Journal of Traffic and Transportation ~ngineeringCEnglish Edition)

2014,1(1) :62-71

Behavior of composite rigid frame bridge under

bi-directional seismic excitations

Xiaogang Liu, Jiansheng Fan *, Jianguo Nie, Guo Li

Department of Civil Engineering, Tsinghua University, Beijing, China

Abstract: Pushover analysis and time history analysis are conducted to explore the bi-directional
seismic behavior of composite steel-concrete rigid frame bridge, which is composed of RC piers
and steel-concrete composite girders. Both longitudinal and transverse directions excitations are
investigated using OpenSees. Firstly, the applicability of pushover analysis based on the funda-
mental mode is discussed. Secondly, an improved pushover analysis method considering the
contribution of higher modes is proposed, and the applicability on composite rigid frame bridg-
es under bi-directional earthquake is verified. Based on this method, an approach to predict the
displacement responses of composite rigid frame bridge under random bi-directional seismic ex-
citations by revising the elasto-plastic demand curve is also proposed. It is observed that the de-
veloped method yield a good estimate on the responses of composite rigid frame bridges under
bi-directional seismic excitations.

Key words: composite rigid frame bridge; bi-directional seismic excitation; pushover analysis;
time history analysis

height of girders correspondingly. Composite rig-

1 Introduction
Continuous rigid frame bridge has a rigid connec-
tion between the girders and piers, making them
work together under lateral loads. Meanwhile,
the piers are subjected to both axial force and
flexural moment, especially, generate negative
flexural moment in the end of girders. For this
benefit, the positive flexural moment at mid-span
will decrease. This will reduce the cross-sectional
id frame bridge comprises of steel-concrete girder
and RC piers or composite piers. The advantages
of such structural solution include better seismic
performance and convenience in constructionCNie
2011). Rigid frame bridges are widely used in
highway overpasses, viaducts and railway bridg-
es, especially in Chinese western areas. As these
bridges are usually used in the areas with great re-
quirement for seismic resistance, their seismic

Corresponding author: Jiansheng Fan, PhD, Professor.

E-mail: fanjsh@tsinghua.edu.cn.

?1994-2014 China Academic Journal Electronic Publishing House. All rights reserved. http://www.cnki.net
 Journal of Traffic and Transportation EngineeringCEnglish Edition)
performances need further research.
Although the direction of seismic excitation is
randomly varied. most of investigations only fo-
cused on uni-directional responses along bridge
longi tudinal or transverse direction. Previous
studies have confirmed that bi-directional seismic
excitations have significant influence on structure
responses. especially when the bridge structures
are under elasto-plastic stage CDe Stefano et al.
1998; Heredia-Zavoni and Machicao-Borrniuevo
2004). Furthermore. the influence of bi-direction
seismic excitations is also relative to seismic inten-
sity and bridge structural features CLopez and
Torres 2000; Wang et al. 2004). The responses of
bridge structures are dependent on the variation
of seismic excitation input direction. Usually, the
most unfavorable input direction does not coin-
cide with the bridge longitudinal or transverse di-
rectionC Li et al. 2010; Lopez and Torres 1997;
Wilson et al. 1995). Therefore. it is necessary to
conduct further analysis on the seismic behavior
of composite rigid frame bridge under bi-direc-
tional seismic excitations.
Some methods can be used for bridge seismic
behavior analysis. Time history analysis has the
best accuracy, but the computational cost is also
the greatest. Relatively. pushover analysis has
lower computational cost as well as acceptable ac-
curacyCYang et al. 2000). Pushover analysis was
first proposed in 1970s. With great previous re-
search in past years. it has become a good instru-
ment for seismic analysis. Saiidi and Sozen(1981)
proposed an approach to replace the multi degree
of freedom system (MDOF) by single degree of
freedom system CSDOF). Fajfar and Gaspersic
(1996) improved the application of capacity spec-
trum in pushover analysis. On these bases. Gupta
and Kunnath (2000) proposed a better applicable
pushover analysis approach. To account for the
influence of higher modes. Chopra and Goel
( 1999) proposed model combination pushover
analysis approach. Vidic et al. (1994) and Cho-
pra and Goel ( 1')9') also improved the capacity
spectrum to account for the elasto-plastic behav-
ior. making pushover analysis applicable in elas-
?1994-2014 China Academic Journal Electronic Publishing House. All rights reserved.
63
to-plastic seismiC response. Besides. Krawinkler
and Seneviratna ( 1998 ). Mwafy and Elnashai
(2001) and Yang et al. (2000) also conducted
much work on pushover analysis. including load
patterns. determination of target displacement.
accuracy evaluation and applicable conditions.
Pushover analysis was originally used in seismic
analysis for building structures. After Northridge
and Kobe earthquakes. this approach was also de-
veloped in seismic analysis for bridge structures.
Paraskeva et al. C2(04) first evaluated the appli-
cability of modal pushover analysis in seismic be-
havior evaluation for bridges. and Lu et al.
C2004) evaluated the applicability of pushover
analysis in steel arch bridges. In China. pushover
analysis was also used in seismic behavior evalua-
tion of bridge piers( Qian et al. 2006; Wang et al.
2000; Qin 20(8). However. little research re-
garding pushover analysis of composite rigid
frame bridge can be referred to. Furthermore.
the influence of higher modes and bi-directional
seismic excitations on the seismic behavior of
composite rigid frame bridge is also unclear. In
this study. pushover analysis in both longitudinal
and transverse directions on composite rigid frame
bridge will be conducted. and a new seismic re-
sponses prediction approach under bi-directional
seismic excitations will be proposed using push-
over analysis.
2 Pushover analysis method
Based on site condition and structure damping ra-
tio. the corresponding elastic response spectrum
for specific seismic wave can be obtained. The
S.e-Sde demand curve can be derived by Eq. (1 )
using elastic response spectrum.
Sae=filxolmax
T1 (1)
{ Sde = 4--,Sac
1(-
Applying specific load pattern on structures. an
overall load-displacement curve can be obtained
when conducting pushover analysis using FEA
method. The Sa-Sd capacity curve can be obtained
by Eq. (2) using load-displacement curve.
http://www.cnki.net
 64 Xiaogang Liu et al.

;~
established using OpenSees (Open System for
5 =
(2) Earthquake Engineering Simulation) to conduct

15 d --
r
0
t CPt.,oof

The target displacement is defined as the inter-

elasto-plastic time history analysis and pushover
analysis( Mazzoni et al. 2(06). Piers and girders
were all defined as nonlinear beam element based
section point of demand curve and capacity curve
on cross-sectional fibers. This nonlinear beam el-
in pushover analysis. If the structure is still under
ement has 5 integration points along axial direc-
elastic status, the target displacement can be de-
tion with good accuracy and computational effi-
termined by the intersection point of S.e-Sde de-
ciency. Mass of piers, girders and vehicle loads
mand curve and S,-Sd capacity curve.
were simplified as lumped mass. Rayleigh damp-
When structure is under elasto-plastic status,
ing was used in FE analysis and material damping
the elastic response spectrum needs to be revised
ratio was 5%.
to account for the influence of elasto-plastic be-
Constitutive model of concrete was selected as
havior. Vidic et al. (1994) proposed that the
Kent-Scott-Park modelCMazzoni et al. 20(6), and
elastic response spectrum could be reduced using
the uniaxial stress-strain relationship is illustrated
strength reduction coefficient R and displacement
in Fig. 1 (a). The compressive segment of the
reduction coefficient f1. The elasto-plastic S"P-Sdp
curve is defined by Eqs. (7)-(12), where: K is
demand curve can be obtained by Eqs. (3)-(6),
the concrete strength enhancement factor ac-
where c\' C" CR and Cr are determined by the
hysterical behavior and damping ratio of struc-
counting for the confinement effect; t: = O. Hfeu is
tures, as shown in Tab. 1; T g is the characteristic the concrete cylinder compressive strength; feu is

period of structures. The target displacement can the concrete cubic compressive strength; Z is the

be determined by the intersection point of S"p -SdP softening modulus in compressive softening seg-
demand curve and S. -Sd capacity curve if struc- ment; fYh is the yielding strength of hoops; Ps is
ture is under elasto-plastic status. the volume stirrup ratio; h is the width of core
I

R=c\(/l-l)'RT/T n +l T~Tn 0) concrete measuring from outer edge of stirrups;

R = C\ (/l - 1 )'R + 1 T> T" (4) Sh is the spacing of stirrups; f, is the concrete axi-
Tn = C'/l'T T g (5) al tensile strength and is defined as ft = O. 26f~~J;
JS.P = SR e

(6)
tensile stiffness is E e = 2{1E,,; tensile softening
modulus is selected as E, = O. 1 E e for easier con-
1SdP = f1 S~e = f :;: S.e vergence. The corresponding concrete material in
OpenSees is Concrete D2. Constitutive models of
rebar and steel are selected as Mene-Gotto-Pinto
Tall. 1 J>'lrameters of Vidic model (damping ratio 5 %)
model(Mazzoni et al. 20(6), and the stress-strain
Structure
c, CR relationship is illustrated in Fig. 1 (b), where E =
Hysterical loop Damping ratio
2.06 X 10 5 MPa, E p = O. 01 E. The corresponding
Q style Proportional to mass 1.00 1.00 0.65 O..,0
material in OpenSees is Steel 01.
Proportional to
Q style
instantaneous stiffness
0.75 1.000.650.30
(J, = Kr[2e - (~)'J (7)
o o

Bilinear Proportional to mass 1.35 0.95 0.75 0.20

(Je = Kf[l - Z(, -n)]~O. 2K!, "<"<u
Bilinear
Proportional to
1. 1() 0.95 0.75 0.20 (Je = D. 2Kf~ e~u (9)
instantaneous stiffness
K = 1 + PsfYh ( 10)
f.
3 Finite element model
Z = O. 5( 3 +0. 29f + 0 75. W - 0.-
O()? K ) - \ (10
The composite rigid frame bridge FE model was 145f, - 1000 . p, '\j s;:

?1994-2014 China Academic Journal Electronic Publishing House. All rights reserved. http://www.cnki.net
 ..IoJrnal ot Traffic and Transportation EngineeringCEllglish Edition) 65

E
0,2

"".
o
E
,.,
_.. -,_ ... f. :;;

, IIi'I'i~v"
.... I. -0.1
,,_-o. 179.r

-0.2
,
" " n.)
30
" "
f, .... E.
... (.) T.ft

0.2

(b) Mene-lIo11o,P,nlo $leel model

:;;

,
Fig. I Conslllluhc laws of n!;llcri:t1, -0.2
"_-0.JS7g

20 30 40
(
', =O.U02K
= O. 00" + 0. 9p./ ]l)()
( 12) " n) " "
(b) EICUlro

4 Composite rigid frame bridge pushover '.6

analysis

Composite rigid frame bridges usually have large

:;;
,.,"
span and relatively long natural vibration period. ,
making them located at the decay segment of the
-0.2
response spectrum curve. Therefore. three typical
seismic waves with different intensities lire select- -0.4 , 20 30 40
ed for an<llysis. Seismic records and response
(e)
n)
Norlhndgc
" "
spectrums for the waves arc illustrated in Fig.2.
With the variation of peak ground acceleration I'
---Tin
... III Centro
(PGA). piers will be subjected to different de-
grees of plastic deformation.
.., - Norlhridlle

Five composite rigid frame bridges. as illustra- ,:' '1.j ':

O.j ." .,', ";'. .
ted in Fig.3. are selected to conduct time history
anal)si~ and pusho\'cr anal)si~ u~ing the ~i~mic
, '" .,N':~_ ~
..... '. . .,-~':"'';'''':
~ - - - - _..: ::::.: ::..: '-= :....
\\a\cs illustrated in Fig.2. All bridge piers com-
,'----,---t,===~,=""'~.
n.)
prise of RC piers and steel-concrele composile (d) Response lpeclmms

girder~. All RC piers ha\c the same rectangular Fi&.2 Ro:ord of ~Ism". ""a\CS
cross-section of 2.6 m x 3.5 m. The reinforce
ment rallO of piers is (J. H6o. and the \olume sllr- in Tab.2. Similar results call be obtained for
rup ratio is O.H2o . Material slTcngths of concrete Bridges 2 001 .5 " .
/ .... steel plate /. and rebar {.. arc ~ll MPa. 3~5
4.1 Pushover analysis in transverse direction
MP,I and JJ5 MPa. rcspccti\cl}. In FE model.
lumped milSS model is used and mass of girders Pushover all<lIysis in tTllllSVCrSC direction is can-
and picrs arc simplified as lumped m,ISS in FE ele- dueted on all bridges using the I" modal load.
ments. Modal analysis result of Bridge 1" is shown Transverse displacement at the lOp of the middle

71994-2014 China Academic Journal Electronic Publishing House. All rights reserved. http://www.cnki.net
 66 Xiaogang Liu el al

mic wave can be obtained by procedures detailed

" " , in Section 2. Pusho\'er analysis and time history

If] I I analysis are conducted using each seismic wave.

and the comparison of target displacements by
pushover analysis and the maximum displaccmcni
(.) Bridlf ," by lime history analysis are shown in Tab.3.

" " " hb.) Mesult tomparl_ or pllliho.ft" analysis and II..,., hlstor,
analysis in bridJ:" tno'I!i"~ dindi9n

Middle pier top

Bridge Seismic
displacement( 10m) Dc"iation( <v.)
(b) Dridll~ 2" No. wa,c
l'ushover Time histroy
" '1' " " Taft 20J \(,0
"
. .,
o IJridgc

'"
El Cemro

Nonhridgc ,'"
J55
"
25

Taft 30' 2201 J5

,. -lS'}
",
E1 Cenlro

823
"
Northridge
'" 25

." . , '"
Tafl 26J J5
J. E1 (;(cntro J77 JJ

Nonhridgc "0 16

Taf! 101 ,W ,
I " El Centro

Northridge
272

-150
2111

"0
J
,
,.
Taft 212
'" J2

m
Fig.] ijridge mo<Jcls(Unit, m)
El Centro
'" 29

Tab. '1 Vihrati"" l)('riuoJ~ "nd "willes

Northridge
'"' '" "
Mode No. Vibration pcriod(sl Vibration mode If piers are not quite high. the target displace-
4.2Sfl.g Transverse ment by pushover analysis using the ,. modal load
, 2.1!175 Transverse will fit well with the maximum displacement by
J I.H7J Trans'crsc time history analysis. The deviations of pushover
, 1.7JJJ Longitudinal
anal)'sis and time history analysis for Bridge -I'~ do
not exceed 10010 for all seismic waves. Besides.
5 0.%75 Tr~nsn"sc
the increase of span number will have linle influ-
,
O.72'}2 Trlns\'crsc
ence on the deviations if lhe maximum span is un-
O.70H Trallli'"enc
, 0.558l Longitudinal
changed. However. Wilh the increase of pier
height and maximum span. the analysis resull by
9 0.45UO Longitudinal
pushover analysis becomes significantly gre,lter

'" O.41{>.l
pier and transverse bottom shear force of the mid-
dle pier are selected as the [oad-displacement-
curve. The larget displllCClTlCnl for specific seis-
Longitudinal than that by time history analysis. The devialions
of Bridges 1". 2.... 31<1 and 5'~ ,Ill exceeds 20% for
all seismic waves. which indicates thai the influ-
ence of higher modes of vibration cannOI be ncg-
lected.

11994-2014 China Academic Journal Electronic Publi5hing House. All rights roscrvcd. http://www.coki.nct
 Journal of Traffic and Transportation EngineeringCEnglish Edition) 67

To account for the influence of higher vibration
modes, an adaptive pushover analysis method was
proposed( Gupta et al. 20(0) and a modal push-
over analysis method was also proposed (Chopra
et al. 2(02). The procedures of the former meth-
od are very complex, making it difficult to be ap-
Obviously, the result of this new pushover meth-
od accounting for the influence of higher vibra-
tion modals fits well with that of time history
analysis.
Tab. 4 Result comparison of pushover analysis considering

plied. The latter method usually overestimates contribution of higher vibration modals and time history
analysis in bridge transverse direction
structural response. Therefore, a new pushover
method similar with modal pushover analysis Middle pier top
Bridge Seismic Deviation
displacement( mm)
method is proposed. The influences of the target No. wave (%)
displacements for major vibration modals are con- Pushover Timehistroy
sidered, and the final target displacement is ob- Taft 175 160 9.0
tained by the combination of target displacements lSI El Centro 390 355 10.0
for major vibration modals according to the con- Northridge 654 607 8.0
tribution of each modal (equivalent mass coeffi- Taft 251 224 12.0
cient). The calculation procedure is detailed in
2 nd
El Centro 38<) 384 1.0
Eqs. (13)-(15).
N
Northridge 6'>7 65') 6.0
(2..:mj!pji)2 Taft 226 1')5 16.D
j=1
N
( 13) yd El Centro 423 377 12.D
2..: m i!P]i
j= 1
Northridge 771 665 16.0

Mt Taft 118 110 7.0

-N-- ( 14)
El Centro 312 281 11.0
2..: mi
;=1 Northridge 485 440 10.0

Taft 183 161 14.0

d = . :. i-_- 1,--_
II
(15) El Centro 375 368 0.2

~ri Northridge 688 643 7.0

i =1
where Mt is the modal mass for the i lb vibration
modal; m; is the lumped mass; !pj; is the standard
vibration model coefficient; r; is the contribution
coefficient for the i Ib vibration modal (equivalent
mass coefficient); d; is the target displacement
by pushover analysis for major vibration modal;
d is the final target displacement by pushover
analysis method.
It should be noted that the contribution coeffi-
cient of transverse vibrations should be zero for
longitudinal vibration, and the contribution coef-
ficient of longitudinal vibrations should be zero
for transverse vibration.
New pushover analysis by the proposed method
using the transverse vibration modals in the first
10 vi bra tion modals is conducted for each bridge.
The final target displacements for all bridges by
this new pushover method are shown in Tab.4.
4.2 Pushover analysis in longitudinal direction
Pushover analysis in longitudinal direction is con-
ducted on all bridges using the 151 modal load. As pier
top displacements along longitudinal direction are al-
most the same for all piers in one bridge, longitudinal
displacement at the top of middle pier and the sum of
longitudinal bottom shear force of all the piers are se-
lected as the load-displacement for capacity curve.
The target displacement for specific seismic wave
can be the obtained by procedures detailed in Sec-
tion 2. Pushover analysis and time history analysis
are conducted using each seismic wave, and the
comparison of target displacement by pushover
analysis and the maximum displacement by time
history analysis are illustrated in Fig. 4. As can be
seen, the target displacement of the pushover

?1994-2014 China Academic Journal Electronic Publishing House. All rights reserved. http://www.cnki.net
 68 Xiaogaog Uti el at

-. ...
Fig. 5. Obviously. the result of this new pushover
_Pusbov~.
method considering the influence of higher vibra-
i, ... _Timehislory lion modals fits better wilh that of time history
analysis.
i ,,.
.;;

f &SSJ Pusbovcr
. . Timc ~1510ry

-. ...
(.)

IilliilI Pushover
~,, .,. tim Time hinory

T.n 1 Centro Norlhridge

.! Sd,micwlvc
".
0

c6 (M) Il'idgc I"

~
em Pushover
m:ra Time ~i510ry

-., ...
(b)

,...
I'lI::3 r ...boo"c,
~ _ Ti_ hillOry

,,.
.;;
Tlfi 1 Cntn!
~;,mlcwlvc

(b) Brid,e 2"

~Pusbo"cr
_Time ~'5tOry

-. .,.
f ),. _ r".hovc.

, ,,. _ Time hillory

~ ToO El Centro NClrlhridgc

'00 Seismiewue

~ -. '00
(e) Bridge)"

(5;SS!Pushovcr
~ _ Time: hiltory
c )00
Fig. 4 Comparison of pushO"cr anal)"sjSllnd time hislor}'
~

'
200
analysis in bridge lon&"IlGe di.oxlion

r". 100
analysis is obviously lower Ihan the lime history
analysis result. ovcreslimating Siructural seismic
EI CnitO
!kismie ...ve
perrorm,tOce. (d) Brodie'

Similar 10 pushover analysis accounting for Ihe fiB.:; Comparison of pu~ho,cr lInal)'Sl$ conSidering contribuuon
influence of higher vibration modals. pushover or h'Bhcr vibn'lllon modals lind liD><' h,story analysis in
analysis using the proposed method in Section -I. I bridge lon&l1udc dm~cllon

is conducted. The longilUdinal vibration rnodals

in the firsl 10 vibration madals arc conducted for
bridges. The final target displacement considering
high vibration influence modal is illustrated in
5 Pushover analysis under bi-directional
seismic excitations
There is significant difference between seismic

11994-2014 China Academic Journal Electronic Publi5hing House. All rights reserved. http://www.cnki.nct
 Journal of Traffic and Transportation Engineering( English Edition) 69

performances of longitudinal direction and trans- considering high vibration modal is conducted for
verse direction. Thus, bridge structural response all bridges in Tab.5 using the first 10 vibration
may vary with the input direction of seismic exci- modes following the procedures detailed in Sec-
tations. As input direction of actual seismic waves tions 4. 1 and 4. 2, and the final target displace-
is random, the influence of bi-directional seismic men ts E xl and E yl for longitudinal and transverse
excitations may be significant (De Stefano et al. direction respectively can be obtained. The target
1998; Heredia-Zavoni and Machicao-Borrniuevo displacement E maxl considering bi-directional seis-
2004). Previous research has confirmed that the mic excitations is defined as SRSS combination of
maximum seismic response of bridges may not oc- E x, and E y ! , namely E max1 = I E~, + E;1 . The com-
cur when seismic input in the longitudinal or parison of bi-directional pushover analysis and
transverse direction alone and the maximum seis- time history analysis is illustrated in Fig. 6, where
mic responses may be 30% larger when seismic in- E max is the maximum displacement response by
put direction is random. The results of SRSS com- time history. As can be seen, the prediction re-
bination, EuroCode-8 and Code for seismic design sults of maximum displacement response by bi-di-
of buildings may be un-conservative (Li et al. rectional pushover analysis are conservative
2011) .
enough. As the results of bi-directional pushover
The influence of bi-directional seismic excita- analysis are the biggest displacement response
tions on the reduction coefficient of elasto-plastic considering the random input of seismic excita-
response spectrum has the statistical laws for vari-
tions, it is a bit conservative in comparison with
ous vibration periods, structural displacement
time history analysis when input direction of seis-
ductility coefficients and damping ratios. There-
mic waves<Tab. 6) is not the most unfavorable.
fore, the demand curve can be revised to account
for bi-directional seismic excitations, and SRSS Tab.S Analysis models of composite rigid frame bridge
combination of the final longitudinal and trans- Span(m)
Model No. Pier height(m)
verse target displacements can be selected as the
Basic model 50 + 75 + 50 40 + 60 + 60 + 40
prediction of the maximum displacement re-
Contrast model 1" 50 + 75 + 50 60 + 60 + 60 + 611
sponse. The Sapb-Sdpb demand curve considering
Contrast model 2 nd 50 + 75 + 50 40 + 611 + 611 + 611
bi-directional seismic excitations can be derived
Contrast model yd 50 + 75 + 75 40 + 60 + 60 + 411
from Vidic model in Section 2, revising strength
reduction coefficient R and displacement ductility Contrast model 4 th 75 + 75 + 75 40 + 60 + 611 + 411

reduction coefficient p., as shown in Eqs. (16)- Contrast model 5 th 50 + 75 + 50 40 + 60 + 60 + 40

(19) . Contrast model 6 th 25 + 37.5 + 25 20 + 30 + 311 + 20

R b =1.4[cl (p.-1)C R T/T ll +1] T~Tll (16)

R b =1.4[c, (p.-1)c R +1] T>T" (17) Tab. 6 Records of seismic waves

(18) PGA(gJ
Seismic wave

i
5ae Direction x Direction y
Sapb = If
(19) Hollywood Storage. 1952 0.059 0.042
_ Sde _ l!:... 41C 2 Taft Lincoln School. 1952 0.105 0.156
SdPb - p. If - R T2 5 ae
San Fernando. 1971 0.134 0.271
Pushover analysis is conducted on composite Mexico City.1985 0.171 o.1110
rigid frame bridges in Tab. 5 following the procedures Lorna Prieta. 1989 0.220 0.276
detailed in Section 2. All parameters of piers and Parkfield Cholame. 1966 D.237 0.275
composite girders are same as detailed in Section 3. El Centro Site. 1'140 0.357 0.214
The Sapb-Sdpb demand curve considering bi-directional James RD.1979 0.478 0.367

seismic excitations is adopted and seismic waves in Northridge. 1'194 0.604 0.344

Tab. 6 are used for analysis. Pushover analysis Bonds Corner El Centro. 197,} 0.778 0.5'15

?1994-2014 China Academic Journal Electronic Publishing House. All rights reserved. http://www.cnki.net
 70
Buio;:model _Coot'ltllDodtl.-
.Conlr.1Imodell- ~ Conlrut model S
,., "ConuUI model 2"
",eOIlUlSl model)"
+ContNlIl model 6-
0.,
t
~
J
0.,
I
t
1
I
...
" "~ ,,;
= ,
~ -' u
"
u
" "
0 (No.5113S007.51222810).
" !k'JlDle ....,es
z
MS.6 ComparISOn of pusho\er anal)sis and l,rot history
anal)"Is under bi-dm:cllonal Kism".. UClIations
6 Conclusions
Pushover analysis and time history analysis arc
conducted for composite rigid frame bridge using
OpcnSecs. 'ew pushover analysis method consid-
ering the influence of higher vibration modal b
proposed. and its applicability lind accuracy in
longitudinal and transverse direction arc evalua-
ted. On this basis. the revision of the claslO-plas-
tic demand curve is proposed considering bi-direc-
tional seismic excitations. A bi-directional push-
over analysis prediction approach for target dis-
placement response is also proposed. which is
suiwblc for seismic excitations in random input
directions. Conclusions arc drawn as follows:
(I )The influence of high vibration modals
should be considered for composite rigid framc
bridge with high piers and large sp'lns. The new
pushover amllysis method by the combination of
targct disp!;lcements of major modals according to
equivalent mass coefficiclH can predict the dis-
placement response of rigid frame bridge in longi-
tudinal and transverse direction.
(2)The elasto-plastic demand curc should be rc-
\'iscd to account for the influencc of bi-directional
seismic excitations. The target displacements E
and E y1 for longitudinal and transverse direction
respcctively can be obtained using the revised Sorb-
Sdfil> demand curve. and the maximum target dis-
placement response considering bi-directional
seismic excitations can be obtained by SRSS com-
11994-2014 China Academic Journal Electronic Publishing Hoose. Ali rights reserved.
Xlaogang liu el at
bination of E and E ... This bi-directional push-
J ,
over analysis method considering the random in-
put of seismic excitations gives the acceptable pre-
diction of target displacement response.
Acknowledgments
The authors gr,lIefully appreciate the financial sup-
{X)rt provided by the National Science and Technolo-
gy Sup{X)rt Program (No. 2011BAJOlJB(2) and the
National Nlltural Scicncc Foundation of China
Rertrtnees
Chopn. A. K.. Gocl. R. K.. I'm. Capaelly demand diagram
mel hods ror eSlimaling seismic ddormalion of inelastic strue_
lures SDF Sy~lCm. R~port No. l'EER.19')'). 02. I'aeific Canh.
quake Engineering RCSCitrch Cen(er. Univcrsily of California.
HcrkclC$.
ChOrlll. A. K.. God. R. K.. 2002. A modal pusho,'cr anal)s.s
procrourc ror atimaling sci,mic demands for btllldmJS
~nhqu.a.ke Engmccrlng al'ld 5uucluTllJ D)"I'IamiC'S. 31 (3),
S61S1l2.
IX Stdaoo. M.. Faella. G .. Ramasco. R.. t'-"JIl. Inclaslic scis
mlc response of one_way planuymmclrie under bidimen~ion
al ground malions. Earthqu:.kc Engineering anll SlruclUral
Inamic. 27(4),363376.
Fajf,.. I'.. GnperSle_ P.. I'J'H>. N2 mC!hod ror the SClSmlC dam-
age anal)'IJ$ or RC b.llld,np. Eanhqu.a.ke EnSlnccnng and
SITUClural Dynamics. 2:;(1), .1146.
Gupla. B.. Kunnalh. 5. K. 2111111. Adaptile spectrabasal push
o"cr procedure ror seismiC naluallon or Slruetures. Earth.
quake Spcclra. 1M2), 3(,73')1.
Ilcredia-ZIlvoni. E.. /<.lachic;.o-llorrniuevo. C 2(~14. Response
10 onhogonal componenlS of ground mOIlOn and assc~smenl of
pcreenlage combil13l1on rules. E<!nhqu.a.ke Enginttring and
51",Clural D)namlC'!. J3L2l, HI2M.
Kr...mk1cr. II.. Scnevlratna. G D. P. K .1"9"1. Pros al'ld cons
Or a puslm"er analysis or seismic performance C,...Iu.a.lion. En
g.ne<:ring SUUClurcs. 2U( -I_t.l, -152....6-1.
Li. GIIO. Fan. Jlansheng. L,. Quanwiong. 21'11. Scism,e response
analysis of eoml)()Sile rigid frame brillge unllcr bidlreClional
e~eilalions. Journal of Harbtn InSlilUle of l'cehnology. 43
(52), 32~-USl.
L,. Quan",-ang. hn. Jiansheng. I\IC. Jianguo. 21110. Efrcci of di
J<.'C"lioa.al unccnalnlY of eanhquake &round mali<><! on SiNe
J , tural responses. Enginccnng Mechames. 27{ Ill, 135_1 -II).
Lopez. O. A .. Chopra. A. K.. Ilcrnanllez. J. J.. 2{)OIl. Crnical
rcsponst' of SlruC!urcs 10 mlllli_eomponenl eanhquake cxcila
lion. Ennhqunkc Engineering anll SlruC1Ur;.! Oynamin. 29
(11),175')1778.
Lopez. O. A .. lanes. R .. 1'J'J7, The critical angle or >ci>mic in.
eldence and lhe max.mum suuclural response. Earthquake
bttp:Ilwww.coki.net
 Journal of Traffic and Transportation Engineering(English Edition)
Engineering and Structural Dynamics, 26, 881-894.
Lu, Z. H . Usami. T . Ge. H. B., 2004. Seismic performance
evaluation of steel arch bridges against major earthquakes.
Part2: simplified verification procedure. Earthquake Engi-
neering and Structural Dynamics. 33(14): 1355-1372.
Mazzoni. S. Mckenna. F. Scott. M. H . et al. 2006. The open
sees command language manual. University of California.
Berkeles.
Mwafy. A. M. Elnashai. A. S. 2001. Static pushover versus dy-
namic collapse analysis of RC buildings. Engineering Struc-
tures. 23, 407-424.
Nie. Jianguo. 2011. Steel-concrete composite bridge. People's
Transportation Press. Beijing.
Paraskeva. T. S. Kappos. A. J . Sextos. A. G . 2006. Exten-
sion of modal pushover analysis to seismic assessment of bridg-
es. Earthquake Engineering and Struetural Dynamies. 35
(10), 1269-1293.
Qian. Jiaru. Kang. Zhao. Zhao. Zuozhou. et al.. 2006. Dis-
placement-based seismic safety assessment of existing munici-
pal bridges. Engineering Mechanics. 23(SD: 194-202.
Qin. Sifeng. 2008. Static nonlinear analysis methods for estimating
?1994-2014 China Academic Journal Electronic Publishing House. All rights reserved.
71
seismic performance for bridges. Dalian University of Tech-
nology. Dalian.
Saiidi. M. Sozen. M. A . 1981. Simple nolinear seismic analysis
of RC structures. Journal of the Structural Division-ASCE.
107(5): 937-952.
Vidic. T . Fajfar. P. Fischinger. M. 1994. Consistent inelastic
design spectra: strength and displacement. Earthquake Engi-
neering and Structural Dynamics. 23(5): 507-521.
Wang. Dongsheng. Li. Hongnan. Wang. Guoxin. 2004. Pseudo-
constant ductility inelastic spectra for bi-directional ground
motions. Earthquake Engineering and Enghineering Vibra-
tion. 24(4): 25-31.
Wang. Dongsheng. Di. Tong. Guo. Mingzhu. 2000. Estimated
seismic vulnerability of bredges by push-over method. World
Information on Earthquake Engineering. 16(2): 47-519.
Wilson. E. L.. Suharwardy. A. Habibullah. A . 1995. Aclari-
fication of the orthogonal effects in a tree-dimensional seismic
analysis. Earthquake Spectra. 11 (4): 659-666.
Yang. Pu. Li. Yingming. Wang. Yayong. 2000. Improved meth-
od of structure pushover analysis. Journal of Building Struc-
tures, 2l( 1) , 44-51.
http://www.cnki.net