-to-floor heights (for analysis purposes measured from center-of-beam to center1 2 Brandow & Johnston Associates.4 COMPARATIVE EVALUATION OF ANALYSIS PROCEDURES 3.18) to obtain an estimate of the peak value ro of the total response. was designed by Brandow & Johnston Associates1 for the SAC2 Phase II Steel Project.15 m (30 ft) on center. CA 94804-4698. with five bays each in the north-south (N-S) and east-west (E-W) directions.2). shown in Fig. Richmond. with the ninth level being the roof. SAC is a joint venture of three non-profit organizations: The Structural Engineers Association of California (SEAOC). the Applied Technology Council (ATC).73 m (150 ft) by 45. The bays are 9. 1301 S. The building’s lateral load-resisting system is composed of steel perimeter moment-resisting frames (MRFS) with simple framing on the farthest south E-W frame. CA 90017. SAC Steel Project Technical Office.s* = mφ n n (3. denoted B1. Consulting Structural Engineers. Obviously Dn and An are available from the response (or design) spectrum.. 46th Street. and California Universities for Research in Earthquake Engineering (CUREE). The columns are 345 MPa (50 ksi) steel wideflange sections. urno . (3. 13 .20) and the structure is pushed to the roof displacement.73 m (150 ft) in plan. 3. the peak value of the roof displacement due to the nth-mode. The peak modal responses. The interior bays of the structure contain simple framing with composite floors. rno . The building has a basement level. can be combined according to Eq. (3. each determined by one pushover analysis.1 System and Excitation Considered The 9-story structure. in both directions.2. The levels of the 9-story building are numbered with respect to the ground level (see Fig. This modal pushover analysis (MPA) for linearly elastic systems is equivalent to the well-known RSA procedure (Section 3.2). 1660 W. 3. 3.12) is urno = Gnf rn Dn (3. which from Eq. Third St.19 m (122 ft) in elevation. region. A benchmark structure for the SAC project. Although not actually constructed. Los Angeles.21) 2 where Dn = An ω n . this building is 45. and 37. this structure meets seismic code and represents typical medium-rise buildings designed for the Los Angeles.

20). floor slabs.4. 3. The simple model adopted here is sufficient for the objectives of this study. The building is modeled in DRAIN-2DX (Allahabadi and Powell. which are seismic (tension) splices to carry bending and uplift forces. 0. dimension.27. Each frame resists one half of the seismic mass associated with the entire structure.49 m (18 ft). The first three vibration modes and periods of the building for linearly elastic vibration are shown in Fig.of-beam) are 3. 3. third. The strength.0 kips-sec2/ft). The 9-story N-S MRF is depicted in Fig. and for the ninth level is 1. including the steel framing. we select a weak ground motion: the northsouth component of the El Centro (1940) ground motion scaled down by a factor of 4. Concrete foundation walls and surrounding soil are assumed to restrain the structure at the ground level from horizontal displacement. Column splices. This model is based on centerline dimensions of the bare frame in which beams and columns extend from centerline to centerline. The column bases are modeled as pinned and secured to the ground (at the B-1 level). the vibration periods are 2. 14 .49 sec. such as those described in Gupta and Krawinkler (1999) can be used.65 m (12 ft) and for the first floor is 5.2 kips-sec2/ft). To ensure that this structure remains elastic. monolithic column pieces are connected every two levels beginning with the first level.96 m (13 ft). fifth.e.01×106 kg (69.3.07×106 kg (73. roofing and a penthouse located on the roof. partitions. 3. for the first three modes are shown in Fig. The column lines employ two-tier construction. ceiling/flooring. if desired more complex models. and 0.65×105 kg (66. for the first level is 1. mechanical/electrical.00×106 kg (616 kips. These force n distributions will be used in the pushover analysis to be presented later. and seventh levels at 1. s* (Eq.2.85. for the second through eighth levels is 9.89×105 kg (67.sec2/ft). The floor system is composed of 248 MPa (36 ksi) steel wide-flange beams in acting composite action with the floor slab. and shear distortion of panel zones are neglected but large deformation (P∆) effects are included. 1988) using the M1 model developed by Krawinkler and Gupta (1998). The seismic mass of the structure is due to various components of the structure.0 kips-sec2/ft).83 m (6 ft) above the center-line of the beam to column joint.7 kips-sec2/ft). i. The floor-to-floor height of the basement level is 3. The seismic mass of the ground level is 9. 3. are located on the first. The seismic mass of the above ground levels of the entire structure is 9.. respectively. The force distributions.

Fig.. Nine-story building [adapted from Ohtori et al.2. 3. 2000] 15 .

n = 1.03 −1.4.49 sec 3 T = 0.72 −2.5 Mode Shape Component 1 1.13 −1.1 3.44 1.5 0 0.67 −1.3.93 −1.05 1. determined by RHA [Eqs.5.51 0. respectively. 3.38 0.85 sec 2 Floor 5th 4th 3rd 2nd 1st T1 = 2.05 s * 1 s * 2 s * 3 Fig.796 0.13)].5 −1 −0.2 Response History Analysis The structural response due to individual vibration modes. 2. 3. Each figure is organized in four parts: (a) shows the roof displacement urn t .4. 3. is shown in Figs.12 0.61 2. and 3.04 1.728 2.5 Fig. and 3 n 3. (3.39 3.8 −2. and 3 .9th 8th 7th 6th T = 0.33 2.487 −1.12) and (3. Force distributions s* = mφn .94 2.31 −0.1 −2. 3.27 sec Ground −1. First three natural-vibration periods and modes of the 9story building 3. (b) shows the base shear Vbn t normalized relative to the weight W of the building.7.75 1.37 2.6.0272 −2. (c) shows the joint rotation q n t of an external joint at 16 bg bg bg . n = 1.05 2. 2.

2.8. The errors due to truncating the contributions of vibration modes beyond the third mode are negligible. consistent with the increasing significance of the higher mode response among these three sets of response quantities.2. respectively. the exact response considering all modes. and ur 3o = 0. with the first three modes provide essentially the exact response. and rotations of external joints with moment connections are presented in Tables 3. As expected. 17 .16)]. errors are smallest in floor displacements.8. in particular. 3. the results for the roof displacement and top-story drift are shown in Fig.422 cm.15) and (3. the peak roof displacement due to each of three modes is ur1o = 9. Combining the modal response histories for all modes gives the total response [Eqs. are presented in Fig. (3. especially in estimating the story drifts. where the second and third modal responses are a larger percentage of the top story drift compared to roof displacement.urn relation. The linear relationship between the base shear and roof displacement for each mode implies that the structure did not yield.12 cm. two. and 3. or all modes. and three vibration modes. The peak values of displacements of all floors. and even larger in joint rotations. but three modes—perhaps even two modes—are sufficient. two. which are listed in Tables 3. 3. 3.1. three.the roof level. errors generally decrease as response contributions of more modes are included. ur 2o = 2. larger in story drifts. Also included are the combined response due to one.3. 3. and the percentage errors due to truncation of higher modal contributions.23 cm. The peak values of floor displacements and story drifts determined by RHA. 3.1. drifts in all stories.9. This is illustrated in Fig. The peak values of the various response quantities are noted in these figures. and (d) shows the Vbn . It is apparent that the first mode alone is inadequate. For a fixed number of modes included. and 3. including one. The same method was used to determine the peak values of many response quantities. respectively.3.

7 -50.01E-04 -2.91E-04 1.071 0.183 0.89E-03 1.089 0.1 -0.3 -3.156 0.4 -10.9 2.1 -14.31E-03 Mode 2 -1.009 -0.235 0.2 -20.2 -1.156 0.9 3.157 0.1 Table 3.14 m3) from RHA for 0.062 -0.042 0.203 0.28E-04 1.45E-03 3.44E-03 1.72E-03 3.259 0.99E-03 Mode 3 3.088 -0.9 -22.7 4.6 0.09E-03 1.038 0.85E-03 3.245 0.023 0.130 0.22E-03 2.0 -46.229 0.74E-04 6.4 -53.325 0.8 1.055 0.090 0.3 Table 3.011 0.125 0.310 0.0 7.124 0.7 2.407 0.44E-03 3.29E-03 2.03E-03 3 Modes 2.4 0.227 0.2 1.152 0.03E-03 1.173 0.1 -11.64E-03 3.2 9.202 0.133 -0.65E-03 RHA (all modes) 2.152 0.413 RHA (all modes) 0.253 0.88E-03 2.125 0.124 0.406 0.181 0.022 0.6 0.88E-03 2.4 -41.333 0.258 0.09E-03 1.09E-03 Error (%) 1 Mode -23.78E-04 -3.032 -0.1 Building height is measured from the ground floor to the 9th floor.63E-03 2.264 Error (%) 1 Mode 2 Modes 3 Modes -23.1 -2.4 -1.7 -19.202 0.1 -2.74E-03 1.8 -56.475 0.90E-03 3.370 0.11E-04 -5.466 0.0 -2.275 0.015 -0.060 -0.054 0.03E-03 -6.3 Peak values of joint rotations (radians) from RHA for 0.1 Peak values of floor displacements (as % of building height = 37.26E-04 -5.303 0.400 0.8 -5.08E-03 2.282 0.282 0.265 0.001 -0.401 Error (%) 1 Mode 2 Modes 3 Modes -23.253 0.38E-03 2.0 3 Modes -5.9 -24.Table 3.002 -0.15E-03 4.74E-04 9.058 -0.4 -0.6 -1.263 0.99E-03 2.069 0.307 0.043 0.25 × El Centro ground motion Floor 1st 2nd 3rd 4th 5th 6th 7th 8th 9th 3 Modal Response Mode 1 2.9 1.47E-03 1.14E-03 2.09E-03 2.9 -15.012 0.024 -0.062 0.199 0.300 0.4 -7.6 11.1 -2.5 18.266 0.7 7.197 0.336 0.25 × El Centro ground motion Displacement /Height (%) Modal Response Combined (RHA) Mode 1 Mode 2 Mode 3 1 Mode 2 Modes 3 Modes Floor 1st 2nd 3rd 4th 5th 6th 7th 8th 9th 0.1 4.192 0.097 0.069 0.8 -1.080 0.378 0.63E-03 2.6 9.03E-03 1.6 4.295 -0.56E-03 2.179 0.38E-04 2 Modes 2.117 0.006 0.89E-03 1.364 0.2 Peak values of story drift ratios (as % of story height) from RHA for 0.9 -16.2 0.8 -10.012 -0.231 -0.76E-03 1.76E-03 1.50E-03 4.13E-03 2.24E-03 2.311 0.33E-03 2.045 0.026 0.74E-03 1.350 0.3 -8.50E-03 2.121 0.13E-04 9.311 0.177 0.265 RHA (all modes) 0.3 -0.5 -2.8 1.66E-05 -3.1 -19.00E-03 1.042 0.060 0.010 -0.260 0.65E-03 2.008 -0.035 0.9 -23.2 -57.259 0.011 0.321 0.226 0.225 0.266 0.173 0.6 1.229 0.9 9.4 -22.1 3.202 0.453 0.2 -2.00E-03 2.31E-03 Joint Rotation (rad) Combined (RHA) 1 Mode 2.7 2 Modes -3.5 -1.226 0. 18 .2 -4.8 -15.25 × El Centro ground motion Story 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Modal Response Mode 1 Mode 2 Drift Ratio (%) Combined (RHA) 1 Mode 2 Modes 3 Modes Mode 3 0.097 0.6 -0.205 0.00E-03 1.0 -2.261 -0.3 19.0 -10.227 0.097 0.73E-03 3.4 -3.237 0.159 0.260 0.177 0.237 0.01E-04 3.060 -0.235 0.9 8.5 0.0 -0.6 -1.4 -6.245 0.3 -0.4 -1.7 3.11E-03 1.94E-03 2.3 -33.399 0.003 0.266 0.42E-04 1.

(b) base shear.1 /W 0. and (e) pushover curve.5.12 • 0 (a) r1 −15 0.5 15 −0. Excitation is 0.25 × El Centro ground motion 19 . 3.12 0 (d) 0 (e) 15 Fig.0435 • 0 (b) V b1 −0. Response due to first mode: (a) roof displacement.12 −0.1 −15 −7.0435 Vb1 / W 0.1 5 x 10 −3 θ (rad) 0.1 −15 −7.5 0 u (cm) r1 7.1 0.1 0.0435 9. (d) force-deformation history.5 9.5 0 u (cm) r1 7. (c) joint rotation.00174 • 0 (c) r1 −5 0 5 10 15 Time (sec) 20 25 30 0.15 u (cm) 9.

3.1 0.23 0.5 u (cm) r2 0 • 2.23 −5 (a) 0.5 0 u (cm) r2 2. and (e) pushover curve.5 5 −0. (b) base shear.1 Vb2 / W 0.00244 • 0 (c) r2 −5 0 5 10 15 Time (sec) 20 25 30 0. (c) joint rotation.05 /W 0.5 5 Fig.0252 (d) 0 0 (e) −0.05 5 x 10 −3 b2 θ (rad) 0.1 −5 −2.0252 −2.23 −2.25 × El Centro ground motion 20 .0252 • 0 (b) V −0. Response due to second mode: (a) roof displacement. Excitation is 0. (d) force-deformation history.6.5 0 u (cm) r2 2.1 −5 −2.

422 (e) −0.02 −1 −0.5 1 −0.422 −1 (a) 0. Response due to third mode: (a) roof displacement.422 /W V b3 0 (d) 0 −0. (d) force-deformation history.5 0 u (cm) r2 0.02 −0.7.02 /W V b3 0 • 0. and (e) pushover curve. (c) joint rotation. 3.02 −1 −0.000913 • 0 (c) r3 −2 0 5 10 15 Time (sec) 20 25 30 0.0108 −0.02 2 θ (rad) 0.0108 −0.5 1 Fig.1 u (cm) r3 0 • 0.25 × El Centro ground motion 21 . Excitation is 0.5 0 u (cm) r3 0. (b) base shear.02 0.0108 x 10 −3 (b) −0.

and the errors in these estimates relative to the exact response from RHA considering all modes.20) with n = 2 and 3 up to roof displacements ur 2 o = 2. 3. the value determined by RHA (Fig. using the force distribution of Eq. i.12).21). story drifts. These values of the response quantities are identical to the peak response values determined from RHA (Tables 3. and 3.5. As for the first mode.e. respectively.9). 3. pushing the structure using the force distribution of Eq. the errors in the MPA results are generally larger than in RHA (Fig. respectively.6. 3. these pushover curves are consistent with the Vb .4. 3.6. 3.3).3). and 3. 3.1. i. (3. and 3. 3. implying that the modal truncation errors are small if at 22 . (3.4.23 cm . In practical application.2. pushing the structure. The floor displacements. although both analyses led to identical peak values of the individual modal responses.8) leads to the pushover curve shown in Fig. This pushover curve is consistent with the relationship between the base shear and roof displacement determined by RHA (Fig. and 3. considering one. For a fixed number of modes included considered. and external joint rotations in Tables 3.4) to roof displacement ur1o = 9.ur relations determined by RHA (Figs.3. confirming that MPA gives the exact values of the individual modal responses.5. leads to the pushover curves shown in Figs.4. story drifts.7d). and 3. and it is apparent in the example considered that three modes provide most of the response (Fig. 3.4.1.3 Modal Pushover Analysis Implementing MPA for the fundamental vibration mode. In RHA the errors arise only from truncating the responses due to higher modes. and external joint rotations computed by pushover analysis are presented in Tables 3. As suggested by Eq. Observe that the target roof displacement in each pushover analysis is identical to its exact value determined by RHA.6e and 3. and ur 3o = 0. which would provide the Dn value to be substituted in Eq. Implementing pushover analysis for the second and third modes..5d). this value would be determined directly from the response (or design) spectrum.1 through 3. or three vibration modes.10 and Tables 3. (3.6 present estimates of the combined response according to Eq. except for the algebraic sign associated with Γ1 and minor round-off errors. 3.3).18). 3.5. Figure 3.422 cm .7e and to the floor displacements..20) with n = 1 (Fig. the floor displacements are proportional to the mode shape φ 1 because the structure remains elastic.12 cm. 3. 3.5e. and the computed response values are identical to the peak response values determined from RHA (Tables 3. (3. 3.6d and 3.9 and Tables 3. respectively. (3. two.e.2.

8.1 0.83 • r3 0 • 0.23 r2 −15 15 0 • 0.25 × El Centro ground motion: first three modal responses and total (all modes) response (a) Floor Displacements 9th 8th 7th 6th 9th 8th 7th 6th (b) Story Drift Ratios Floor 4th 3rd 2nd 1st Floor 5th RHA All Modes 1 Mode 2 Modes 3 Modes 5th 4th 3rd 2nd 1st RHA All Modes 1 Mode 2 Modes 3 Modes fig3_9a.25 × El Centro ground motion 23 .5 0.5 Ground 0 0.3 0.4 Displacement/Height (%) 0.48 −3 3 1.eps r 5 10 15 20 Time.2 0. sec 25 30 Fig.03 −3 3 Mode 3 ∆ (cm) u (cm) r3 −15 15 9.9.4 Story Drift Ratio (%) 0.eps Ground 0 0. 3. Heightwise variation of floor displacements and story drifts from RHA for 0.3 0.422 r2 0 • 1.eps fig3_9b.6 Fig.12 • r1 −15 15 0 r1 0 0 0.2 0. sec 25 30 r 5 10 15 20 Time.685 • −3 3 Mode 2 ∆ (cm) u (cm) • 2. (a) Roof Displacement 15 (b) Top Story Drift 3 Mode 1 ∆ (cm) u (cm) 9.1 0.59 • Total (All Modes) 0 −15 0 ∆ (cm) u (cm) 0 −3 0 fig3_8. Additional errors are introduced in pushover analysis due to the approximation inherent in modal combination rules. Response histories of roof displacement and top-story drift from RHA for 0. 3.least three modes are included.

133 0.002 0.05E-03 3.5 -16.274 0.9 3 Modes -12.76E-03 -1.466 0.179 0.062 0.08E-03 2.8 -15.071 0.227 0.89E-03 2.47E-03 1.024 0.385 0.370 0.259 0.235 0.33E-04 5.203 0.9 -12.4 -5.332 0.3 1.31E-03 2.4 1.44E-03 3.260 0.3 -14.9 -13.285 0.276 0.89E-03 2.3 Table 3.03E-03 1.31E-03 Error (%) RHA (all modes) 2.270 0.73E-03 3.229 0.74E-03 -1.2 -0.125 0.060 -0.7 -19.74E-04 -6.264 Error (%) Modal Response 1 Mode 2 Modes 3 Modes -23.042 0.097 0.230 -0.09E-04 -3.117 0.203 0.9 -15.036 -0.6 -17.060 0.237 0.282 -0.2 -11.088 -0.6 -19.173 0.260 0.63E-03 -2.156 0.90E-04 -9.321 0.90E-03 1.9 -14.00E-03 -1.89E-03 1.09E-03 Mode 2 1.267 -0.157 0.152 0.203 0.173 0.25 × El Centro ground motion Floor 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Displacement /Height (%) Modal Response Combined (MPA) Mode 1 Mode 2 Mode 3 1 Mode 2 Modes 3 Modes 0.407 0.3 -41.336 0.3 -19.4 -11.00E-03 2.3 -13.25 × El Centro ground motion Floor 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Modal Response Mode 1 -2.78E-04 2 Modes 2.8 -22.9 -15.097 0.7 -15.99E-03 Mode 3 -3.15E-03 2.2 1.94E-03 2.245 0.12E-03 1.282 0.03E-03 6.63E-03 2.3 -12.069 0.28E-03 2.2 Table 3.231 -0.1 -18.009 0.011 -0.7 -21.40E-04 5.003 -0.0 -2.0 -18.069 0.4 -14.267 0.079 0.0 -0.7 2 Modes -13.4 -2.1 -11.9 -14.090 0.38E-03 2.9 -14.322 0.9 -15.8 -14.38E-03 3 Modes 2.245 -0.042 0.03E-03 3.237 -0.4 -53.3 -2.6 -16.253 0.43E-04 -1.260 -0.286 0.133 -0.04E-03 3.80E-04 3.011 0.048 0.5 Peak values of story drift ratios (as % of story height) from MPA for 0.22E-04 2.267 0.4 -19.89E-03 -2.57E-03 1 Mode -23.7 -50.7 24 .227 0.272 0.008 -0.300 0.25 × El Centro ground motion Story 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Modal Response Mode 1 Mode 2 Drift Ratio (%) Combined (MPA) 1 Mode 2 Modes 3 Modes Mode 3 -0.374 0.043 -0.76E-03 1.121 0.328 0.89E-03 -1.6 Peak values of joint rotation (radians) from MPA for 0.31E-03 Joint Rotation (rad) Combined (MPA) 1 Mode 2.4 0.058 -0.253 RHA (All Modes) 0.177 0.00E-03 1.3 -33.9 -13.4 Peak values of floor displacements (as % of building height = 37.203 0.2 -12.4 -22.9 -13.106 0.055 0.9 -16.080 0.235 -0.03E-03 -1.045 -0.4 -9.3 -9.270 0.048 0.8 -23.00E-03 2.078 0.96E-03 2.062 -0.Table 3.012 -0.401 Error (%) 1 Mode 2 Modes 3 Modes -23.006 -0.015 -0.0 -10.9 -18.6 -15.133 0.09E-03 1.022 0.4 -7.14 m) from MPA for 0.152 0.72E-03 3.9 -11.2 -57.9 -8.023 -0.012 0.9 -15.261 -0.032 0.1 -0.157 0.203 0.2 -20.73E-05 3.253 -0.65E-03 2.1 -19.2 -4.331 0.181 0.1 -11.156 0.0 -46.4 -22.22E-03 -2.313 0.038 -0.266 0.106 0.4 -4.44E-03 -1.92E-04 -1.179 0.24E-03 -2.9 -24.2 -16.197 0.74E-03 1.4 -9.8 -15.230 0.259 -0.7 2.001 0.00E-03 2.336 RHA (all modes) 0.09E-03 -1.8 -56.253 0.177 0.270 0.9 -13.226 0.0 -16.4 -4.229 0.010 0.97E-03 1.42E-04 -1.296 -0.53E-04 -9.026 0.00E-03 3.124 0.310 0.15E-03 1.125 0.

eps fig3_10b.(a) Floor Displacements 9th 8th 7th 6th 9th 8th 7th 6th (b) Story Drift Ratios Floor 4th 3rd 2nd 1st Floor 5th RHA All Modes MPA 1 Mode 2 Modes 3 Modes 5th 4th 3rd 2nd 1st RHA All Modes MPA 1 Mode 2 Modes 3 Modes fig3_10a.2 0. Heightwise variation of floor displacements and story drifts from MPA for 0.3 0.4 Story Drift Ratio (%) 0. 3.10.1 0.3 0. shading indicates modal combination error 25 .6 Fig.1 0.25 × El Centro ground motion.eps Ground 0 0.2 0.4 Displacement/Height (%) 0.5 0.5 Ground 0 0.

26 .

sign u b g g bg (4.1) is not valid for inelastic systems.2) The standard approach is to solve directly these coupled equations. it is useful for later reference to transform Eq. Thus. (4. Although classical modal analysis (Section 3. Expanding the displacements of the inelastic system in terms of the natural vibration modes of the corresponding linear system we get ut = bg n =1 Â φ nqn bt g 27 N (4.3) . Each structural element of this elastic system is defined to have the same stiffness as the initial stiffness of the structural element of the inelastic system. leading to the “exact” nonlinear response history analysis (RHA).4 4. (3. Therefore.1 Inelastic Multistory Buildings RESPONSE HISTORY ANALYSIS For each structural element of a building. the initial loading curve is idealized as bilinear. Eq. the natural vibration periods and modes of the corresponding linear system are the same as the vibration properties of the inelastic system undergoing small oscillations (within the linear range). sign u = .mι ug t b (4.1) With this generalization for inelastic systems.2) to the modal coordinates of the corresponding linear system. the relations between lateral forces f s at the N floor levels and the lateral displacements u are not single valued. Both systems have the same mass and damping. and the unloading and reloading curves differ from the initial loading branch. but depend on the history of the displacements: f s = f s u.1) becomes mu + cu + f s u.

premultiplying by φ T . The spatial distribution s of the effective earthquake forces is expanded into the modal contributions sn according to Eq. (3. This approximate RHA procedure is the preliminary step in developing a modal pushover analysis procedure for inelastic systems. (4.2).4)] leads to the uncoupled modal response history analysis (UMRHA) procedure.2). Simultaneously solving these coupled equations and using Eq. However. Unlike Eq.8) because qr t will generally be nonzero for “modes” other than the nth “mode. 2. (4. sign qn = φ n f s u n . For linear systems. qr t = 0 for all 28 bg . (4.9) for linearly elastic systems.2).9) involves T Fsn = Fsn qn .6) for inelastic systems will no longer be described by Eq.6) The solution of Eq. The equations governing the response of the inelastic system to p eff.Substituting Eq. implying coupling of modal coordinates because of yielding of the structure. (4.n t given by Eq.3) in Eq. however.sn ug t bg b g bg bg (4. (3. (3. where φ n are now the modes of the corresponding linear system. and using the mass.3) will.4) represents N equations in the modal coordinates qn . give the same results for u t as obtained directly from Eq.” implying that other “modes” will also contribute to the solution.4) is rarely solved because it offers no particular advantage over Eq.… N (4. 4. (3.4) where the only term that differs from Eq. (4. Equation (4.2 UNCOUPLED MODAL RESPONSE HISTORY ANALYSIS bg Neglecting the coupling of the N equations in modal coordinates [Eq. (4. (3.3). sign u n b g b g bg (4.5) This resisting force depends on all modal coordinates qn t .and classical n damping-orthogonality property of modes gives F qn + 2z nw n qn + sn = . (4. sign u = . in principle. Eq.6b) are mu + cu + f s u. these equations are coupled for inelastic systems.Gn ug t Mn bg n = 1. (4.

(a) p 15

eff,1

= −s × 0.25 × El Centro

1

(b) p 5

eff,2

= −s × 0.25 × El Centro

2

ur1 (cm)

0

ur1 (cm)

9.112 •

Mode 1

Mode 1

0

−15 15 Mode 2

−5 5 Mode 2

ur2 (cm)

0

ur2 (cm)

0 • 2.226 5 10 15 20 25 30

−15 0 15

5

10

15

20

25

30

Mode 3

−5 0 5

Mode 3

ur3 (cm)

0

ur3 (cm)

5 10 15 20 Time (sec) 25 30

0

−15 0

−5 0

5

10 15 20 Time (sec)

25

30

Fig. 4.1. Modal decomposition of the roof displacement due to (a) p eff,1 t = - s1 ¥ 0.25

**¥ El Centro ground motion; and (b) p eff,2 t = - s 2 ¥ 0.25 ¥ El Centro ground
**

motion

af

af

modes other than the nth-mode; therefore, it is reasonable to expect that the nth “mode” should be dominant even for inelastic systems. These assertions are illustrated numerically in Figs. 4.1 and 4.2 for the selected 9-story building. Equation (4.6) was solved by nonlinear RHA, and the resulting roof displacement history was decomposed into its “modal” components. The modal decomposition of the roof displacement for the first three modes due to 0.25 ¥ El Centro ground motion demonstrates that because the building does not yield during this weak ground motion, the response to excitation p eff,n t is all in the nth-mode (Fig. 4.1). The structure yields when subjected to the strong excitation of 1.5 ¥ El Centro ground motion, and the modes other than the nth-mode contribute to the response. The second and third modes start responding to excitation p eff,1 t at about 29

bg

bg

(a) p 80

eff,1

= −s × 1.5 × El Centro

1

(b) p 20

eff,2

= −s × 1.5 × El Centro

2

Mode 1

Mode 1 • 1.437

u (cm)

r1

0 • 48.24

ur1 (cm)

Mode 2

0

−80 80

−20 20 Mode 2

u (cm)

r2

0

3.37 •

ur2 (cm)

0 • 11.62 5 10 15 20 25 30

−80 0 80

5

10

15

20

25

30

Mode 3 0.4931 •

−20 0 20

Mode 3 • 0.7783

u (cm)

r3

0

ur3 (cm)

25 30

0

−80 0

5

10 15 20 Time (sec)

−20 0

5

10 15 20 Time (sec)

25

30

Fig. 4.2. Modal decomposition of the roof displacement due to: (a) p eff,1 t = - s1 ¥ 15 ¥ . El Centro ground motion; and (b) p eff,2 t = - s2 ¥ 15 ¥ El Centro ground . motion

af

af

5.2 sec, the instant the structure first yields; however, their contributions to the roof displacement are only 7% and 1%, respectively, of the first mode response (Fig. 4.2a). The first and third modes start responding to excitation p eff,2 t at about 4.2 sec, the instant the structure first yields; however, their contributions to the roof displacement are 12% and 7%, respectively, of the second mode response (Fig. 4.2b). Approximating the response of the structure to excitation p eff,n t

bg

bg

by Eq. (3.8),

substituting Eq. (3.8) in Eq. (4.6) and premultiplying by φ T gives Eq. (4.4), except for the n important approximation that Fsn now depends only on one modal coordinate, qn :

T Fsn = Fsn qn , sign qn = φ n f s qn , sign qn

b

g

b

g

(4.7)

30

**With this approximation, solution of Eq. (4.4) can be expressed by Eq. (3.11) where Dn t is governed by F Dn + 2z nw n Dn + sn = - ug t Ln and
**

T Fsn = Fsn Dn , sign Dn = φ n f s Dn , sign Dn

bg

bg c h

(4.8)

c

h

(4.9)

is related to Fsn qn , sign qn because of Eq. (3.11). Equation (4.8) may be interpreted as the governing equation for the nth-“mode” inelastic SDF system, an SDF system with (1) small amplitude vibration properties—natural frequency

b

g

**w n and damping ratio z n —of the nth-mode of the corresponding linear MDF system; (2) unit
**

mass; and (3) Fsn Ln - Dn relation between resisting force Fsn Ln and modal coordinate Dn defined by Eq. (4.9). Although Eq. (4.4) can be solved in its original form, Eq. (4.8) can be solved conveniently by standard software because it is of the same form as the SDF system [Eq. (2.3)], and the peak value of Dn t can be estimated from the inelastic response (or design) spectrum (Chopra, 2001; Sections 7.6 and 7.12.1). Introducing the nth-“mode” inelastic SDF system also permitted extension of the well-established concepts for elastic systems to inelastic systems. Compare Eqs. (4.4) and (4.8) to Eqs. (3.9) and (3.10): note that Eq. (3.11) applies to both systems.4 Solution of the nonlinear Eq. (4.8) formulated in this manner provides Dn t , which substituted into Eq. (3.12) gives the floor displacements of the structure associated with the nth“mode” inelastic SDF system. Any floor displacement, story drift, or another deformation response quantity r t is given by Eqs. (3.13) and (3.14), where An t is now the pseudost acceleration response of the nth-“mode” inelastic SDF system. The two analyses that lead to rn

bg

bg

bg

bg

b g are shown schematically in Fig. 4.3. Equations (3.13) and (3.14) represent the response of the inelastic MDF system to p eff,n bt g , the nth-mode contribution to p eff bt g . Therefore the response of the system to the total excitation p eff bt g is given by Eqs. (3.15) and

and An t (3.16). This is the UMRHA procedure.

4

Equivalent inelastic SDF systems have been defined differently by other researchers (Villaverde, 1996; Han and Wen, 1997).

31

n t = .s n u g t . and 3. Roof displacement due to p eff.0 ¥ El Centro ground motion: (a) exact solution by NL-RHA.4. Conceptual explanation of uncoupled modal response history analysis of inelastic MDF systems (a) Nonlinear RHA 150 n=1 150 (b) UMRHA n=1 ur1 (cm) • 75. and (b) approximate solution by UMRHA bg af bg 32 .575 5 10 15 20 Time (sec) 25 30 ur3 (cm) 0 • 5.Forces sn Unit mass An(t ) st rn ωn. Fsn / Ln ¨ g(t ) u (a) Static Analysis of Structure (b) Dynamic Analysis of Inelastic SDF System Fig.02 −150 n=2 −150 50 n=2 14.9 • 0 −50 10 −50 10 n=3 ur3 (cm) 0 • 5. ζn.3.51 • ur2 (cm) 0 ur2 (cm) n=3 14.89 50 r1 0 u (cm) 0 • 78. n = 1.112 5 10 15 20 Time (sec) 25 30 −10 0 −10 0 Fig. where u g t = 3. 2. 4. 4.

the errors in either response quantity are only a few percent.744 • 0 −20 10 n=3 −20 10 n=3 ∆r3 (cm) 0 • 5. but even for this very intense excitation.4) and (4.0 ¥ El Centro ground motion. n = 1. 4.0 ¥ El Centro ground motion: (a) exact solution by NL-RHA. where u g t = 3.5)] is neglected. (4. Top story drift due to p eff.817 5 10 15 20 Time (sec) 25 30 −10 0 −10 0 Fig. (4.s n u g t .4 and 4.15) is strictly valid only for linearly elastic systems.5. Such comparison for roof displacement and top-story drift is presented in Figs.2. both for 3. (2) the superposition of responses to p eff. (3. respectively.n t (n = 1.6) by UMRHA is compared with the “exact” solution by nonlinear RHA. and 3.Dn relation is 33 bg bg . and (b) approximate solution by UMRHA af bg bg 4.855 ∆r1 (cm) 0 ∆r1 (cm) n=2 • 6.956 5 10 15 20 Time (sec) 25 30 ∆r3 (cm) 0 • 5.2… N ) according to Eq. and (3) the Fsn Ln . These errors arise from the following assumptions and approximations: (1) the coupling among modal coordinates qn t arising from yielding of the system [recall Eqs. 2.008 • 6.33 0 −20 20 −20 20 n=2 ∆r2 (cm) 0 ∆r2 (cm) 7. The errors are slightly larger in drift than in displacement. 4.(a) Nonlinear RHA 20 n=1 20 (b) UMRHA n=1 • 5.5.1 Underlying Assumptions and Accuracy The approximate solution of Eq. this intense excitation was chosen to ensure that the structure is excited well beyond its linear elastic limit.n t = .

8) governing Dn t is based on Eq. within the linearly elastic range of the structure. and the base shear Vbn .approximated by a bilinear curve to facilitate solution of Eq. it can conduct a force-controlled nonlinear static analysis with an invariant distribution of lateral forces. 4. is plotted against roof displacement urn .12) for floor displacements.2 Properties of the nth-“mode” Inelastic SDF System How is the Fsn Ln .6a.Dn relation to be determined in Eq.urn pushover curve to the Fsn Ln . such nonlinear static analysis provides the so-called pushover curve. A bilinear idealization of this pushover curve for the nth-“mode” is shown in Fig. (4. the base shear is Vbny and roof displacement is urny . (4.9). What is an appropriate invariant distribution of lateral forces to determine Fsn ? For an inelastic system no invariant distribution of forces can produce displacements proportional to f n at all displacements or force levels.2. However.Dn relation? The two sets of forces and displacements are related as follows: V Fsn = bn Γn Dn = urn Γ nf rn 34 (4. The structure is pushed using the force distribution of Eq. (4. (4.20). (3. Therefore. the relationship between lateral forces f s and Dn in Eq. when specialized for linearly elastic systems it is identical to the RHA procedure of Section 3. At the yield point. Although several approximations are inherent in this UMRHA procedure. this distribution seems to be a rational choiceeven for inelastic systemsto determine Fsn in Eq.4. (4. How to convert this Vbn .20) to some bg pre-determined roof displacement. 4. Although most commercially available software cannot implement such displacement-controlled analysis.9) should be determined by nonlinear static analysis of the structure as the structure undergoes displacements u = Dnφ n with increasing Dn . (3.10) .Dn curve. (3. When implemented by commercially available software.8) in UMRHA.8) before it can be solved? Because Eq. which is different than the Fsn Ln . The overall errors in the UMRHA procedure are documented in the examples presented in Section 4. Therefore we impose this constraint in developing the UMRHA procedure in this section and modal pushover analysis in the next section. the only force distribution that produces displacements proportional to f n is given by Eq.1.

4. 4. 4. The two are related through Fsny Ln 2 = w n Dny (4.8). (4.13) This value of Tn . where the yield values of Fsn Ln and Dn are Fsny Ln = Vbny * Mn Dny = urny Γ nf rn (4. 2001. Section 13. (4. should be used in Eq. which may differ from the period of the corresponding linear system.11) * in which M n = Ln Γn is the effective modal mass (Chopra.V bn (a) Idealized Pushover Curve F sn /L n (b) F sn / L − D Relationship n n Idealized Vbny Actual 1 αnkn V bny /M * n 1 α ω2 n n kn 1 1 ωn 2 u u rny rn Dny = ur n y / Γn φ r n D n Fig. 35 .10 enables conversion of the pushover curve to the desired Fsn Ln − Dn relation shown in Fig. 4.6b is w 2 . which is not a meaningful quantity.12) implying that the initial slope of the curve in Fig. the elastic vibration period Tn of the nth-mode SDF system is computed from F Ln Dny I 1/ 2 Tn = 2p G H Fsny JK (4.6. Properties of the nth-“mode” inelastic SDF system from the pushover curve Equation 4.2. the initial slope of the pushover curve in Fig.6b.11). Knowing Fsny Ln and Dny from n Eq. In contrast.6a is 2 kn = ω n Ln .5).

4. 7. for linearly-elastic vibration of the building. n (3. the first two or three modes will suffice.3 Summary The inelastic response of an N-story building with plan symmetric about two orthogonal axes to earthquake ground motion along an axis of symmetry can be estimated as a function of time by the UMRHA procedure just developed.13). Consider a nonlinear static analysis of the structure subjected to lateral forces distributed over the building height according to s* [Eq.20)]. ωn . details are available in Appendix A: 1. 4. (3. 8.2. Dn (t ) .6a). the peak value of Dn t .n t . An t . For the nth-mode. Convert the idealized pushover curve to the Fsn Ln − Dn relation (Fig. Idealize the pushover curve as a bilinear curve (Fig. 6. and modes. is now determined by 36 bg bg . 4. 4. 4. with the structure is pushed to the roof displacement urno . of the total response r t obtained in Step 8. Repeat Steps 2 to 6 for as many modes as required for sufficient accuracy. Calculate the peak value. Compute the natural frequencies. Combine the “modal” responses using Eqs.6b) by utilizing Eq. (3. (3.4.21) where Dn . of the nth-“mode” inelastic SDF system (Fig. and pseudo-acceleration history. n 3. 5. This value of the roof displacement is given by Eq.6b.15) and (3. 4. φn .16) to determine the total response. Calculate histories of various responses by Eqs. 9. (4. 2.20)]. Compute the deformation history. (3.3b) with force-deformation relation of Fig.3 MODAL PUSHOVER ANALYSIS bg bg A pushover analysis procedure is presented next to estimate the peak response rno of the inelastic MDF system to effective earthquake forces p eff . Typically. r o . develop the base-shear – roof-displacement ( Vbn − urn ) pushover curve for the force distribution s* [Eq.11).12) and (3. which is summarized next as a sequence of steps.

12). Calculate the peak roof displacement urno associated with the nth-“mode” inelastic SDF system from Eq. details are available in Appendix B. The peak “modal” responses rno . As shown in Sections 3.8). 4..6b by solving Eq.6).2. e.3. alternatively. However.18). which is summarized next as a sequence of steps. each determined by one pushover analysis. and (2) it provides the exact modal response for elastic systems.g.2 and 3. 6. 37 .8). Dn .3b) with force-deformation relation of Fig.solving Eq. etc. the lateral force distribution used possesses two properties: (1) it appears to be the most rational choice among all invariant distribution of forces. It provides results for elastic buildings that are identical to the well-known RSA procedure (Section 3. the pushover analysis provides an estimate of the peak value rno of any response rn t : floor displacements.3) because. joint rotations. it seems reasonable because it provides results for elastic buildings that are identical to the well-known RSA procedure described in Section 3. we will refer to rno as the peak “modal” response even in the case of inelastic systems. 5.1 Summary The peak inelastic response of a building to earthquake excitation can be estimated by the MPA procedure just developed. (3. Steps 1 to 4 of the MPA are same as those for UMRHA.4. At this roof displacement. seems reasonable. story drifts. Eq.n t . The response value rno is an estimate of the peak value of the response of the inelastic system to p eff. or from the inelastic response (or design) spectrum. (4. bg bg bg bg Thus. as mentioned earlier. 2001. (4. This pushover analysis. (4. plastic hinge rotations. it can be determined from the inelastic response (or design) spectrum (Chopra. as described in Section 4.6 and 7. (3. rno also represents the exact peak value of the nth-mode contribution rn t to response r t . This application of modal combination rules to inelastic systems obviously lacks a theoretical basis.3.2. Sections 7. although somewhat intuitive for inelastic buildings. of the nth-“mode” inelastic SDF system (Fig. Compute the peak deformation. for elastic systems. 4. are combined using an appropriate modal combination rule.21). to obtain an estimate of the peak value ro of the total response. 4. governed by Eq.

4. The peak values of floor displacements and story drifts including one. Figure 4. 4. and three modes are compared with the “exact” values in Fig. rno . also included are the combined responses due to one.5 ¥ the El Centro ground motion including the response contributions associated with three “modal” inelastic SDF systems. (3. and the errors in the approximate results are shown in Fig.7 shows the individual “modal” responses.4. and the “exact” response from nonlinear RHA for the roof displacement and top-story drift.2. the “exact” results. the UMRHA procedure lacks a rigorous theory. Determine the total response by combining the peak “modal” responses using the SRSS combination rule of Eq. The peak values of displacements of all floors and drifts in all stories are presented in Tables 4. two. and three “modes”. The peak values of response are as noted. two.0. the combined response due to three “modes”.7 cm. the 38 .2). 8. 4. the first two or three “modes” will suffice.8. with.53 cm.1). determined by the UMRHA procedure.4. Observe that errors tend to decrease as response contributions of more “modes” are included. respectively. 4. in particular. in contrast to modal analysis (Section 3. This deficiency also implies that. This is to be expected. ur 2o = 11. and the percentage errors in the approximate results. and compared with the results of a rigorous nonlinear RHA using the DRAIN-2DX computer program. Typically. 9.4 COMPARATIVE EVALUATION OF ANALYSIS PROCEDURES The response of the 9-story building described earlier is determined by the two approximate methods: UMRHA and MPA. From the total rotation of a plastic hinge. is presented next.0 to 3.1 and 4.7. although the trends are not systematic as when the system remained elastic (Section 3. say. and ur 3o = 2. extract from the pushover database values of other desired responses.1 Uncoupled Modal Response History Analysis The structural response to 1. subtract the yield value of hinge rotation to determine the hinge plastic rotation.18). To ensure that this structure responds well into the inelastic range the El Centro ground motion is scaled up by a factor varying from 1.9. At urno . the peak roof displacement due to each of the three “modes” is ur1o = 48. three “modes” included.3 cm. Repeat Steps 3 to 8 for as many “modes” as required for sufficient accuracy.

recall that the computed errors have been presented earlier for ground motion multipliers 0.25. However. For this purpose the UMRHA and exact analyses were repeated for ground motions of varying intensity. For each excitation.5 × El Centro ground motion: individual “modal” responses and combined response from UMRHA. Next.24 • NL−RHA r −12 12 ∆r (cm) u (cm) 0 5 10 15 20 Time.38 6. defined as the El Centro ground motion multiplied by 0. 0. 0. we investigate how the errors in the UMRHA vary with the deformation demands imposed by the ground motion.88 u (cm) "Mode" 3 −80 80 0 −80 80 0 −80 0 • 44. 2. sec 25 30 • 48.2).0.2). sec 25 30 r −12 0 Fig. Response histories of roof displacement and top-story drift due to 1.1 and 4. 1.response is much less accurate (Tables 4.5.1 and 3.25 (Tables 3.1 −12 12 ∆r (cm) u (cm) UMRHA 3 "Modes" 0 • 7.75.0. just as for elastic systems. (a) Roof Displacement (b) Top Story Drift 80 12 0 • 3.3 11. for a fixed number of “modes” included.53 ∆r3 (cm) 12 0 • 2.1 and 4.0. in particular.7.6 5 10 15 20 Time.7 • −80 80 ∆r1 (cm) u (cm) "Mode" 1 −12 r2 0 ∆r2 (cm) 12 0 u (cm) −80 80 −12 r3 0 • 2.85. the errors in responses computed by UMRHA including three “modes” relative to the “exact” response were determined. 1.1 and 3. the degree to which the system deforms beyond its elastic limit. 4.5.2) if the system yields significantly versus if the system remains within the elastic range (Tables 3. the errors in story drifts are larger compared to floor displacements.62 5. 0. and total response from NL-RHA 39 . and 3.44 • "Mode" 2 r1 0 • 48.5 (Tables 4.2) and 1.

(a) Floor Displacements 9th 8th 7th 6th 9th 8th 7th 6th NL−RHA UMRHA 1 "Mode" 2 "Modes" 3 "Modes" (b) Story Drift Ratios Floor Floor 5th 4th 3rd 2nd 1st 5th 4th 3rd 2nd 1st NL−RHA UMRHA 1 "Mode" 2 "Modes" 3 "Modes" Ground 0 0.5 1 1.5 Fig. Shown is the error in each floor displacement (Fig.8. (3. Heightwise variation of floor displacements and story drift ratios from UMRHA and NL-RHA for 1.5 Displacement/Height (%) 2 Ground 0 0. it will be useful to know the deformation of the system relative to its yield deformation.5 1 1. 4.5 Story Drift Ratio (%) 2 2.5 ¥ El Centro ground motion Figure 4. Heightwise variation of error in floor displacements and story drifts estimated by UMRHA including one.5 × El Centro ground motion (a) Floor Displacements 9th 8th 7th 6th 9th 8th 7th 6th (b) Story Drift Ratios Floor 4th 3rd 2nd 1st UMRHA 1 "Mode" 2 "Modes" 3 "Modes" Floor 5th 5th 4th 3rd 2nd 1st UMRHA 1 "Mode" 2 "Modes" 3 "Modes" Ground −60 −40 −20 0 Error (%) 20 40 60 Ground −60 −40 −20 0 Error (%) 20 40 60 Fig.10b).10 summarizes the error in UMRHA as a function of ground motion intensity. For this purpose. Two versions of the pushover curve are 40 . the pushover curves using force distributions s* [Eq. indicated by a ground motion multiplier. 4.11. and the error envelope for each case.10a). with the peak displacement of each “modal” SDF system noted for each ground motion multiplier. or three “modes” for 1. 4. 4. two.9. 4.20)] for the first three n modes of the system are shown in Fig. in each story drift (Fig. To interpret these results.

Second. 4. Additional errors are introduced in UMRHA of systems responding beyond the linearly elastic limit for at least two reasons. Each of these three pushover analyses provides the pushover curve (Fig. the pushover curve for each “mode” is idealized by a bilinear curve in solving Eq.10) for each “modal” inelastic SDF system (Figs 4.6 and 4. UMRHA lacks a rigorous theory and is based on several approximations.1). Figure 4.10 permits the following observations regarding the accuracy of the UMRHA procedure: the errors (1) are small (less than 5%) for ground motion multipliers up to 0. The location of plastic hinges and their rotations. even though the system remains essentially elastic.11a).5 ¥ the El Centro ground motion. 4. The idealized curve for the first “mode” deviates most from the actual curve near the peak displacement corresponding to ground motion multiplier 1.20) with n = 1. and 3 (Fig. (4.included: the actual curve and its idealized bilinear version. 2.5).7). the peak values of displacements at all floors (Table 4. Perhaps this explains why the errors are large at this excitation intensity. 11.2. the errors in truncating the higher mode contributions are negligible.3 cm. 4. and (4) are larger in story drift compared to floor displacements.2.75. (2) increase rapidly as the ground motion multiplier increases to 1. The structure is pushed using the force distribution of Eq. the ductility factor for the first mode system is only 1.2 Modal Pushover Analysis The results of modal pushover analysis procedure considering the response due to the first three “modes” was implemented for the selected building subjected to 1. drifts in all stories (Table 4. 3.4) to roof displacements urno = 48. and 2.0. as mentioned in Section 3. (3) maintain roughly similar values for more intense ground motions. respectively.01 (Fig. determined from “exact” analyses. the values determined by RHA of the nth-mode inelastic SDF system (Fig. (3. and. were noted but are not shown here.11). For more intense excitations.0. The system remains elastic up to ground motion multiplier 0. the first reason mentioned above seems to be the primary source for the errors. 4. First.4). 41 .53 cm.3).7 cm.4.75.4. and plastic hinge rotations at the external beam end at each floor level (Table 4. as mentioned in Section 4.

5 28.942 -0.1 3.5 18.554 1.169 0.7 Table 4.350 -0.371 0.256 -1.5 × El Centro ground motion Floor 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Displacement /Height (%) “Modal” Response Combined (UMRHA) “Mode” 1 “Mode” 2 “Mode” 3 1 “Mode” 2 “Modes” -0.315 -0.727 1.410 1.8 1.044 1.407 -10.5 2 GM Multiplier 2.133 1.668 -23.2 22.201 -0.852 1.256 1.2 4.616 -0.663 -0.410 -1.379 1.5 1 1.003 -31.130 0.6 2.6 1.298 0.676 0.072 -1.371 -0.3 42 .5 9.2 1.3 1.484 0.5 -3.982 9.057 -0.2 6.068 0.490 1.009 -0.200 8.945 -37.154 0.811 1.9 5.1 Peak values of floor displacements (as % of building height = 37.372 1.121 -0.338 -1.914 2.863 0.877 0.9 31.473 -22.4 -1.751 1.9 12.663 0.372 -1.3 6.376 -1.333 0.088 10.055 -0.844 -25.226 -0.003 0.135 9.010 0.366 -0. Noted 7 6 Error (%) 30 1 Error (%) 30 5 8 20 6 5 2 20 2 4 9 9 10 0 0 10 3 8 7 3 1 4 0.3 25.526 -0.214 -0.126 0.0 11.202 11.5 9.707 1.698 1.138 1.370 -0.820 -0.942 1.214 0.049 -0.0 -9.5 × El Centro ground motion Story 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Modal Response “Mode” 1 “Mode” 2 Drift Ratio (%) Combined (UMRHA) 1 “Mode” 2 “Modes” “Mode” 3 -1.722 0.033 0.430 1.256 1.478 0.071 -0.2 Peak values of story drift ratios (as % of story height) from UMRHA for 1.187 -0. Errors in UMRHA as a function of ground motion intensity: (a) floor displacements.540 0.0 11.241 -1.241 1.5 3 Fig.5 2 GM Multiplier 2.235 -0.0 2.4 1.298 -0.248 Error (%) NL 3 1 2 3 RHA “Mode” “Modes” “Modes” “Modes” 0.938 -1.5 28. 4.366 0.4 -7.018 0.201 1.495 1.8 1.0 1.0 7.291 0.8 1.8 1. and (b) story drifts Table 4.072 1.971 1.1 1.9 12.6 4.5 3 0 0 0.0 9.376 1.1 8.096 0.490 -1.900 -10.8 14.513 -0.201 -1.806 0.1 0.9 16.4 4.1 1.260 -15.819 2.373 -0.070 1.513 0.065 0.472 1.914 -0.136 1.856 2.079 0.120 1.772 Error (%) NL 3 1 2 3 RHA “Mode” “Modes” “Modes” “Modes” 1.9 31.216 1.8 1.2 12.(a) Floor Displacements 60 50 40 Error Envelope Error for Floor No.5 1 1.938 1.220 -0.426 -1.3 8.220 0.820 -19.575 -41.10.5 10.317 0.293 1.3 0.283 1.763 -15.14 m) from UMRHA for 1.863 1.806 -0.338 1.071 0.7 31.8 0.7 14.2 1.983 1.104 0. Noted 60 50 40 (b) Story Drifts Error Envelope Error for Story No.

Fortuitously.18).” and nonlinear RHA. and three “modes. considering one.13 and Tables 4.11c). 4. however.12 and 4. 4.1 and 4.” and three “modes. Observe that the primary contributor to plastic hinge rotations in the lower stories is the first “mode” and the second “mode” in the upper stories. but was more successful when the excitation was more intense (results are not included). overestimated by up to 18% in the middle stories. significantly smaller than in UMRHA (compare Fig. for two or three modes included.” two “modes.” respectively.13 with Fig.3 and 4.4). were determined by four analyses: MPA considering one “mode.4).5).14. and are within a few percent of the exact values for the upper stories. significant improvement is achieved by including response contributions due to the second “mode”. Furthermore. The second “mode” was necessary to identify hinges in the upper stories. even if three “modes” are included (Fig. However. the additional errors due to the approximation inherent in modal combination rules tend to cancel out the errors due to the various approximation contained in the UMRHA. 4. 4.3 and 4. 4. 43 . The drifts are underestimated by up to 13% in the lower stories.2). although the error is recorded as 100% if MPA estimates zero rotation when the nonlinear RHA computes a non-zero value. the third “mode” does not contribute because this SDF system remains elastic (Fig.3 through 4. Pushover analysis seems to be inherently limited in computing accurately hinge plastic rotations.14.Figures 4. but the third “mode” contributions do not seem especially important (Fig. One “mode” pushover analysis was unable to identify the plastic hinges in the upper stories where higher mode contributions to response are known to be more significant.3 and 4. For example.4. Obviously. in general. especially in estimating the story drifts (Fig.13 and Table 4. the errors in the modal pushover results are.5 present estimates of the combined response according to Eq.9 and Tables 4. The locations of plastic hinges shown in Fig. 4. MPA including three “modes” underestimates the displacements of the lower floors by up to 8% and overestimates the upper floor displacements by up to 14%.13 and Tables 4.3 and 4. this error is not especially significant because the hinge plastic rotation is very small. The first “mode” alone is inadequate. 4. 4. The errors are especially large in the hinge plastic rotations estimated by the MPA procedure. As shown in Figs.12 and Tables 4. 4. (3.4 with Tables 4.12 and Tables 4. MPA failed to identify the plastic hinges at the column bases in Fig. the results were not always accurate. the hinges identified in beams at the 6th floor were at variance with the “exact” results. two. and the errors in these estimates relative to the exact response from nonlinear RHA.

2 cm. V = 7616 kN.85 Base Shear (kN) 8000 6000 4000 2000 0 0 0.25. and 3.0 × El Centro ground motion 44 . 0. 4. 1.5 0.85 0.(a) "Mode" 1 Pushover Curve 12000 10000 u = 36.5 0.25 1 2 4 6 Roof Displacement (cm) 8 10 Fig. V = 5210 kN. “Modal” pushover curves with peak roof displacements identified for 0.0.5 1 0. α = 0.11.25 0.5 0. V = 4952 kN.9 cm.75.75 0.0.5.85 0.25 5 10 15 20 Roof Displacement (cm) 25 (c) "Mode" 3 Pushover Curve 12000 10000 u = 4.75 Actual Idealized 20 40 60 Roof Displacement (cm) (b) "Mode" 2 Pushover Curve 80 12000 10000 u = 9. α = 0.5 2 3 Actual Idealized 2000 0 0 0. 1.13 y by Base Shear (kN) 8000 6000 1 1. 2.5 2 3 4000 2000 0 0 Actual Idealized 0.14 y by Base Shear (kN) 8000 6000 4000 1. 0.6 cm.19 y by 3 2 1. α = 0.5.75 0.

Shown is the error in each floor displacement (Fig. implying excitations intense enough to cause significant yielding of the structure.3 for elastic systems (or weak ground motions). The errors are only weakly dependent on ground motion intensity (Fig. indicated by a ground motion multiplier. As mentioned in Section 3. the errors in MPA were fortuitously smaller than in UMRHA (compare Figs. 4.15a). and the error envelope for each case. While various sources of errors in UMRHA. each story drift (Fig.15b). as discussed in Sections 3.(a) Floor Displacements 9th 8th 7th 6th 9th 8th 7th 6th NL−RHA MPA 1 "Mode" 2 "Modes" 3 "Modes" (b) Story Drift Ratios Floor Floor 5th 4th 3rd 2nd 1st 5th 4th 3rd 2nd 1st NL−RHA MPA 1 "Mode" 2 "Modes" 3 "Modes" Ground 0 0.10 and 4.4 also apply to MPA. In this case. UMRHA is essentially exact. 4. 4. whereas MPA contains errors inherent in modal combination rules. implying excitations weak enough to limit the response in the elastic range of the structure.4. implying that the modal combination errors contained in these procedures are acceptable. the MPA procedure is equivalent to the RSA procedure.75.5 1 1.15).12 Heightwise variation of floor displacements and story drift ratios from MPA and NL-RHA for 1. However.5 1 1. The fact that MPA is able to estimate the response of buildings responding well into the inelastic range to a similar degree of accuracy indicates that this procedure is accurate enough for practical application in building retrofit and design. 4.0. an observation with practical implications. 45 .4. 4.5 Story Drift Ratio (%) 2 2.15 summarizes the error in MPA considering three “modes” as a function of ground motion intensity.5 × El Centro ground motion. shading indicates errors in MPA including three “modes” Figure 4. now standard in engineering practice.3. the errors in MPA were larger for ground motion multipliers less than 0.5 Fig. identified in Section 3.15) for ground multipliers larger than 1.2 and 3.5 Displacement/Height (%) 2 Ground 0 0.

and hinge plastic rotations estimated by MPA including one. Errors in floor displacements. 4. two. and three “modes” for 1.13.(a) Floor Displacements 9th 8th 7th 6th Floor 5th 4th 3rd 2nd 1st MPA 1 "Mode" 2 "Modes" 3 "Modes" Ground −60 −40 −20 0 Error (%) 20 40 60 (b) Story Drift Ratios 9th 8th 7th 6th Floor 5th 4th 3rd 2nd 1st MPA 1 "Mode" 2 "Modes" 3 "Modes" Ground −60 −40 −20 0 Error (%) 20 40 60 (c) Hinge Plastic Rotations 9th 8th 7th 6th Floor 5th 4th 3rd 2nd 1st MPA 1 "Mode" 2 "Modes" 3 "Modes" Ground −120 −80 −40 0 Error (%) 40 80 120 Fig.5 ¥ El Centro ground motion 46 . story drifts.

503 -1.8 -12.3 Peak values of floor displacements (as % of building height = 37.76E-03 4.338 1.8 7.02E-03 3.2 -100.331 1.945 -49.72E-03 7.6 13.220 1.99E-03 6.752 1.3 11.55E-03 3.007 1.00E+00 0.1 -8.071 -0.910 1.176 0.37E-03 1.00E+00 0.9 1.1 13.737 1.233 1.527 -0.37E-04 Error (%) “Modal” Response 1 “Mode” -32.641 1.2 -100.9 7.00E+00 0.5 × El Centro ground motion Story 1st 2nd 3rd 4th 5th 6th 7th 8th 9th “Modal” Response “Mode” 1 “Mode” 2 Drift Ratio (%) Combined (MPA) 1 “Mode” 2 “Modes” 3 “Modes” Error (%) NL 1 2 3 RHA “Mode” “Modes” “Modes” 1.60E-03 2.50E-10 3.02E-03 0.00E+00 0.00E+00 0.209 1.640 -1.00E+00 0.351 -0.5 × El Centro ground motion Displacement /Height (%) “Modal” Response Combined (MPA) “Mode” 1 “Mode” 2 “Mode” 3 1 “Mode” 2 “Modes” Floor 1st 2nd 3rd 4th 5th 6th 7th 8th 9th 0.0 -5.00E+00 0.7 1.429 0.0 3 “Modes” -32.372 0.666 Table 4.298 -0.2 0.911 0.1 46.514 -1.1 1.6 -44.737 0.36E-03 6.4 1.00E+00 0.6 -8.0 -100.05E-03 2.8 -29.9 2.53E-03 7.250 0.116 1.055 0.3 -3.0 -100.0 1.088 12.00E+00 0.498 1.012 1.2 6.9 -100.00E+00 0.116 1.2 1.5 Peak values of hinge plastic rotations (radians) from MPA for 1.260 -14.88E-03 0.003 -16.3 -3.60E-04 7.336 Error (%) NL 3 1 2 3 RHA “Mode” “Modes” “Modes” “Modes” 0.049 -0.1 62.8 1.980 0.745 1.125 -1.19E-10 3.705 1.018 0.694 1.781 0.101 -0.5 7.9 -4.407 -27.066 -0.00E+00 0.581 0.233 1.156 -0.0 -50.1 62.76E-03 4.6 -44.130 0.640 1.00E+00 3 “Modes” 7.00E+00 3.756 0.0 47 .053 -1.8 17.154 0.473 -15.22E-10 NL RHA 1.399 0.5 10.00E+00 1 “Mode” 7.982 13.76E-03 4.594 -1.1 62.0 15.733 1.781 0.197 -0.611 0.5 1.071 0.37E-03 1.14 m) from MPA for 1.02E-03 0.668 -13.36E-03 6.18E-03 7.652 1.10E-02 9.36E-03 6.9 5.053 1.00E+00 0.5 Table 4.435 0.503 1.018 -0.5 10.305 -0.311 0.55E-03 3.00E+00 0.190 -0.015 0.36E-03 6.00E+00 Floor 1st 2nd 3rd 4th 5th 6th 7th 8th 9th “Mode” 2 0.3 13.3 1.266 -0.518 1.6 7.9 0.614 0.820 -7.8 1.222 0.00E+00 0.068 0.7 1.168 -0.02E-03 3.018 -0.116 1.478 0.517 1.8 -29.895 1.26E-04 9.9 1.756 0.2 -4.200 8.259 1.315 0.00E+00 0.00E+00 0.88E-03 1.371 -0.009 0.0 -100.0 2 “Modes” -32.8 -29.2 -3.980 -0.222 0.298 0.5 2.72E-03 7.399 0.683 1.4 Peak values of story drift ratios (as % of story height) from MPA for 1.118 0.687 0.900 -0.1 46.6 13.4 -8.88E-03 0.6 1.244 0.033 -0.2 0.8 0.8 -6.007 1.1 18.895 1.55E-03 3.00E+00 2 “Modes” 7.2 11.763 -14.7 -12.8 -4.0 -50.015 0.844 -7.Table 4.426 15.516 0.00E+00 “Mode” 3 0.37E-03 1.057 0.804 1.705 -1.00E+00 0.9 -100.5 × El Centro ground motion Hinge Plastic Rotation (rad) “Modal” Response Combined (MPA) “Mode” 1 7.8 “Mode” 3 -1.00E+00 0.5 2.667 1.220 1.8 -6.6 -7.728 1.304 -1.5 -6.5 2.72E-03 7.414 1.066 -0.581 0.105 0.135 -7.879 1.2 -100.667 -1.37E-03 1.76E-03 4.1 46.00E+00 0.72E-03 7.942 6.202 8.738 1.304 1.6 -9.575 -53.

” however. two. Figures 4. 4.18) considering one.6 and 4. and three “modes. The pushover curves for the second and third modes included in Fig.17 and Tables 4.20) with n = 1 to a target roof displacement. the third “mode” contributions do not seem especially important (Fig.3 Modal Pushover Analysis with Gravity Loads To evaluate the accuracy of the dynamic response of the system.5 × El Centro ground motion are presented below.18 and 48 .5). Significant improvement is achieved by including response contributions due to the second “mode.7).8).7.0 cm.4) to roof displacement ur1o = 52. (3. Starting with its initial state under gravity loads. 4. 3. 4. (3. MPA including two “modes” underestimates the displacements of lower floors by up to 6% and overestimates the upper floor displacements by up to 22%. and plastic hinge rotations at selected external beam end at each floor level (Table 4. and the structure is pushed using the force distribution of Eq. The errors are especially large in the plastic hinge rotations estimated by the MPA procedure even if three “modes” are included (Fig. 4. Results of such analyses for the selected building subject to 1.11a). the structure is pushed using the force distribution of Eq.6 and 4. as are the roof displacements ur 2o = 11.6). As shown in Fig. drifts in all stories (Table 4.18 and Tables 4. The first “mode” alone provides adequate estimates of floor displacements. 4.” but not the contributions of higher “modes. They are now included in the pushover analysis of the structure for the first “mode” only.7).16a—which is slightly different than the one excluding gravity loads (Fig.16 are unchanged. resulting in the pushover curve shown in Fig.7 cm and ur 3o = 2. (3.17 and Tables 4.6 through 4. The response contributions of the second and third modes are the same as before (Tables 4.53 cm.17 and 4.8 present estimates of the combined response according to Eq. The story drifts are underestimated by up to 7% in the lower stories.3 . the results presented so far did not include gravity load effects.18 and Tables 4.7). Also presented at these roof displacements are the displacements at all floors (Table 4.” and the errors in these estimates relative to the exact response from nonlinear RHA.20) with n = 1 (Fig.6 and 4.4. and overestimated by up to 29% in the middle stories and by up to 13% in the upper stories.4. Obviously gravity load effects will influence the seismic demands due to the first “mode.” these effects will modify the combined modal response as well as the results of nonlinear RHA. but it is inadequate in estimating the story drifts (Fig. 4.4. 4. Static analysis of the structure for gravity loads provides the initial state—forces and deformations—of the structure.

Errors in MPA as a function of ground motion intensity: (a) floor displacements.5 × El Centro ground motion (a) Floor Displacements 60 50 40 Error Envelope Error for Floor No.Table 4. and (b) story drifts 49 .5 1 1. Noted Error (%) 30 20 10 0 0 2 8 3 4 1 9 Error (%) 30 20 5 4 6 5 7 2 9 10 0 0 1 8 7 3 6 0.5 1 1. (a) MPA.5 2 GM Multiplier 2.5 3 0.5 2 GM Multiplier 2. 4. Locations of plastic hinges determined by MPA considering one.14.15. Noted 60 50 40 (b) Story Drifts Error Envelope Error for Story No.5 3 Fig. and three “modes” and by NL-RHA for 1. 2−"Modes" •• •• •• •• •• •• •• •• •• •• •• •• •• •• • • • • •• • •• (c) 3−"Modes" • • • • • • • •• •• • •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• • • • • •• • •• • • • • • • • • • (d) Nonlinear RHA •• •• • •• •• •• •• • •• •• •• •• •• •• •• • •• •• • •• •• •• •• • • • • • • •• •• •• Fig.8). 1−"Mode" • • • • • • • • •• •• •• •• •• •• •• •• •• •• •• •• •• •• • • •• • •• • • • • • •• •• • •• •• •• •• (b) MPA. 4. two. Most pushover analysis procedures do not seem to compute to acceptable accuracy plastic hinge rotations.

15. MPA provides response values accurate enough for practical application in building retrofit or design. Figure 4. Shown is the error in each floor displacement (Fig. and nonlinear RHA. this procedure is able to predict plastic hinge locations essentially consistent with nonlinear RHA.19 were determined by four analyses: MPA considering one “mode. and the error envelope for each case. 4. These errors are only slightly larger than those in Fig. 50 .The locations of plastic hinges shown in Fig. each story drift (Fig.20b).20a). One “mode” pushover analysis is unable to identify the plastic hinges in the upper stories where higher mode contributions to response are known to be more significant. 4. and the errors are only weakly dependent on ground motion intensity. 4.” two “modes”. excluding gravity load effects. With two modes included in MPA. indicated by a ground motion multiplier. 4. The second “mode” is necessary to identify hinges in the upper stories.20 summarizes the error in MPA considering three “modes” as a function of ground motion intensity.

V = 7433 kN.5 2 3 4000 2000 0 0 Actual Idealized 0. 0.85 0.9 cm. “Modal” pushover curves with gravity loads included. V = 4952 kN.5 0. α = 0.25 5 10 15 20 Roof Displacement (cm) 25 (c) "Mode" 3 Pushover Curve 12000 10000 u = 4.5 0.25 1 2 4 6 Roof Displacement (cm) 8 10 Fig.0. noted are peak values of roof displacement for 0.(a) "Mode" 1 Pushover Curve 12000 10000 u = 36.75 Actual Idealized 20 40 60 Roof Displacement (cm) 80 (b) "Mode" 2 Pushover Curve 12000 10000 u = 9.5 0.5 2 3 Actual Idealized 2000 0 0 0.14 y by Base Shear (kN) 8000 6000 4000 1.85.16.75 0. 2. 4.85 Base Shear (kN) 8000 6000 4000 2000 0 0 0.0 × El Centro ground motion 51 .25.85 0.75 0.13 y by Base Shear (kN) 8000 6000 1 1. α = 0.19 y by 3 2 1. 0.50.5 1 0. and 3. 0.3 cm. α = 0.6 cm.75. 1.25 0. V = 5210 kN.0.

(a) Floor Displacements 9th 8th 7th 6th 9th 8th 7th 6th NL−RHA MPA 1 "Mode" 2 "Modes" 3 "Modes" (b) Story Drifts Floor Floor 5th 4th 3rd 2nd 1st 5th 4th 3rd 2nd 1st NL−RHA MPA 1 "Mode" 2 "Modes" 3 "Modes" Ground 0 0.17. 4.5 × El Centro ground motion.5 Fig. and 52 .5 Story Drift Ratio (%) 2 2.5 Displacement/Height (%) 2 Ground 0 0. Heightwise variation of floor displacements and story drift ratios from MPA and NL-RHA for 1. two.5 1 1. and hinge plastic rotations estimated by MPA including one. shading indicates errors in MPA including three “modes” (a) Floor Displacements 9th 8th 7th 6th 9th 8th 7th 6th (b) Story Drift Ratios Floor 4th 3rd 2nd 1st MPA 1 "Mode" 2 "Modes" 3 "Modes" Floor 5th 5th 4th 3rd 2nd 1st MPA 1 "Mode" 2 "Modes" 3 "Modes" Ground −60 −40 −20 0 Error (%) 9th 8th 7th 6th 20 40 60 Ground −60 −40 −20 0 Error (%) 20 40 60 (c) Hinge Plastic Rotations Floor 5th 4th 3rd 2nd 1st MPA 1 "Mode" 2 "Modes" 3 "Modes" Ground −120 −80 −40 0 Error (%) 40 80 120 Fig.5 1 1. Errors in floor displacements. 4. gravity loads included.18. story drifts.

20. and three “modes” and by NL-RHA for 1. Locations of plastic hinges determined by MPA considering one.5 × El Centro ground motion. two.5 3 Fig. Noted 60 50 40 (a) Story Drifts Error Envelope Error for Story No.5 1 1. 4.5 1 1. 1−"Mode" • • • • • • • •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• • • •• •• •• • • • • • •• •• •• •• •• •• •• (b) MPA. Noted Error (%) 30 20 6 8 1 5 2 4 3 9 7 Error (%) 30 4 5 20 6 9 7 1 10 0 0 10 0 0 8 2 3 0. and (b) story drifts. 4. Errors in MPA as a function of ground motion intensity: (a) floor displacements. gravity loads included (a) Floor Displacements 60 50 40 Error Envelope Error for Floor No. 2−"Modes" •• •• •• •• •• •• •• •• •• •• •• •• •• •• • • • • •• •• •• (c) MPA. 3−"Modes" (d) Nonlinear RHA • • • • • • • •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• • • • • •• •• •• • • • • • • • • •• •• • •• •• •• •• • • •• •• • •• •• •• •• • • •• •• • •• •• •• •• • • • • • • •• •• •• Fig.(a) MPA.5 3 0.5 2 GM Multiplier 2. gravity loads included 53 .19.5 2 GM Multiplier 2.

927 1.687 0.00E+00 0.6 Peak values of floor displacements (as % of building height = 37.351 -0.7 19.270 -12.8 -32.00E+00 0.55E-03 3.414 28.018 -0.5 1.213 1.7 1.033 -0.00E+00 0.128 -1.00E+00 0.037 -0.00E+00 0.35E-03 8.105 0.78E-03 1.102 1.5 × El Centro ground motion.353 -23.0 -30.888 1.436 1.19E-04 5.071 0.23E-03 0.197 -0.207 18.996 -0.00E+00 0.4 -4.19E-03 1.037 0.831 0.00E+00 3.8 Peak values of hinge plastic rotations (radians) from MPA for 1.3 9.049 -0.637 0.372 0.35E-03 8.0 2 “Modes” -32.938 1.11E-03 9.3 -100.00E+00 0.00E+00 2 “Mode” 8.1 4.101 -0.7 -2.8 28.168 -0.2 -2.490 -11.2 1.071 -0.00E+00 0.4 -6.263 1.603 -1.3 -22.19E-03 1.7 -2.237 0.237 0.35E-03 8.836 -0.057 0.527 -0.8 Table 4.860 1.263 0.998 21.55E-03 3.125 -1.04E-10 3.330 1.11E-03 9.783 1.5 54 .996 0.156 -0.00E+00 0.8 0.068 0.114 1.7 16.434 Error (%) NL 3 1 2 3 RHA “Mode” “Modes” “Modes” “Modes” 0.1 13.830 -12.2 9.00E+00 0.6 2.8 1.130 0.850 1.9 31.466 0.0 37.2 9.00E+00 0.26E-03 3.5 × El Centro ground motion.23E-02 1.17E-03 9.998 1.23E-03 3.00E+00 0.319 1.214 0.88E-03 1.109 0.7 Peak values of story drift ratios (as % of story height) from MPA for 1.3 -100.686 -7.850 -1.515 -50.00E+00 0.00E-03 5.199 16.0 -100.009 0.429 1.154 0.75E-03 0.2 4.1 1.00E+00 Error (%) “Modal” Response 1 “Mode” -32.102 1.854 0.19E-03 1.983 1.0 37.04E-02 8.637 0.Table 4.9 -3.305 -0.14 m) from MPA for 1.00E+00 0.015 0.913 7.00E-10 NL RHA 1.00E+00 3 “Modes” 8.2 9.055 0.5 21.2 12.5 -5.673 Table 4.3 1.831 0.00E+00 “Mode” 3 0.6 19.2 16.311 0. gravity loads included Story 1st 2nd 3rd 4th 5th 6th 7th 8th 9th “Modal” Response “Mode” 1 “Mode” 2 Drift Ratio (%) Combined (MPA) 1 “Mode” 2 “Modes” 3 “Modes” Error (%) NL 1 2 3 RHA “Mode” “Modes” “Modes” 1.00E+00 “Mode” 2 0. gravity loads included Floor 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Displacement /Height (%) “Modal” Response Combined (MPA) “Mode” 1 “Mode” 2 “Mode” 3 1 “Mode” 2 “Modes” 0.667 0.00E-03 5.516 0.4 -4.5 “Mode” 3 -1.399 -0.594 -1.88E-03 0.310 1.176 0.2 21.454 1.3 -22.3 13.00E-03 5.983 1.190 -0.461 0.399 0.2 1.5 1.6 0.8 -32.603 1.107 1.854 0.064 -10.908 -1.315 0.35E-10 3.266 -0.5 0.4 20.5 21.822 1.478 0.4 1.00E+00 0.2 2.507 1.728 1.114 -1.2 4.7 4.00E+00 0.066 -0.8 9.11E-03 9.514 -1.2 -0.23E-03 0.5 × El Centro ground motion.8 0.813 1.754 0.434 0.0 -100.3 -22.213 1.953 15.2 4.4 26.998 0.23E-03 3.754 1.6 2.88E-03 0.00E-03 5.2 -3.665 0.530 1.55E-03 3.9 -6.098 20.429 -1.2 0.908 1.434 0.933 1.5 3 “Modes” -32. gravity loads included Hinge Plastic Rotation (rad) Floor “Modal” Response “Mode” 1 1st 2nd 3rd 4th 5th 6th 7th 8th 9th 8.4 1.9 -4.0 16.0 -30.530 1.13E-03 5.11E-03 9.00E+00 0.821 1.00E+00 Combined (MPA) 1 “Mode” 8.821 -1.319 1.3 -100.066 -0.35E-03 8.371 -0.744 1.5 10.0 37.6 1.330 1.00E+00 0.257 0.213 1.1 21.877 -46.530 1.921 1.00E+00 0.858 2.19E-03 1.

Specified in FEMA-273 are three distributions for lateral forces: 1. 5. Comparison of Modal and FEMA Pushover Analyses 5.5 sec . and the exponent k = 1 for fundament period T1 £ 0. plastic hinge rotations. j 2. and 3. etc. a plot of base shear versus roof displacement.5 sec . and varies linearly in between. SRSS distribution: s* is defined by the lateral forces back-calculated from the story shears determined by response spectrum analysis of the structure. “Uniform” distribution: s* = m j (where the floor number j = 1.. 2… N ).2 COMPARATIVE EVALUATION Compared in this section are the earthquake-induced demands for the selected building determined by five analyses: pushover analysis using the three force distributions in FEMA-273. gravity load effects were included in all 55 . MPA considering three “modes.1 FEMA-273 PUSHOVER ANALYSIS In this investigation we focus on one step in the nonlinear static procedure in the FEMA-273 document [Building Seismic Safety Council.5. assumed to be linearly elastic. story drifts. is determined by nonlinear static analysis of the structure subjected to lateral forces with invariant distribution over height but gradually increasing values until a target value of roof displacement is reached. computed at the target displacement represent the earthquakeinduced demands on the structure. The floor displacements. joint rotations. Equivalent lateral force (ELF) distribution: s* = m j h k where h j is the height of the j j jth floor above the base. 1997] The pushover curve. k = 2 for T1 ≥ 2.” and nonlinear RHA.

0981 0. The “uniform” distribution overestimates all floor displacements by 17-28%.11 0. The pushover curves are given in Fig.1.1.3a and Table 5.0654 0.analyses.21 0.1. (b) ELF.3.0896 0. 5.0702 0. the floor displacement demands in Fig. The three FEMA force distributions are presented in Fig.367 (a) Uniform (b) ELF (c) SRSS Fig.3a and 5. and the displacements of upper floors are overestimated by 9-22%. wherein the first two are obvious and the third is determined from response spectrum analysis of the building (Appendix C).119 0.1 through 5.11 0.00719 0. 5.11 0.042 0.112 0.0466 0.0 cm. Force distributions in FEMA-273: (a) “uniform”.062 0.5 times the El Centro ground motion.0913 0. Using each of these force distributions. 5. the value determined from RHA of the first-mode inelastic SDF system for 1. pushover analyses are implemented for a target roof displacement of 52.4.0381 0.126 0.3. 5. the story drift demands in Fig. The MPA procedure is more accurate than all the FEMA force distributions.3.1 demonstrate that the displacement demands are underestimated by the ELF and SRSS force distributions by 12 to 30 % at the lower six floors of the building.11 0.0197 0. with the errors being larger for the SRSS distribution. 5. and the locations of all plastic hinges in Fig. The errors in the FEMA and MPA estimates of seismic demands relative to the “exact” demands are presented in Fig.177 0. and Table 5. 5.11 0. and (c) SRSS 56 .4. 0.2. plastic hinge rotation demands in Table 5. Figures 5.0446 0.165 0.4 and Tables 5.3b and Table 5.281 0. The first four floor displacements are within 5% of the “exact” value. Also included in these presentations are the MPA results considering three “modes” and the “exact” demands from nonlinear RHA. both presented in Section 4.4a.11 0.2. 5.11 0.

α = 0.2.24 y by Base Shear (kN) 8000 6000 4000 2000 0 0 Actual Idealized 10 20 30 40 Roof Displacement (cm) 50 60 Fig. α = 0. gravity loads are included 57 . Pushover curves using three force distributions in FEMA-273: (a) “uniform”. (b) ELF.8 cm. and (c) SRSS.18 y by Base Shear (kN) 8000 6000 4000 2000 0 0 Actual Idealized 10 20 30 40 Roof Displacement (cm) (c) SRSS Distribution 50 60 12000 10000 u = 39. V = 8530 kN. V = 7456 kN. V = 6897 kN.14 y by Base Shear (kN) 8000 6000 4000 2000 0 0 Actual Idealized 10 20 30 40 Roof Displacement (cm) (b) ELF Distribution 50 60 12000 10000 u = 38. α = 0.(a) Uniform Distribution 12000 10000 u = 33 cm.3 cm. 5.

Figure 5. (The 100% error in a 6th floor hinge is ignored as it simply represents that the MPA estimated zero rotation. Modal pushover analysis procedure gives estimates better than all the FEMA force distributions. and Table 5. at most. MPA including three “modes. errors are largest in the upper stories.4b. (a) Floor Displacements 9th 8th 7th 6th 9th 8th 7th 6th (b) Story Drift Ratios Floor 4th 3rd 2nd 1st NL−RHA MPA.5 Story Drift Ratio (%) 2 2. with story drifts under estimated by. In contrast. 7%. the errors are largest in the lower stories. This seems to be an inherent limitation of pushover analysis. but it is still inaccurate. reaching 35%. 5.” and NLRHA.3 demonstrate that the hinge plastic rotations estimated by all three FEMA force distributions contain unacceptably large errors. gravity loads included 58 . 3 "Modes" FEMA Uniform ELF SRSS 0. such as hinge plastic rotations.2 demonstrate that the story drift demands are greatly underestimated by all the FEMA force distributions. For the SRSS distribution.4c and Table 5. 3 "Modes" FEMA Uniform ELF SRSS 0.Figures 5.3b and 5. larger errors are noted in the upper and lower stories. the MPA procedure is more accurate than all the FEMA force distributions. For the uniform distribution.3.5 1 1. For the ELF distribution.5 1 1. and overestimated by no more than 32%.5 Ground 0 Ground 0 Fig. reaching 31%.) The pushover analysis procedures considered seem incapable of computing accurately local response quantities. with errors reaching 37% in this example.5 Displacement/Height (%) 2 Floor 5th 5th 4th 3rd 2nd 1st NL−RHA MPA. Heightwise variation of floor displacements and story drift ratios estimated using FEMA-273 force distributions. reaching 64%. whereas nonlinear RHA computed an insignificantly small value.

The locations of the plastic hinges are not identified correctly by the FEMA force distributions.5 were determined by five analyses: MPA considering the three “modes. the “uniform” distribution fails to identify yielding of the beams above the fourth floor. and hinge plastic rotations estimated using FEMA-273 force distributions and MPA (including three “modes”). and the three FEMA analyses. and the ELF distribution fails 59 . story drifts. The locations of plastic hinges shown in Fig.4. While pushover estimates for floor displacements are more accurate.(a) Floor Displacements 9th 8th 7th 6th MPA FEMA Uniform ELF SRSS 9th 8th 7th 6th (b) Story Drift Ratios MPA FEMA Uniform ELF SRSS Floor 4th 3rd 2nd 1st Floor −40 −20 0 Error(%) 9th 8th 7th 6th 20 40 60 5th 5th 4th 3rd 2nd 1st Ground −60 Ground −80 −60 −40 −20 0 20 Error (%) 40 60 80 (c) Hinge Plastic Rotations MPA FEMA Uniform ELF SRSS Floor 5th 4th 3rd 2nd 1st Ground −200 −150 −100 −50 0 50 Error (%) 100 150 200 Fig. it appears that structural performance evaluation should be based on story drifts that are known to be closely related to damage and can be estimated to a higher degree of accuracy by pushover analyses.” nonlinear RHA (“exact”). Errors in floor displacements. 5. gravity loads included The structural engineering profession is now comparing these hinge plastic rotations against rotation limits established in FEMA-273 to judge structural component performance. the SRSS distribution fails to identify yielding of beams in the middle floors. Based on the results presented here. 5. they are not good indicators of damage.

but fails to predict yielding in a few locations. The MPA procedure is superior to FEMA-273 analyses over the entire range of ground motion intensities considered.to identify yielding in some locations.6 and 5. especially in estimating story drifts. and the error envelope for each case.” The MPA procedure provides estimates of earthquake demands that are significantly more accurate than all FEMA273 analyses. Figures 5. The MPA procedure identifies yielding in most locations predicted by “exact” analysis. 60 .7 summarize the error in FEMA analyses relative to the “exact” demands as a function of ground motion intensity indicated by a ground motion multiplier. Included for comparison is the error in MPA with three “modes. Shown is the error in each floor displacement and each story drift.

35E-10 3.330 1.0 -100.45E-03 3.9 0.3 23.62E-03 0.6 4.6 -25.0 -30.03E-03 5.9 16.00E+00 4.0 Table 5.524 0.7 21.8 -35.2 Peak values of story drift ratios (as % of story height) from FEMA force distributions and MPA.6 -17.9 -20.007 1.355 0.199 27.6 -73.263 0.8 7.3 -100.3 31.221 1.975 1.294 1.984 1.7 -15.5 -29.0 -63.8 2.4 -12.399 1.6 21.4 0.4 -23.351 0.414 1.5 61 .11E-03 9.0 -100.7 26.00E-10 Floor 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Uniform 1.888 0.7 -28.8 -2.547 -27.9 -70.877 1.011 1. gravity loads included Displacement /Height (%) FEMA ELF SRSS MPA 0.098 1.58E-04 6.55E-03 3.23E-02 1.1 163.2 5.858 1.9 11.0 -100.992 1.00E+00 0.2 -4.2 9.4 16.8 -32.13E-03 5.2 16.51E-03 4.515 FEMA Uniform ELF 27.8 Floor 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Uniform 0.335 2.207 1.00E+00 0.913 0.234 1.7 MPA -2.178 1.736 0.466 0.667 0.168 1.6 -6.836 0.0 -100.7 7.417 1.344 0.064 1.367 1.341 1.490 0.3 -4.19E-04 5.4 16.04E-02 8.6 -41.262 1.2 19.061 1.2 6.52E-04 1.789 0.4 -50.3 29.23E-03 3.953 0.938 1.93E-03 1.00E+00 0.6 14. gravity loads included Displacement /Height (%) FEMA ELF SRSS MPA 1.65E-03 7.0 37.5 -3.5 Story 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Uniform 2.5 -33.00E+00 0.372 1.94E-03 2.2 -12.1 0.8 44.8 -63.323 1.6 -11.00E+00 8.26E-03 3.19E-03 1.860 1.830 2.4 -55.8 -100.5 10.9 16.1 10.16E-03 0.5 -29.686 0.8 -100.310 1.3 -11.314 1.7 -60.730 1.5 26.7 Table 5.7 19.839 0.09E-03 4.9 SRSS -22.2 16.462 1.109 1.6 22.5 -27.673 Error (%) NL RHA 1.998 1.083 1.10E-02 7.2 4.5 22.84 1.487 0.998 1.3 -22.611 0.5 -33.2 13.3 -14.089 1.067 0.306 1.0 -57.50E-03 0.00E-03 5.672 1.436 Error (%) NL Uniform RHA 0.809 0.015 0.8 17.78E-03 0.14 m) from FEMA force distributions and MPA.00E+00 NL RHA 1.5 -32.Table 5.34E-03 2.318 2.623 1.00E+00 0.927 1.7 FEMA ELF SRSS -22.353 1.9 28.2 10.399 1.3 Peak values of hinge plastic rotations (radians) from FEMA force distributions and MPA Hinge Plastic Rotation (rad) MPA FEMA ELF SRSS 4.6 17.59E-03 5.560 1.597 0.00E+00 0.00E+00 2.854 0.2 -100.562 1.209 0. Peak values of floor displacements (as % of building height = 37.7 -29.1.154 1.724 0.4 21.875 0.2 9.195 0.530 1.8 16.35E-03 8.53E-02 1.75E-03 0.783 1.399 -27.1 -26.78E-03 1.530 1.17E-03 9.0 -71.270 0.4 MPA -2.0 -59.214 1.9 15.566 1.00E+00 Error (%) FEMA Uniform ELF SRSS MPA 24.88E-03 1.128 1.9 -77.708 0.5 15.1 -3.6 -4.

5.5. Locations of plastic hinges determined from three force distributions in FEMA-273. gravity loads included 62 .5 × El Centro ground motion. MPA including three “modes” and NL-RHA for 1.(a) FEMA: Uniform • • • • • • • • •• •• •• •• •• •• (b) FEMA: ELF • •• •• •• •• •• •• • •• • •• •• •• •• •• • • • • • •• • • • • • • • •• •• •• •• • •• •• •• •• • •• •• •• •• • • •• •• •• • • • • • •• •• (c) FEMA: SRSS • • •• •• •• •• • • •• •• •• • • •• •• •• • • • • •• •• •• • • • • • • • • • (d) MPA with 3 "Modes" • • • • • • • •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• • • • • •• •• •• (e) Nonlinear RHA • • • • • • • • •• •• • •• •• •• •• • • •• •• • •• •• •• •• • • •• •• • •• •• •• •• • • • • • • •• •• •• Fig.

5 2 GM Multiplier 2. Errors in floor displacements from three force distributions in FEMA-273 and from MPA including three “modes”.5 1 1.5 1 1.5 3 Fig.5 3 (c) FEMA: SRSS 60 50 40 60 50 40 2 3 1 5 4 (d) MPA (3 "Modes") Error (%) 30 20 10 0 0 8 Error (%) 30 20 6 8 1 5 2 4 3 9 7 6 7 9 10 0 0 0.5 2 GM Multiplier 2.(a) FEMA: Uniform 60 50 40 60 50 40 3 (b) FEMA: ELF Error Envelope Error for Floor No. 5.5 3 0 0 0. Noted 2 Error (%) Error (%) 1 2 30 20 10 0 0 1 4 6 7 30 20 10 8 3 5 8 4 9 7 9 6 5 0.5 2 GM Multiplier 2.5 3 0. gravity loads included 63 .5 2 GM Multiplier 2.5 1 1.5 1 1.6.

5 3 0 0 0.5 1 1.5 1 1.5 1 1.5 3 (c) FEMA: SRSS 60 50 40 60 50 40 3 2 1 (d) MPA (3 "Modes") Error (%) 30 20 Error (%) 30 5 4 8 9 20 9 6 7 7 10 0 0 4 10 2 3 1 8 5 6 0.5 3 Fig.5 2 GM Multiplier 2.(a) FEMA: Uniform 100 60 50 9 8 (b) FEMA: ELF Error Envelope Error for Story No. Error in story drifts from three force distributions in FEMA-273 and from MPA including three “modes”. 5. Noted 80 40 9 1 Error (%) 7 Error (%) 60 30 20 10 2 5 6 4 8 40 1 6 20 2 5 3 4 3 7 0 0 0.5 3 0 0 0.7. gravity loads included 64 .5 2 GM Multiplier 2.5 2 GM Multiplier 2.5 1 1.5 2 GM Multiplier 2.

. For this system. e. now common in structural engineering practice. It has led to the following conclusions: 1. SRSS) leads to the modal pushover analysis (MPA) procedure. This system has vibration properties—natural frequency w n and damping ratio. joint rotations.g. ζ n —of the nth-mode of the MDF system.6 Conclusions This investigation was aimed toward developing an improved pushover analysis procedure based on structural dynamics theory. hinge plastic rotation. which retains the conceptual simplicity and computational attractiveness of the procedure with invariant force distribution. 3. where the nature and magnitude of errors arising from approximate modal combination rules are well understood. This MPA procedure for elastic buildings is shown to be equivalent to the standard response spectrum analysis (RSA) procedure. 2.. pushover analysis is inherently limited in the sense that it cannot provide any cumulative measure of response.g. Pushover analysis of a one-story inelastic system predicts perfectly peak seismic demands: deformation. where m is the mass matrix n of the building and φ n its nth mode. However. and the structure is pushed to the roof displacement determined from the peak deformation Dn of the nth-mode elastic SDF system. Dn is available from the elastic response (or design) spectrum. etc. 65 . The peak response of an elastic multistory building due to its nth vibration mode can be exactly determined by static analysis of the structure subjected to lateral forces distributed over the building height according to s* = mφ n . Combining these peak modal responses by an appropriate modal combination rule (e. the energy dissipated in yielding or the cumulative rotation of a plastic hinge.

5. (ii) Idealize the pushover curve as a bilinear curve and convert it to the bilinear force-deformation relation for the nth-“mode” inelastic SDF system. 6. p eff (t ) = . These underlying assumptions and approximations are evaluated and the errors in the UMRHA procedure relative to the “exact” nonlinear response history analysis are documented. n where φ n is now the nth natural mode for small-amplitude linear vibration. and (2) superposing responses of the inelastic MDF system to individual terms.urn bg bg bg bg b g curve from a pushover analysis of the structure for the force distribution s* = mφ n . 2.n t = . and (ii) the total response ro is determined by combining the rno (n = 1.mι ug t .4. p eff . The response value rno is determined by implementing the following steps: (i) Develop the base-shearroof-displacement Vbn . The MPA procedure developed to estimate the seismic demands for inelastic buildings consists of two phases: (i) the peak response rno of the inelastic MDF system to effective earthquake forces p eff . the SRSS rule). …) according to an appropriate combination rule (e.g.n t is determined by pushover analysis. To enable systematic extension of the MPA procedure to inelastic systems.sn ug t —n denotes mode number—in the modal expansion of the effective earthquake forces. (iii) Compute the peak deformation Dn of this system with unit mass by nonlinear response history analysis or from the inelastic response (or design) spectrum. an uncoupled modal response history analysis (UMRHA) is developed by (1) neglecting the coupling among modal coordinates arising from yielding of the system. with vibration properties in the linear range same as those of the nth-mode elastic SDF system. 66 ..

The MPA procedure is more accurate than all the FEMA force distributions in estimating floor displacements. 8. and identifying locations of most plastic hinges. Pushover analyses seem to be inherently limited in computing accurately hinge plastic rotations. plastic hinge rotations. it fails to compute with acceptable accuracy plastic rotations of the hinges.25 to 3. the errors in the MPA procedure are shown to be only weakly dependent on ground motion intensity. Thus the MPA procedure is accurate enough for practical application in building evaluation and design.0. the pushover analysis provides the peak value rno of any response quantity: floor displacements. and hinge plastic rotations. This implies that MPA is able to estimate the response of buildings responding well into the inelastic range to a similar degree of accuracy when compared to standard RSA for estimating the peak response of elastic systems. Comparing the earthquake-induced demands for the selected 9-story building determined by pushover analysis using three force distributions in FEMA-273. and nonlinear RHA. MPA. Based on results presented for El Centro ground motion scaled by factors varying from 0.(iv) At the roof displacement determined from Dn . 67 . demonstrates that the FEMA force distributions greatly underestimate the story drift demands. and lead to unacceptably large errors in the hinge plastic rotations. joint rotations. story drifts. The initial state—forces and deformations—of the structure can be considered in the MPA procedure by including these gravity load effects in pushover analysis of the structure only for the first “mode.” 10. such as hinge plastic rotations. all pushover analysis procedures do not seem to compute with acceptable accuracy local response quantities. 7. 9. etc. story drifts. However. Comparing the peak inelastic response of a 9-story SAC building determined by the approximate MPA procedure—including only the first two or three rno terms—with rigorous nonlinear RHA demonstrates that the approximate procedure while providing good estimates of floor displacements and story drifts.

This report has focused on development of the MPA procedure and its initial evaluation in estimating the seismic demands on a building imposed by a selected ground motion.11. 68 . they are not good indicators of damage. Instead. structural performance evaluation should be based on story drifts that are known to be closely related to damage and can be estimated to a higher degree of accuracy by pushover analyses. with the excitation scaled to cover a wide range of ground motion intensities and building response. such as life safety and collapse prevention. This new method for estimating seismic demands at low performance levels. While pushover estimates for floor displacements are even more accurate. should obviously be evaluated for a wide range of buildings and ground motion ensembles. The present trend in the structural engineering profession of comparing computed hinge plastic rotations against rotation limits established in FEMA-273 to judge structural performance does not seem prudent.

Estimation of seismic drift demands for frame structures. Nat. Report No. Krawinkler. Engng. Washington.C. Spectra. (1999). Stanford University. A.K. Earthq. S. and Wen. D. 123:256-265.K. DRAIN-2DX user guide. Engrg. H. Conf. 15:417-434. 9th World Conf. Tokyo-Kyoto. A. Struct. Y. (2000). A. ASCE 123(1):3-10. Report No.. Washington. Seismic performance and retrofit evaluation for reinforced concrete structures. 20(4-6):452-464. and Kunnath.. ASCE. on Earthq. Calif. H.W. J.D.7 References Allahabadi. Krawinkler.4. B. P. Earthquake Engineering Research Center. Bracci. Engrg. Calif.M. Story drift demands for steel moment frame structures in different seismic regions. J. Engrg. FEMA-273. and Reinhorn. Engrg. S.. Federal Emergency Management Agency. 6th U. Method of reliability-based seismic design. and Fischinger. (2001). Dyn. Proc. and Krawinkler... Spectra. 16(2):367-392 Han. (1988). S. John A. J. J. Dynamics of Structures: Theory and Applications to Earthquake Engineering.P. and D’Amore. 69 . University of California. (1998).. Proc.. Pros and cons of a pushover analysis of seismic performance evaluation. and Gupta. E. and Powell. Berkeley. A.. UCB/EERC-88/06. and Krawinkler. Kunnath. Gupta. Earthq. Struct. (1999). G. Seismic demands for performance evaluation of steel moment resisting frame structures (SAC Task 5. 132. Building Seismic Safety Council (1997). S. Engrg. and Seneviratna. 5:111-116. Gupta. 29:1287-1305.M. Struct.H. (1997). (1988).S. Struc. (1997). (2000). H.K.. G.K. Adaptive spectra-based pushover procedure for seismic evaluation of structures. N2—a method for nonlinear seismic analysis of regular structures. Prentice Hall: New Jersey. Gupta. Japan. Earthq. Fajfar. M. I: Equivalent nonlinear system.K. Earthq. A.. Blume Earthquake Engineering Center. Stanford.3). Pushover analysis procedure in earthquake engineering. Chopra. R. NEHRP Guidelines for the Seismic Rehabilitation of Buildings. Seattle. (1998). Kim. H.

Skokan. and Sozen. Earthq. (1999). 11th World Conf.K.. Washington. Multimode pushover procedure (MMP)—A method to identify the effects of higher modes in a pushover analysis. D. (2000). Vance. H.A.S. Sasaki. Struct.. Ph.. Saiidi. Proc. and Krawinkler. (1994). and Gupta.S. Validity of deformation demand estimates using nonlinear static procedures.edu/~quake/. Univ. 15(4):765-789. Villaverde. of Civil Engrg.A.S. F. H. Ohtori. B. Nat.. Structures. Notre Dame University.E. Japan Workshop on Performance-Based Engineering for Reinforced Concrete Building Structures. (1996). (1981). Mech. Proc. U.. Paper No. Paret. S.. (1999). S. R. 122:282-285. K. Approximate inelastic procedures to identify failure mechanisms from higher mode effects. and Bonowitz. M. Maison. D. Conf. Auckland.F. J.M. Reliability of nonlinear static methods for the seismic performance prediction of steel frame buildings. (1996). (1998). (1991). D. Simplified response spectrum seismic analysis of nonlinear structures.. Naiem.. 966. Earthq. Calif. Acapulco.. Engrg.. Y. SEAOC 1998 Convention. G. V. 5th U. Methodology for R/C Bldg. Simple nonlinear seismic analysis of R/C structures. of Calif. 70 . (2000). Engrg.F.J. U. B. 6th U. R.. Proc.K. Spectra. 79-94. M. 1972.. Spencer. K. Engrg. Sasaki. Eilbekc. J. on Earthq.C. Jr. Christenson. Shiohara.F.S. and Dyke.Kunnath. Proc. and Freeman. Div. Earthq. Berkeley. Maui. T.K. T. R. ASCE.. Common pitfalls in pushover analysis. Proc. Engrg. R. Freeman. and Paret.... T.. 107(ST5):937-951.-Japan Workshop on Performance-Based Earthq. S.. Mexico. S. Sapporo. Seismic evaluation and upgrading of existing buildings.A. Matsumori. Proc. Engrg.nd.. pp. E.. and Lobo. Conf. Earthquake member deformation demands in reinforced concrete frame structures. Seattle.S. Hawaii Miranda. Dissertation. T. New Zealand. Nonlinear static pushover analysis—why.H. B. Benchmark Control Problems for Seismically Excited Nonlinear Buildings.. Indiana. ASCE. (2000). http://www. 1:283-292. and Hart. Hokkaido. Paper No. How safe are pre-Northridge WSMFs? A case study of the SAC Los Angeles Nine-Story Building. when and how?.. Proc. Structural Engineers of California. Japan. Earthq. and Kabeyasawa. S. Engrg. Dept. (1998). Otani. M. Lawson. 12th World Conf.J...

1 incrementally and record the base shears and associated roof displacements. respectively. iterations may be necessary. 3. 3.20): s* = mφn n n 2. (3. Apply force distribution of Step 2.g. A. 3.Appendix A Uncoupled Modal Response History Analysis A.g.urn ) pushover curve for the force distribution s* : n 2. 1. DRAIN2DX (Allahabadi and Powell. B. using any numerical integration method.1) using the FEMA-273 procedure (Building Seismic Safety Council. trapezoidal rule. The structure should be pushed just beyond the target (or expected) roof displacement in the selected mode. and modes. Define the anchor point. 2.2. for linear-elastic vibration of the building. e. Since the target roof displacement may not be known at the start of the procedure.1 STEP-BY-STEP PROCEDURE A detailed step-by-step implementation of the uncoupled modal response history analysis (UMRHA) procedure is presented in this section and illustrated by an example in the next section. develop the base-shear – roof-displacement (Vbn .1. Idealize the pushover curve as a bilinear curve (Fig.. φn . Let the roof displacement and base shear at the anchor point be urno and Vbno . Calculate the area under the actual pushover curve.1.. e.2. Apn . 1988). This step can be conveniently implemented in any commercially available software. 71 . For the nth-“mode”. Compute natural frequencies. Define the force distribution s* from Eq. ω n . 1977). of the bilinear curve at the target roof displacement.

Vbny . Replace i+1 with i and repeat Steps 3. i +1 i i 3.5. Draw the curve OAB by connecting the three points O. O.0. and B with straight-line segments to obtain the idealized bilinear curve.2. by connecting a straight line between origin.1.9.6 × Vbny . obtained by judgment.6 . Calculate Vbny = Vbny × Apn Abn . i 3.9. iterations are necessary. Calculate the post-yielding strain-hardening ratio. If desired.1. Calculate the slope. will be refined by an iterative procedure that seeks to equate areas under the actual and the idealized pushover curves.4. Calculate the yield displacement. 3. and a point on the actual pushover curve with base shear equal to i 0. A. Calculate area under the bilinear curve OAB. i 3.6 . 72 . Let the point with base shear = Vbny and roof displacement = urny be denoted as A. 3. 3.4. If the error exceeds some pre-specified tolerance. This step gives the secant stiffness at a base shear equal to 60% of the yield base shear. i i 3.7. Vbny .4.6 × Vbny .2.4 to 3. i i i αn = Vbno Vbny − 1 urno urny − 1 i 3. urny = Vbny kn . determine the roof displacement. Abn . at base shear i equal to 0. urn.8. other appropriate methods can be ( used. Calculate initial slope of the idealized bilinear curve. Estimate the yield base shear.6. corresponding to the estimated yield i i i base shear. i 3.9. Calculate the error = 100 × Abn − Apn ( ) ( ) ( ) ) Apn .3. i i i 3.i 3.6 × Vbny ( ) i urn . From the pushover data. kn = 0. This value. kn .8.0.

2. 4. Calculate peak values. 7. An (t ) . (3. 4. A. Compute deformation history.6 ur n y ur n o ur n Fig. Scale the horizontal axis by Γnφrn to obtain Dno = urno Γ nφrn and Dny = urny Γ n φrn (Eqs.11b). 4. Scale the vertical axis with * Mn to obtain * Fsno Ln = Vbno M n and * Fsny Ln = Vbny M n (Eqs. 5.15) and (3. of the combined responses obtained in Step 8. * 4.10a and 4.eps ur n.1. A. (3.2.3b) with unit-mass and force-deformation relation of Fig.11a).10b and 4.6 × Vbny A • Idealized αnkn B • 1 Actual k 1 • O n figA_1. of the nth“mode” inelastic SDF system (Fig. Repeat Steps 3 to 6 for as many modes as required for sufficient accuracy. 9. 6.2). Dn (t ) .1 Idealization of nth-“mode” pushover curve 73 . 4. A.16). 4. and pseudo-acceleration history. Develop the Fsn Ln − Dn relation (Fig. r o . In general first two or three modes will suffice.3.4. Vbn Idealization of Pushover Curve Vbno Vbny 0.12) and (3. 8.13). (3. Compute the Ln and Γ n from Eq. Combine the “modal” responses using Eqs.0.4) and effective modal mass from M n = Ln Γn . Calculate histories of various responses using Eqs.

At this point. The following steps illustrate the procedure to develop the idealize curve for the first “mode”.2. Area under the actual pushover curve. Idealized bilinear curves for each of the three modes are included in Fig. A. 1.3. and 12. The base-shear – roof-displacement (Vbn .1. Target displacements used to generate these pushover curves are 63. Following is step-by-step implementation of the procedure described in Section A. n 2.5 cm (25 in. Ap1 = 360777 kN-cm.2 EXAMPLE The UMRHA procedure is implemented to calculate the response of the 9-story building described in Section 3. for the first.4. The anchor point. ur1o = 63.). 25.6 kN. B. A.4.1.4 cm (10 in.). 3. are shown in Fig.5.5 cm and Vb1o = 8729.urn ) pushover curve for the force distribution s* : n 2.eps Dny = ur n y / Γnφr n Dno = ur n o / Γnφr n Dn Fig. computed for the first three modes are shown in Fig. s* .2 Properties of the nth-“mode” inelastic SDF system A. 74 .). 3.3.1. respectively. A. and third mode. is defined at the target roof displacement.3. The pushover curves for the first three modes. to the north-south component of the El Centro (1940) ground motion scaled up by a factor of 1.7 cm (5 in. second. The force distributions. generated using DRAIN-2DX. 3. 2.Fsn / Ln Fsn / Ln − Dn Relationship Vbno / M* n Vbny / Mn * 1 α ω2 n n ω2 1 n figA_2. 3. First three mode shapes and frequencies of the selected building were computed and are shown in Fig.2. 3.

4 210.4. A. Error = 100 × (365100 − 360777 ) 360777 = 1. L1 = 2736789 kg.01%.4 kN. i i 3.6 kN. The procedure converged after nineteen cycles to give ur1 y = 36.3666. This value exceeds the prespecified tolerance of 0.0. i 3. Vb1 y = 7615.4.0.3666 = 3740189 kg. α1 = Vb1o Vb1 y − 1 ur1o ur1 y − 1 = ( ) ( ) (8729. Determined from the pushover database. 3.23 cm. 3.86 cm at 0.6 = 22. * 4.09 cm.4 kN.86 = 210.6 × Vb1 y ( ) i ur1. 3.5.2.1.4 × (360777 365100 ) = 7911. Also included are the modal damping ratios and the periods calculated from Eq.4 ) − 1 (63.9 kN.9. Area under the bilinear curve OAB.6 × Vb1 y = 4803. The Fs1 L1 − D1 relation for the first “mode” is developed as follows.4. The results of the iterative procedure are summarized in Table A. The yield displacement.8 kN. and B with straight-line segments gives the idealized bilinear curve. ur1. The initial slope of the idealized bilinear curve.9. ur1 y = Vb1 y k1 = 8006.6 8006.2.198%.2.6 = 4803.7. k1 = 0.09 ) − 1 = 0. The next estimate of the yield shear is Vbny = 8006.09 cm and Vb1 y = 8006. Therefore.18 = 38. k1 .3.194.18 kN/cm.6. The curve OAB obtained by connecting the three points O. i +1 3.i 3. The post-yielding strain-hardening ratio. i 3.8. i i i 3. Γ1 = 1. The first estimate of the yield base shear Vb1 y = 8006.1. i i 3. The point A on the i i bilinear curve is defined by ur1 y = 38.135. 4. Ab1 = 365100 kN-cm.5 38.1. and α1 = 0. is calculated as follows. (4.13). iterations are necessary.9. 75 . The results for other modes are summarized in Table A.8 22. and M1 = 2736789 × 1. i i i 3.1.

2.7. 6.4.4. 7. Scaling the horizontal axis by Γ1φr1 gives D1o = 46. 8. The peak values are computed and are summarized in Tables 4. Scaling the vertical axis by Fs1 y L1 = 203. 5.51 cm.46 cm and D1y = 26. Deformation and pseudo-acceleration histories of the inelastic SDF systems for the first “mode” with unit mass and force-deformation relation developed in Step 4 are plotted in Fig. 4. The combined modal responses are presented in Fig.8. 4.7. 4.62 (cm/sec2). 9.40 (cm/sec2) and 76 . A. Histories of roof displacement and top story drifts for the first “mode” are computed and presented in Fig. The results were generated for first three “modes’ and are included in Fig.1 and 4.7. 4. 4. * M1 gives Fs1o L1 = 233.3. The peak values are also plotted in Fig.2.

9 cm. V = 7616 kN. α = 0. V = 5210 kN. A. α = 0.eps 10 20 30 40 50 Roof Displacement (cm) (b) "Mode" 2 Pushover Curve 60 8000 u = 9.eps 0 0 2 4 6 8 10 Roof Displacement (cm) 12 Fig.6 cm.14 y by 6000 Base Shear (kN) 4000 2000 Actual Idealized figA_3c.19 y by Base Shear (kN) 8000 6000 4000 2000 0 0 Actual Idealized figA_3a.2 cm.(a) "Mode" 1 Pushover Curve 12000 10000 u = 36.eps 0 0 5 10 15 20 Roof Displacement (cm) (c) "Mode" 3 Pushover Curve 25 8000 u = 4. “Modal” pushover curves for the example building 77 .13 6000 Base Shear (kN) 4000 2000 Actual Idealized figA_3b. V y by = 4952 kN.3. α = 0.

223 −80 80 n=3 −3 3 n=3 D3 (cm) 0 A3 (g) • 10.5 × El Centro ground motion for the first “mode.” and third “mode” inelastic SDF systems.4.06 0 • 1. A. Histories of deformation and pseudo-acceleration due to 1.2769 D1 (cm) 0 • 35.33 A1 (g) n=2 0 −80 80 −3 3 n=2 D2 (cm) 0 A1 (g) • 22.52 5 10 15 20 Time (sec) 25 30 0 • 1.” second “mode.m −80 0 −3 0 25 30 Fig.(a) Deformation 80 n=1 3 (b) Pseuso−acceleration n=1 • 0.747 5 10 15 20 Time (sec) figa_4. 78 .

037 0.191 0.063 0.78 21.198 0.190 0.193 0.2 4671.8 7622.09 37.136 Fsny Ln (cm/sec2) Dny (cm) Fsno Ln (cm/sec ) Dno (cm) Tn (sec) 233.0 4577.188 0.948 ζ n (%) 79 .9 7615.103 “Mode” 3 696400 0.237 0.6 i k1 i ur1 y i α1 i Ab1 Error (%) 1.2671 1.81 21.162 0.1 4574.95 21.3 7658.64 37. Results of iterative procedure to develop the idealized bilinear curve for the first “mode” inelastic SDF system Itr.022 0.90 21.693 0.2 4628.4 7714.18 210.010 (kN) 8006.76 21.186 0.39 36. Properties of “modal” inelastic SDF systems Properties Ln (kg) Γn * M n (kg) “Mode” 1 2736789 1.59 22.3 7745.18 210.25 36.8525 1.529 0.2 7690.12 3876.09 18.70 36.18 210.77 21.4927 1.1.28 36.38 22.017 0.18 210.910 0.79 21.62 26.74 21.1 4569.18 210.11 22.86 21.35 36.193 0.18 210.85 36.51 2 “Mode” 2 -920860 -0.18 210.194 (kN-cm) 365100 364060 363276 362684 362235 361892 361631 361432 361279 361162 361073 361004 360951 360911 360880 360856 360838 360824 360813 Table A.151 0.3666 3740189 203.05 36.194 0.170 0.18 210.5309 488839.139 0.2 4571.46 2.5 3109.26 36.176 0.40 46.8 7618.59 36.85 0.75 21.18 210.29 36.30 37.18 210.65 1226.194 0.107 0.56 47.83 21.9 4573.5 4614.32 36.048 0.182 0.4 7647.6 × Vb1 y i ur1.1 1013.5 7624.4 7911.135 0. No.9 4570.6 4583.18 210.23 22.3 7628.18 (cm) 38.74 (kN/cm) 210.18 210.404 0.184 0.4 7672.7 4580.8 4647.18 210.18 210.6 7840.2.18 210.2406 167531.5 7633.Table A.192 0.18 210.25 36.0 4704.02 21.3 4603.44 36.56 19.082 0.1 7616.24 36.7 7639.0 4588.50 36. i 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 i Vb1 y i 0.8 4747.5 (cm) 22.86 22.0.9 4570.23 0.9 (kN) 4803.309 0.75 21.029 0.180 0.18 210.0 7619.013 0.79 0.194 0.3 7786.193 0.4 4595.05 52.18 210.

80 .

Following is step-by-step implementation of the procedure described in Section B. or from the inelastic response (or design) spectrum. Determine the total response by combining the peak “modal” responses using the SRSS combination rule of Eq. 11. (4. Typically. of the nth-“mode” inelastic SDF system (Fig. Calculate the peak roof displacement urno associated with the nth-“mode” inelastic SDF system from Eq.1 STEP-BY-STEP PROCEDURE A detailed step-by-step implementation of the modal pushover analysis (MPA) procedure is presented in this section and illustrated by an example in the next section. 6.1 to the north-south component of the El Centro (1940) ground motion scaled up by a factor of 1. B. This is the same example as solved in Appendix A. Compute the peak deformation. 81 . Repeat Steps 3 to 7 for as many “modes” as required for sufficient accuracy.8). 12. rno . 13.33 cm. 10.1.4. extract from the pushover database values of other desired responses.21).6b by solving Eq. (4.Appendix B Modal Pushover Analysis B. (3. subtract the yield hinge rotation to determine the plastic hinge rotation. Dn .8) for the peak deformation of the first-“mode” inelastic SDF system with unit mass and force-deformation relation developed in Step 4 gives D1 = 35.2 EXAMPLE The MPA procedure is implemented to calculate the response of the 9-story building described in Section 3. the first two or three “modes” will suffice.3b) with unit mass and force-deformation relation of Fig. Solving Eq. Steps 1 to 4 of the MPA are the same as those for UMRHA presented in Appendix A. From the total hinge rotation. 10. At urno . 4.18). (3. 4.5.

The total response computed by combining the peak “modal” responses using the SRSS combination rule of Eq. Peak roof displacement ur1o = Γ1 × φr1 × D1 = 1.4 for the story drifts normalized by the story height. (3. Steps 3 to 7 are repeated for first three “modes. Also included in Table 4. where results for other ground motion intensities are also included. 8. 12.18) are also included in Tables 4. The values are summarized in Table 4.366 × 1 × 35.4.” and the results are included in Tables 4.11.3 and 4. 82 .3 for the floor displacements normalized by the building height and in Table 4. At ur1o = 48. 9.3 and 4.5 are the plastic hinge rotations computed by subtracting the yield hinge rotation from the total hinge rotations.28 cm. values of floor displacements and story drifts are extracted.4.28 cm.33 = 48.1. Results of Steps 5 and 6 are summarized in Table B.

5 µ = Dn Dny urno = Γnφrn Dn (cm) Dn (cm) 2.85 µ = Dn Dny urno = Γnφrn Dn (cm) Dn (cm) 1.690 10.691 0.755 27.332 48.07 “Mode” 2 4.1.35 0.268 0.0 µ = Dn Dny urno = Γnφrn Dn (cm) Dn (cm) 3.535 14.75 µ = Dn Dny urno = Γnφrn Dn (cm) Dn (cm) 0.735 3.577 16.267 5.Table B.312 1.139 µ = Dn Dny urno = Γnφrn Dn (cm) Dn (cm) 0.023 0.551 2.252 9.185 11.513 0.33 1.184 0.229 8.27 0.18 27.007 36.25 Quantity Dn (cm) “Mode” 1 6.06 1.5 µ = Dn Dny urno = Γnφrn Dn (cm) Dn (cm) 0.38 22.676 6.13 2.766 7.969 0.126 13.913 22.52 “Mode” 3 1.35 1.755 0.379 21.332 13.457 12.660 14.467 14.59 0.154 78.678 0.117 5.50 35.82 1.79 0.36 1.504 18.0 µ = Dn Dny urno = Γnφrn Dn (cm) 83 .8451 5.28 46.395 0.4222 3.901 8.275 1.73 24.200 0.71 1.748 63.03 26.436 7.856 31.0 µ = Dn Dny urno = Γnφrn Dn (cm) Dn (cm) 1. Calculation of roof displacements urno from peak deformation of inelastic SDF systems Ground Motion Multiplier 0.37 1.05 0.70 0.450 4.03 0.52 0.225 2.37 57.367 1.25 20.

84

**Appendix C FEMA Force Distribution Calculations
**

Presented in this appendix are the calculations leading to the FEMA-273 force distributions (Fig. 5.1) used in developing the pushover curves (Fig. 5.2). These distributions were described in Section 5.1 C.1 “UNIFORM” DISTRIBUTION

The lateral force at a floor is equal to the mass at that floor, i.e., s* = m j . For convenience, the j floor forces are normalized with the base shear. The results are summarized in Table C.1.

Table C.1. FEMA273 “uniform” lateral force distribution

Floor, j

mj (10-3×kg)

s* = j

∑i mi

0.112 0.110 0.110 0.110 0.110 0.110 0.110 0.110 0.119

mj

1 2 3 4 5 6 7 8 9

503.5 494.7 494.7 494.7 494.7 494.7 494.7 494.7 534.1

C.2

EQUIVALENT LATERAL FORCE (ELF) DISTRIBUTION 85

The lateral force at a floor is computed from s* = m j h k where m j is the mass, h j is the height j j of the jth floor above the base, and the exponent k = 1 for fundamental period T1 ≤ 0.5 sec, k = 2 for fundamental period T1 > 2.5 sec; and varies linearly in between. For the selected building, T1 = 2.27 and k = 1.885. The resulting lateral forces are summarized in Table C.1. For convenience, the floor forces are normalized with the base shear.

Table C.2. FEMA273 equivalent lateral force (ELF) distribution

Floor j 1 2 3 4 5 6 7 8 9

**m jhk j (10 ×kg-m ) 371.0 1015.0 1963.2 3196.8 4707.4 6488.3 8534.2 10840.6 14471.1
**

-3 k

s* j

=

m jhk j

∑i mihik

0.007 0.020 0.038 0.062 0.091 0.126 0.165 0.210 0.281

C.3

SRSS DISTRIBUTION

The calculation of the SRSS distribution is summarized as a series of steps as follows: 1. For the nth-mode calculate the lateral forces, f jn = Γn m jφ jn An in which j denotes the floor number and An is the pseudo-acceleration of the nth-mode SDF elastic system, leading to columns 2 to 4 of Table C.3. 2. Calculate the story shears, V jn = ∑ i = j fin where j is now the story number. Implementing this step gives columns 5 to 7 of Table C.3. 3. Combine the modal story shears using SRSS rule, V j = Table C.3. 4. Back calculate the lateral forces at the floor levels from the combined story shears V j to obtain column 9 of Table C.3. 86

N

∑ n (V jn )

2

to get column 8 of

1 -438.042 0.1 1857.9 -166.3 -646.2 fj = ∑ fi i fj (2) (3) (4) 59.2 285.0 176.4 1759.2 277.8 -326.9 832.5 320.6 -359.2 105. FEMA 273 SRSS force distribution Lateral Forces Floor j (1) 1 2 3 4 5 6 7 8 9 f j1 (kN) f j2 (kN) f j3 (kN) V j1 (kN) (5) 1917.6 1233.0 1476.090 0.2 95.7 (10) 0.5 -320.1 -967.6 V j3 (kN) (7) 478.2 Lateral Force fj (kN) (9) 203.8 430.7 2065.1 87.2 97.7 234.045 0.0 354.8 374.9 -153.070 0.For convenience.367 87 .5 159.6 286.065 0.4 250.3 240.047 0.0 136.7 Vj (kN) (8) 2268.1 Story Shears V j2 (kN) (6) 1114.9 1683.6 -732.0 1231.7 1622.4 1842.3.7 95.5 -973.6 -352.7 1578.0 -5.1 832.6 319.7 355.5 215.2 200.0 980.7 694.4 400.9 1446.3 -6. Table C.2 148.3.0 381.8 1381.3 222.9 446.177 0.098 0.7 525. the lateral forces are normalized by the base shear to obtain column 10 in Table C.7 101.7 -46.9 366.1 -646.6 366.1 -525.7 374.5 -350.9 880.

FHWA Ground Anchor Circular

Work Breakdown Structure

Adam Saatler

design of headed anchor bolts

a fatigue primer for structural engineers

emergency dam assessment

soil structure interaction

SOME ADVANCES IN THE ANALYSIS OF FLUID FLOWS

Optimized Modeling and Design of Structures using SAP2000

flood hydrology manual

World Meteorological Organization Guide to Hydrological Practice Vol II

guide to hydrological practices

MayteRico_10

79127176 Hydraulics

15811820 Groundwater Resources of the World 2004

1411

363

1145

Code of Practice for the Design and Installation of Anchors

2333205 Leri SAP2000 Yap Sistemlerinin Analizi Subat2007

Transformation to Pumped Hydro

Soil Mech Book

10_vol11_6763

H1974a.pdf

- Read and print without ads
- Download to keep your version
- Edit, email or read offline

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

CANCEL

OK

You've been reading!

NO, THANKS

OK

scribd