Investigation of Seismic Isolation for California High-speed Rail Prototype Bridge

perchorangeSoftware and s/w Development

Dec 1, 2013 (3 years and 6 months ago)

72 views

1


Investigation of Seismic
Isolation for California

High
-
s
peed Rail Prototype Bridge


Yong Li

and
Joel P. Conte
1



ABSTRACT


California High
-
speed Rail (CHSR) alignments, initially running from San Francisco to Los Angeles via
the Central Valley and later
extending to Sacramento and San Diego, will be supported on viaducts or bridges beside
the roadbed due to the land features of terrain. Consequently, supporting bridge structures of some branches will be
located in high seismic risk regions
,

and special de
mands and constraints will be imposed on their seismic
performance considering the targeted high
-
speed train service. Seismic isolation systems, which decouple the bridge
substructure and superstructure to some extent, elongat
ing

system period and add
ing

e
nergy dissipation in the form
of hysteretic damping, are
quasi
-
rigid

considered
to achieve target seismic performance. To investigate the effects of
seismic isolation systems for the CHSR Project, a three dimensional nonlinear finite element model of a 9
-
s
pan
prototype bridge is developed using the earthquake engineering simulation software framework OpenSees (Open
System for Earthquake Engineering Simulation) and nonlinear time history analyses are performed for
operating
based
hazard level
earthquake

(OBE
)
. The effects of seismic isolation are analyzed by comparing the simulated
bridge responses with and without seismic isolation (i.e., isolated and non
-
isolated bridge). It is found that the
introduction of isolators into the bridge system significantly re
duces the force demand in the bridge piers and

decreases

the deck acceleration
. However, it increase
s

the deck displacement

and the stresses in the rails
.







Yong Li, Ph.D. Student, Department of Structural Engineering, University of California, San Diego, CA 92093

Joel P. Conte, Professor, Department of
Structural Engineering, University of California, San Diego, CA 92093

2


INTRODUCTION AND MOTIVATION


Inspired by successful high
-
speed train systems worldwide, the electrically
-
powered high
-
speed trains will
help the state of California meet ever
-
growing demands on its
transportation infrastructure
. California High
-
speed
Rail (CHSR) alignments, initially
running from San Francisco to Los Angeles via the Central Valley
,

and later
extending to Sacramento and San Diego, will be supported on viaducts or bridges beside the roadbed due to the land
features of terrain. As the CHSR bridge
-
supporting structures wil
l provide a wide range of functions for the system,
a consistent design with different objectives needs to be applied and enforced. For example
,

on the side of structural
engineers, the seismic design will be a crucial concern, because the proposed CHSR wi
ll be located close to several
major seismic faults like the San Andres and Calaveras faults

(
Petuskey
, 2011
)
. Thus, CHSR bridges or a
e
rial
structures

need to be designed with special considerations to carry dedicated high
-
speed train services during or after
seismic events. As
observed

in practice
, i
solation system
s

provide
a promising alternative

because

seismic isolators
have
demonstrated

extensive su
ccess
es

and
become widely used

throughout
the world
. Seismic isolation systems,
which decouple the bridge substructure and superstructure to some extent, elongating system period and adding
energy dissipation

in the form of hysteretic damping

(
Naeim and Ke
lly,
1999
)
,

represent an attractive alternative
to
achieve target seismic performance in this project

as well
. Consequently,
there is a strong motivation
to investigate
the
potential benefits
of seismic isolation for the CHSR Project, e.g., to
assess its
feasibility and to optimize the
seismic isolation system parameters.



PROTOTYPE BRI
DGE DESIGN AND NUMERICAL MODE
L


Selection and

Design of CHSR Prototype Bridge


With the
support

of Parsons Brinckerhoff (PB), which is assisting California in planning, designing, and
managing the construction of today’s high
-
speed rail systems, a 9
-
span prototype bridge
was selected and
designed
for the
present
study of the feasibility and optimiza
tion of isolation systems in the
context of the
CHSR Project. A
three dimensional (3D) fully nonlinear numerical model of the prototype bridge for the CHSR Project is developed
in
the

earthquake engineering simulation software f
ramework OpenSees (
Mckenna,
1997
)
.
The
prototype bridge
consisting
of 9 spans 110 ft each, is 42 ft

wide (
width of the
top of
the
box girder deck) and 48 ft high
. It

is
comprised
of
three

330
-
ft long
frames
with three spans each and two

interior structural expansion joints between
ad
jacent frames

as well as
expansion joints between each of the end frame and the corresponding abutment
, as
shown in Figure 1.

The seismic i
solation strategy is introduced into
the
CHSR prototype
b
ridge between the superstructure and
substructure, with un
i
-
directional isolators in the longitudinal direction
(
x
-
direction)
at the abutments

(i.e., the
transversal displacement of the deck relative to the abutment is restrained)

and omni
-
directional
(x
-

and y
-
directions)
isolators on top of the piers
.
One

pair
of seismic isolators
(
aligned transversally
)
supports

the superstructure on
top
of
each pier at continuous joint
s
,
while

two pairs of seismic isolators support
adjacent spans
on
top of
the piers
at
interior
expansion joints as illustrated in Figure 2.





Fig
ure

1
.

Schematic view of geometric configuration of the CHSR prototype bridge

3






Transversal section

Around
e
xpansion
j
oint

Continuous
j
oint

Fig
ure

2
.

Box girder
-

pier connection


Also, a pair of slotted hinge joint (SHJ) devices is installed across each
interior expansion joint

to maintain
continuity of the transversal displacement of the bridge deck at these joints, while allowing relative longitudinal
displacements of the two adjacent frames to accommodate deformations due to creep, shrinkage, and thermal
expansion/contraction
.
More information about the design of the superstructure/substructure components, including
the isolators, the SHJ devices, the foundations, the abutments, and the rail
-
structure connections, will be provided
below together with a description of the model
ing aspects of these components.



Numerical Model of CHSR Prototype Bridge


The finite element (
FE)

model of the bridge
takes advantage

of the current modeling capabilities of the
OpenSees framework
,

including the library of existing element and material models, as well as its flexibility and
extensibility in terms of implementing new elements for the modeling purposes
. A general description of the FE
model is presented first before elaborating the de
tails of each component in this model. The model of a

single pier
and its connection to the deck as well as the connection between the deck and the rails
are shown

in Figure
3
.

Figure
4

schematically describes how the rails and the deck are connected along

one span of the bridge. In this paper, a
deck segment refers to the deck portion of a frame (i.e., three continuous spans of bridge deck)
as shown in
Figure
5
.
At
the expansion joint, the adjacent segments of the bridge are connected through a pair of slo
tted hinge joint (SHJ)
devices as illustrated in Figure
6
. The rails are modeled as elastic beam elements and they are continuous across the
interior and abutment expansion joints, since continuous welded rails (CWR) are to be used in CHSR for lower
mainte
nance cost. To model the connections of the track (two rails) system to the box
-
girder deck through direct
fixation fasteners, and the connections between the deck and the top of the piers or abutments through the seismic
isolator bearings, a suite of line
ar elastic beam
-
column elements with exceedingly stiff (quasi
-
rigid) properties are
used herein to define the rigid offsets.



Superstructure & Substructure


The deck of the post
-
tensioned box girder bridge with typical section used for double track non
-
ba
llasted
aerial structures is modeled as a linear elastic beam (Aviram et al., 2008a). Concrete with a
c
ompressive strength
of
6.0 ksi (specified strength) at 28 days is used in the design of the post
-
tensioned box girder. A corresponding
expected strength
of 7.8 ksi based on Caltrans Seismic Design Criteria (
SDC 2010
) is used in the numerical model
(Caltrans, 2010).

Bridge piers of circular cross
-
section 8 ft in diameter are modeled using nonlinear inelastic beam
-
column
elements with fiber sections. The un
i
-
axial stress
-
strain behavior of the unconfined/confined concrete fibers and the
steel fibers is modeled using realistic material cyclic constitutive models. The expected compressive strength of the
concrete and yield strength of the reinforcing steel, fo
r the piers is taken as 6.5 ksi and 68 ksi, respectively.


4




Figure
3
.
Scheme of the FE model for a
single pier of the CHSR prototype bridge

Figure
4
.
Scheme of the FE model for a single
span
of the CHSR
prototype bridge




Figure
5
.
Scheme of the FE model for a single frame of the CHSR prototype bridge



Isolators and Slotted Hinge Joints


Due to the fact that the axial forces in the isolators at
the
expansion joints and
at the
abutments are nearly
half of the axial forces in the isolators at
the
continuous joints, two groups of seismic isolators are used in this
prototype bridge. Each isolator is modeled as a zero
-
length element with
two uncoupled

bilinear inelastic materials
fo
r horizontal behavior: one in the longitudinal direction and the other in the transversal direction of the bridge. For
the

purpose of optimization, the force
-
deformation characteristics of the seismic isolators

are
of interest
,

with
bilinear
inelastic
forc
e
-
deformation relationships, as shown in Figure
7

where
Isolator A

refers to

the
isolators with
smaller axial forces
and
Isolator B

to the isolators
with large
r

axial forces
.

Up to this point, no specific type of
isolators
(e.g.,
elastomeric rubber bearing
s, single/double/triple friction pendulums)
has been
selected
.

A


Rail L
ine 1

Rail Line 4


5


preliminary design of the generic isolators is used for
the
investigation

presented here
and will be used as starting
point for
the
optimization.

The bridge deck is composed of three seismic isolated segments separated by the expansion joints, and a
pair of SHJ devices is incorporated in the design for segmental displacement control in the transversal direction and
as a fuse in the longitudinal dire
ction. At each of the two interior expansion joints, a pair of SHJ devices is installed
between the two adjacent end diaphragms of the bridge superstructure, vertically located at the height of the shear
center of the box girder cross
-
section and at the ou
termost positions in the transversal direction. Each device is
modeled as a zero length element with a gap
-
hook elastic
-
perfectly
-
plastic spring in the longitudinal direction, and
an elastic spring in each of the transversal and vertical directions as illu
strated in Figure
8
.

The longitudinal gap/hook size is designed to be 2 in. on both compression/tension sides, respectively, to
accommodate free deformation of the deck segments under creep, shrinkage, and temperature change. In addition,
gap elements wit
h an elastic impact spring to model the potential pounding between adjacent deck segments are
included (see Figure
8
) to limit the longitudinal compressive deformation of the SHJ devices, which cannot be larger
than the expansion joint gap size. Even thoug
h these impact spring elements will most likely not be activated, they
are still inserted in the FE model of the bridge to consider the possibility of large deformations of the SHJ
connections during long return period (high hazard) earthquakes for the fut
ure work on optimization.




Fig
ure

6
.

Scheme of the FE model for
the
interior
expansion joint
s

of the CHSR prototype bridge






Isolator A (small)

Isolator B (large)

Figure
7
.

Bilinear inelastic models of f
orce
-
deformation

characteristics
of seismic isolators

6



Figure
8
.

Illustration of FE model for
the pair of
SHJ

devices across an interior expansion joint



.

Figure
9
.
Illustration of FE model for each pier foundation



Foundations & Abutments


Drilled shaft foundations are used
for
the CHSR prototype bridge.
An e
lastic foundation
stiffness
matrix
was generated using LPILE by
PB
for
a
single drilled shaft with
a
diameter
of
6.5 ft with swaying and rocking
coupled.
The footing model
is
illustrated

in
F
igure
9
.
This elastic foundation model with coupling between swaying
and rocking was implemented in the OpenSees framework, since it could not be represented by the existing zero
-
length element with uncoupled translational and rotational spring
stiffn
ess’s

(i.e., diagonal foundation stiffness
matrix).
Exceedingly stiff (quasi
-
rigid) b
eams are used to account for the rigid offset
s

in the pile cap

(see Figure
9
)
.
Also, two
linear
elastic lateral springs with stiffness of 450 kip/in are added to the
base node of each pier
to
represent the passive
soil
resistance
encountered by the
10
ft
thick pile cap.

To appropriately model the abutment behavior, the abutment model developed by Aviram et al. (2008b)
denoted as Spring Abutment is used in the model of
the CHSR prototype bridge. This model is illustrated in Figure
1
0
. It includes beam
-
column elements with exceedingly stiff (quasi
-
rigid) properties connected to the bridge deck to
define the rigid offsets needed, and longitudinal, transversal, and vertica
l nonlinear springs defining the abutment
force displacement characteristics.

The embankment springs in the longitudinal direction are defined by an elastic
-
perfectly
-
plastic (EPP) backbone curve with abutment stiffness and ultimate strength obtained from
SDC 2010
. In

 
7



Fig
ure

1
0
.

Illustration of FE model for
the a
butment
s


the

transversal direction, an EPP backbone curve modified from the one for the longitudinal direction is employed to
represent the backfill, wing walls and pile foundation system behavior based on Maroney and Chai (1994). The
abutment model parameters may be
updated later on, since the CHSR Design Criteria has formulated specifications
for the backfill materials behind the abutment (i.e., cement mixed), which should be stronger than Caltrans.



Rails & Connections
between

Rails and Bridge Deck


The r
ails are modeled as
linear
elastic beam
-
column elements with the
section
properties for 141RE.
They
are made of steel with
a yield

strength of 74 ksi,
an
ultimate strength of 142.5 ksi,
and an allowable stress of
± 23 ksi
at the
OBE hazard level.

The non
-
b
allasted tracks are connected to the track base (bridge deck or subgrade beyond the bridge) with
direction fixation fasteners
(
California High
-
speed Rail Authority, 2012; Petrangeli, 2008
)
. A series of coupling
springs on a per track basis to represent pai
r
s

of fasteners are included in the prototype bridge model

(see Figure 11)
.
In the Track
-
Structure Interaction model, rail
-
structure interaction elements discretized/lumped
every 7.9 ft

are
adopted to idealize the
connection
system

between
rail

and deck
, r
esulting in 14 fastener elements per span
consistent with

the fastener spacing modeling requirements specified in the CHST project design criteria. In the
vertical and transversal directions,
the rail fastener elements are modeled as linear
elastic springs

with stiffness of
4,000 kip/ft/ft and 450 kip/ft/ft
,

respectively
,

per foot of track (two rails)
. In the longitudinal direction, the
fasteners
for non
-
ballasted track are represented by
bi
-
linear
inelastic
coupling springs

with parameters depending on the

vertical load (none or train) acting on the rails as reported
in Table
I
.

These bilinear springs
characterize

the
resistance of the bridge deck to the deformation of the rail track
(pair of rails)
with respect to the deck.
Figure 1
2

presents the elevation view of the
modeling
scheme
used for the full
track
-
structure interaction
system in the
longitudinal direction.
Based on the CHST project design criteria, no dampers are incorporated
across the interaction
layer between the rail tra
ck and the bridge deck.

To appropriately model the boundary conditions of the rai
l
-
fastener
system, the rails and coupling fasteners
are extended a
distance of 361 ft

from the face of the abutment
into the embankment (or natural ground)
at both ends
of th
e bridge. And a horizontal boundary spring representing the
rail
-
fastener system behavior beyond
the
boundaries of the model
is taken as
elastic
-
perfectly plastic with
a
stiffness
of
24
,
200 kips/ft and
a strength
capacity
of
40.3 kips. The boundary spring
needs to be verified to
remain
elastic during the
analysis;

otherwise
the
extension
of the rails and coupling fasteners beyond the bridge must be elongated.



T
ABLE

I
.
RAIL
-
STRUCTURE CONNECTION MODELING PROPERTIES

Direction

Material
Type

Stiffness per
Foot

of Track
(kip/ft/ft)

Yield
Displacement

(in)

Actual

Fastener
Spacing (in)

Fastener Spacing

Modeling
Requirement

Longitudinal

EPP

60(120)*

0.02

27

(1) # of Springs per
Span ≥10


(2)

Spacing ≤10 feet

Transversal

Elastic

450

-

27

Vertical

Elastic

4000

-

27

*Note: Inside the parentheses

is the value used for
the
loaded case.

8



Fig
ure

1
1
.


Modeling of
c
onnection between
track (pair of rails) and bridge girder
with
d
irect
f
ixation
f
asteners



Figure

1
2
.


Elevation
v
iew of the
f
ull
t
rack
-
s
tructure
i
nteraction
m
odel for
the
CHSR
p
rototype
b
ridge



GROUND MOTION INPUT AND SEISMIC RESPONSE SIMULATION RESULTS


Ground Motion Input


It is envisioned that seismic isolation systems will provide cost effective solutions for the CHSR bridges located

in
the San Francisco and Los Angeles highly seismic areas. Thus, 40 earthquake records were selected from the NGA
ground motion database based on the results of probabilistic seismic hazard analysis for the bridge site location (San
Jose, 37.33N,
-
121.90W
) and site condition with an average shear wave velocity V
s30

= 300 m/s. This paper
investigates the seismic response of the CHSR prototype bridge to one of these ground motions, namely the 1999
Chi
-
Chi earthquake (M
w

= 7.6) in Taiwan recorded at station T
CU05 and shown in Figure 1
3
. According to the
design criteria of the CHST Project, Track
-
structure interaction analysis has to be performed at the Operating Basis
Earthquake (OBE) hazard level corresponding to a return period of 50 years. To match the OBE
hazard level, the
Chi
-
Chi TCU05 earthquake record is scaled by 0.26 to match the 50
-
year return period spectral acceleration at the
fundamental period T
1

= 0.87 sec of the bridge and a damping ratio of 5%. The first mode of the bridge is in the
transversa
l direction. The two unscaled horizontal ground motion components applied in the longitudinal (TCU065
-
E) and transversal (TCU
-
65
-
N) directions of the bridge are shown in Figure 1
3

together with their 5% damped
displacement response spectra.

9




(a)
Acceleration time history

(b) Displacement response spectra (ξ = 5%)

Figure 1
3
.

Ground motion input: Chi
-
Chi, Taiwan (1999), Station TCU05



Seismic
Response
Simulation Results


The nonlinear model of the seismic isolated CHSR prototype bridge defined above is subjected to the
above defined
ground motion excitation at
the OBE hazard level and
the
simulated seismic response of the bridge is
compared to the response obtained from
the same bridge model but without seismic isolation (i.e., with rigid
connection
s between the bridge girder and the piers).

The longitudinal and transversal
force
-
deformation

response
s

of isolator #13 supporting the middle
deck
segment on top of pier #5
(s
ee Figure 1
2
)
are presented in Figure 1
4
. The
maximum deformation
(relative horizontal displacement across the isolator height) of this isolator is 1.2in

in
the
longitudinal direction, which is
smaller than

abutment gap size
, and 2.5
in

in
the
transversal d
irection.
I
n the case of
this
OBE

event
, there is no pounding at the abutments and the SHJ
devices do
not get engaged in
the
longitudinal
direction.

As illustrated
in the comparative
results of the bridge response
between
the non
-
isolated and isolated
cases
shown
in Figure
s

1
5

and 1
6
,
the

pier top node displacement
relative to the ground
is reduced by
over
50% at the cost
of increasing
the
deck displacement

relative to the ground
. Consequently,
with seismic isolation,
the pier column
remains linear elas
tic under
this
OBE

level

earthquake, and the total base shear force
for
the bridge system is highly
reduced, which is promising in reducing foundation
size and
cost significantly for the CHSR project. At the same
time, the absolute acceleration generated
o
n the bridge
deck is reduced. This is beneficial to the
safety

of
any
train
operating on the
bridge during an earthquake.






(a) Longitudinal direction

(b) Transversal direction

Figure 1
4
.

L
ongitudinal and transversal
force
-
deformation

response
of isolator #13 on top of pier #5

10





(a) Displacement
h
istor
ies

(Relative

to the ground
)

(b) Acceleration
h
istor
ies

(Absolute)

Fig
ure

1
5
.


Comparison of
longitudinal
bridge
r
esponse between
n
on
-
isolated and
i
solated
cases

for pier and deck






( a ) M o m e n t
c
u r v a t u r e o f
b a s e
s
e c t i o n

( b ) T o t a l
b
a s e
s
h e a r
f
o r c e
t
i m e
h
i s t o r y

F i g u r e 1
6
.


Comparison of pier base
s
ection
r
esponse and
t
otal
b
ase
s
hear
f
orce in
the
l
ongitudinal
d
irection

between non
-
isolated and isolated cases



The stresses generated
in the rails during an OBE earthquake are of great concern to the CHST Project
authority. Therefore, it is important to examine the effects of the seismic isolators on the stresses in the rails. Figures
1
7

and 1
8

show the envelope of the peak rail stresses

generated during the OBE earthquake considered here over all
elastic beam elements composing the outer rail line 1 (marked as the pink line in Figures
5

and 1
2
) of the double
track system. The peak stress envelopes due to (a) the axial force P in the rail

(mainly due to the longitudinal seismic
response of the bridge), (b) the bending of the rail in the horizontal plane M
z

(mainly due to the transversal seismic
response of the bridge), and (c) the bending of the rail in the vertical plane M
y

(mainly due to

the longitudinal
seismic response of the bridge) are given in Figures 1
7

(a), (b), and (c), respectively, and overlapped in Figure 1
7
(d).
It is observed that: (1) the peaks of the stress envelope along the rail occur at the abutment and interior expansio
n
joints (2) the peaks of the stress envelope are spikier at the abutment joints, especially for the stress due to
transversal bending, and the stress due to the axial force dominates as seen from the overlap plot in Figure 1
7
(d); (3)
the rail peak stress

envelopes are higher for the seismic isolated bridge than for the non
-
isolated one, but the shapes
of the peak stress envelope along the rail are similar. The envelopes of the peak stress due to the combined effect of
the axial force and one bending comp
onent at a time are plotted in Figure 1
8
, and need to be checked against the
design allowable stress for the OBE hazard level.

11




(a) Envelop
of

p
eak
s
tress
d
ue to
a
xial
f
orce

(b) Envelop
of
p
eak
s
tress
d
ue to
t
ransversal
b
ending

M
z



(c) Envelop
of
p
eak
s
tress
d
ue to
v
ertical
b
ending

M
y

(d) Envelop
c
omparison for
p
eak
s
tress

Fig
ure 1
7
.


Rail
s
tress
c
omparison between
n
on
-
isolated and
i
solated
cases





(a) Envelope of peak stress due to axial force
P
and
vertical bending

M
y

(b) Envelope of peak stress due to axial force P and
vertical bending M
z

Figure 1
8
.

Rail
s
tress
c
omparison between
n
on
-
isolated and
i
solated
cases



CONCLUDING REMARKS AND FUTURE WORK


A
3D nonlinear FE model of a CHSR prototype bridge including track
-
structure interaction is implemented
in OpenSees, and
the bridge
seismic response is simulated
to investigate the effects
of seismic isolators
between the
substructure and superstructure
in
t
his
CHSR
prototype b
ridge
.
The pros and cons of seismic isolation are observed

12


for the Operating Basis Earthquake (OBE) level (return period of 50 years)
,
with significant
potential benefits for
the
bridge
structure

(i.e., reducing the pier deformations and forces)
,
but
higher
stresses are imposed on the rails.
Therefore, a feasibility and optimization study of seismic isolators in the CHSR project needs to be carried out

in a
probabilistic performance
-
based earthqua
ke engineering framework
to optimize the bridge seismic performance,
while satisfying the design constraints including the limitation of the stresses in the rails for OBE events
.
T
he
FE

optimization framework OpenSees
-
SNOPT
(Gu

et al.
, 2012)

will be used as a tool for
solving this
probabilistic
performance
-
based optimum seismic design problem as applied to
the
CHSR prototype bridge

considered in this
paper

and used as a testbed structure
.



A
CKNOWLEGEMENTS


Support of this research by the Pacific Earthquake Engineering Research (PEER) Center’s Transportation Systems
Research Program under Award No.
1107
-
NCTRCJ
is gratefully acknowledged.
The authors wish to thank Prof.
Steve Mahin (U.C. Berkeley) and Thomas B
. Jackson,

Pang Yen Lin,
Kongsak Pugasap (Parsons Brinckerhoff), and
Roy Imbsen (Earthquake Protection System) for providing insightful discussion
s

on the
selection and design of the
CHSR prototype bridge
used
in this study.
Any opinions, findings, conclus
ions, or recommendations expressed in
this publication are those of the authors and do not necessarily reflect the views of the sponsor.



REFERENCES


Petuskey

A.

2011.


Superstructure
I
solation
S
trategies for California High
-
Speed Rail
V
iaducts
,


M.Eng.
thesis
,
University of California, Berkeley.

Naeim F., Kelly J.M.

1999.
Design of
S
eismic
I
solated
S
tructures: from
T
heory to
P
ractice
, John Wiley & Sons.
New York, NY.

Mckenna F.

1997.


Object
-
oriented
F
inite
E
lement
P
rogramming:
F
rameworks for
A
nalysis,
A
lgorithms and
P
arallel
C
omputing
,



Ph.D. Dissertation
, Uni
versity of California, Berkeley
.

Aviram, A., Mackie, K. and Stojadinovic, B.

2008
a
.


Guidelines for
N
onlinear
A
nalysis of
B
ridge
S
tructures in
California
,


Technical Report 2008/03
, Pacific Earth
quake Engineering Research Center, Universi
ty of
California, Berkeley
.

Caltrans SDC.
2010.
Caltrans
S
eismic
D
esign
C
riteria
, California Department of Transportation, Sacramento,
California.

Priestl
e
y, M.
J.N., Seible, F., and Calvi, M.

1996.

Seismic
D
esign and
R
etrofit of
B
ridges
, Jo
hn Wiley and Sons,
New York, NY
.

Aviram, A., Mackie, K.R., and Stojadinovic, B
.

2008
b
. “
Effect of
A
butment
M
odeling on the Seismic
R
esponse of
B
ridge
S
tructures
,



Earthquake Engineering and Engineering Vibration
,

Vol.
7
, No.
4
,
pp.
395
-
402
.

Maroney B
.
H
.

and Chai Y
.
H.

1994. “
Seismic
D
esign and
R
etrofitting of
R
einforced
C
oncrete
B
ridges
,


Proceedings
of 2nd International Workshop, Earthquake Commission of New Zealand
, Queenstown, New Zealand.

California
High
-
s
peed
Rail A
uthority.
2012.
California High
-
speed
Train

Project
D
esign
C
riteria
, San Francisco,
California.

Petrangeli
M
.
,
Tamagno
C
.
, and
Tortolini
P
.

2008. “
Numerical
A
nalysis of
T
rack
-
S
tructure
I
nteraction and
T
ime
D
omain
R
esonance
,


Journal of Rail
and Rapid
Transit
, pp. 345
-
353
.

Gu Q. Barbato M., and Conte J.P.

2012.


OpenSees
-
SNOPT
F
ramework for
F
inite
-
E
lement
-
B
ased
O
ptimization of
S
tructural and
G
eotechnical
S
ystems
,


Journal of Structural Engineering
, ASCE, Vol. 137, No. 6, pp. 822
-
834.