You are on page 1of 10

5th International Conference on

Structural Health Monitoring of Intelligent Infrastructure (SHMII-5) 2011
11-15 December 2011, Cancún, México


- 1 -
DIAGNOSIS AND PROGNOSIS OF
STONECUTTERS BRIDGE
BASED ON STRUCTURAL HEALTH
MONITORING SYSTEM
Songye Zhu

The Hong Kong Polytechnic University,
Hong Kong, China
You-lin Xu

The Hong Kong Polytechnic University,
Hong Kong, China
Kai-yuen Wong

Highways Department, The Government
of Hong Kong S.A.R., China
Yue Zheng

The Hong Kong Polytechnic University,
Hong Kong, China
As a critical type of infrastructure in modern society, long-span bridges always draw
considerable attention to their serviceability and safety. However, they are still exposed
to substantial uncertainties in their long service life, e.g. performance deterioration due
to harsh environment, potential failure due to unexpected natural and manmade hazards.
Therefore, structural health monitoring (SHM) system has been embraced worldwide as
an efficient means to provide long-term diagnosis and prognosis of bridge conditions
and reliabilities. For example, a comprehensive SHM system has been recently installed
on Stonecutters Bridge by Hong Kong’s Highways Department, and it is composed of
totally 1,571 sensors in 15 different types.
Such an SHM system would considerably improve the accuracy of long-term diagnosis
and prognosis of the bridge health. This paper summarizes the work on the safety
evaluation of Stonecutters Bridge under extreme events based on the installed SHM
systems. The paper comprises three major parts: (1) establishing and updating the FEM
of Stonecutters Bridge according to the field measurement data; (2) predicting seismic
performance of Stonecutters Bridge under non-uniform ground motions at multi-
supports, in which conditional simulation of ground motions based on measured records
at limited points takes into account the effect of wave passage, local soil conditions and
incoherence; (3) performing fully coupled buffeting analysis of Stonecutters Bridge
under non-stationary typhoon, in which time-varying features of mean wind speed and
evolutionary power spectra density extracted from field measurement are appropriately
considered. The latter two analyses employ the SHM-oriented FEM developed in Task
1. The effects of non-uniform seismic ground motions and non-stationary typhoon loads
on the bridge’s response and safety are discussed based on the comparison with
conventional methods considering either uniform seismic ground motion or stationary
wind loads.
Corresponding author’s email: Dr. Songye Zhu, ceszhu@polyu.edu.hk
5th International Conference on
Structural Health Monitoring of Intelligent Infrastructure (SHMII-5) 2011
11-15 December 2011, Cancún, México


- 2 -
DIAGNOSIS AND PROGNOSIS OF
STONECUTTERS BRIDGES BASED ON
STRUCTURAL HEALTH MONITORING
SYSTEM
Songye Zhu
1
, You-lin Xu
1
, Kai-yuen Wong
2
, Yue Zheng
1
, Shunlong Li
3
, Wen-feng
Huang
1
, and Liang Hu
1
1
The Hong Kong Polytechnic University, Hong Kong, China
2
Highways Department, The Government of Hong Kong S.A.R., Hong Kong China
3
Harbin Institute of Technology, Harbin, China
ABSTRACT: As a critical type of infrastructure in modern society, long-span bridges
always draw considerable attention to their serviceability and safety. However, they are
still exposed to substantial uncertainties in their long service life, e.g. performance
deterioration due to harsh environment, potential failure due to unexpected natural and
manmade hazards. Therefore, structural health monitoring (SHM) system has been
embraced worldwide as an efficient means to provide long-term diagnosis and
prognosis of bridge conditions and reliabilities. For example, a comprehensive SHM
system has been recently installed on Stonecutters Bridge by Hong Kong’s Highways
Department, and it is composed of totally 1,571 sensors in 15 different types.
Such an SHM system would considerably improve the accuracy of long-term diagnosis
and prognosis of the bridge health in the sense that: (1) it provides comprehensive and
real-time information of various loads, bridge conditions and bridge response; (2) it
provides an efficient way to validate and update the finite element model (FEM) used;
and (3) it is able to timely capture any unexpected catastrophic events and consequent
structural change. This paper summarizes the work on the safety evaluation of
Stonecutters Bridge under extreme events based on the installed SHM systems. The
paper comprises three major parts: (1) establishing and updating the FEM of
Stonecutters Bridge according to the field measurement data; (2) predicting seismic
performance of Stonecutters Bridge under non-uniform ground motions at multi-
supports, in which conditional simulation of ground motions based on measured records
at limited points takes into account the effect of wave passage, local soil conditions and
incoherence; (3) performing fully coupled buffeting analysis of Stonecutters Bridge
under non-stationary typhoon, in which time-varying features of mean wind speed and
evolutionary power spectra density extracted from filed measurement are appropriately
considered. The latter two analyses employ the SHM-oriented FEM developed in Task
1. The effects of non-uniform seismic ground motions and non-stationary typhoon loads
on the bridge’s response and safety are discussed based on the comparison with
conventional methods considering either uniform seismic ground motions or stationary
wind loads.
5th International Conference on
Structural Health Monitoring of Intelligent Infrastructure (SHMII-5) 2011
11-15 December 2011, Cancún, México


- 3 -
1 INTRODUCTION
In their long service life, long-span bridges are always exposed to substantial
uncertainties induced by aging effect, harsh environment, improper overloading,
unexpected natural and manmade hazards, etc. Structural health monitoring system
(SHMS) has been embraced as a cutting-edge technology in the past two decades to
ensure the functionality and safety of bridges as well as other infrastructures. It is
evidenced by a number of SHMS equipped on recently built long-span bridges. Well-
known examples include Tsing-Ma Bridge in Hong Kong, Sutong Bridge in China, Seo-
Hae Bridge in Korea, Akashi Kaikyo Bridge in Japan, and so on. SHMS provides
valuable information to load monitoring and characterization, system identification,
model updating, damage detection, online condition monitoring, life-cycle performance
assessment, design verification, maintenance and management of long-span bridges.
Stonecutters Bridge, a two cable plane cable-stayed bridge, has been recently built in
Hong Kong. As the world’s second longest cable-stayed bridge, it carries dual 3-lane
highway traffic. The main/central span of Stonecutters Bridge is 1018m, and the total
length including side spans is 1596m. The noteworthy structural characteristics of
Stonecutters Bridge include: (1) separated and streamlined steel twin-box girders at
central span; (2) hybrid (concrete/steel) separated box girders at side spans; (3) hybrid
(concrete/stainless steel) single tower leg; (4) floating deck-to-tower connections in
central span and rigid deck-to-pier connections at side spans. A comprehensive SHMS,
termed Structural Health Monitoring and Safety Evaluation System (SHM&SES), has
been deployed for monitoring and evaluation of Stonecutter Bridge under in-service
condition. The SHM&SES is composed of nine interrelated components, namely,
sensory system, data acquisition system, cabling network system, portable data
acquisition system, data processing and control system, portable inspection and
maintenance system, structural health rating system, structural health evaluation system
and data management system (Wong 2010). The sensory system is composed of 1571
sensors in 15 different types, namely anemometers, barometers, hygrometers,
temperature sensors, corrosion cells, accelerometers, dynamic weigh–in-motion
stations, video cameras, dynamic strain gauges, static strain gauges, GPSs, tiltmeters,
bearing sensors, buffer sensors and tension-magnetic sensors (Wong 2010). Figure 1
shows the layout of the sensory system.
The Hong Kong Polytechnic University, in collaboration with Hong Kong’s Highways
Department, has been developing SHM-aided structural diagnosis and prognosis tools
for Stonecutters Bridge under in-service conditions based on since 2010. The project
attempts to take advantage of real-time field measurement data by the aforementioned
comprehensive SHMS and to better evaluate serviceability and safety performance of
Stonecutters Bridge in normal operational conditions and extreme events. This paper
summarizes the work completed in the Phase 1 of the project, and the methodology and
results are presented in three main tasks: (1) establishing and updating finite element
model (FEM) according to the field measurement; (2) predicting seismic performance
under non-uniform ground motions at multiple supports; (3) performing fully coupled
buffeting analysis under non-stationary typhoon. The latter two analyses employed the
SHM-oriented FEM developed in Task 1. The effects of non-uniform seismic ground
motions and non-stationary typhoon loads on the bridge’s response and safety are
discussed in details.
5th International Conference on
Structural Health Monitoring of Intelligent Infrastructure (SHMII-5) 2011
11-15 December 2011, Cancún, México


- 4 -
2 SHM-ORIENTED FINITE ELEMENT MODELING
Although the SHMS on Stonecutters Bridge is one of the most comprehensive systems
in the world, the number of sensors is always limited for such a large-scale structure,
and locations of structural defects or degradation may not be at the same positions as the
sensors. Therefore, the
WIM WIM
WIM WIM
Anemometers Anemometers (24) (24)
Fixed Fixed Sero Sero- -Type Accelerometers Type Accelerometers (58) (58)
Temperature Sensors Temperature Sensors (388) (388)
Dynamic Strain Gauges (678) Dynamic Strain Gauges (678)
Static Strain Gauges (158) Static Strain Gauges (158)
Global Positioning Systems (20) Global Positioning Systems (20)
Displacement Transducers Displacement Transducers (34) (34)
Buffer Sensors (18) Buffer Sensors (18)
Bearing Sensors (12) Bearing Sensors (12)
Electro Electro- -Magnetic Sensors (32) Magnetic Sensors (32)
Barometers, Rainfall Gauges & Hygrometers (28) Barometers, Rainfall Gauges & Hygrometers (28)
Corrosion Cells (33) Corrosion Cells (33)
Digital Video Cameras (18) Digital Video Cameras (18)
Dynamic Weigh Dynamic Weigh- -in in- -Motion Stations (4) Motion Stations (4) WIM WIM
Stonecutters Tower Stonecutters Tower
Tsing Yi Tower Tsing Yi Tower
Instrumentation Layout in Instrumentation Layout in
Stonecutters Bridge Stonecutters Bridge
Total No. of Sensors Total No. of Sensors : 1505 : 1505

Figure 1. Instrumentation Layout in Stonecutters Bridge (Wong 2010)

East
South
North

Figure 2. 3D FEM of Stonecutters Bridge

f
1
= 0.1563 Hz f
2
= 0.2047 Hz f
3
= 0.2105 Hz


f
4
= 0.2119 Hz f
5
= 0.2594 Hz f
6
= 0.3272 Hz
5th International Conference on
Structural Health Monitoring of Intelligent Infrastructure (SHMII-5) 2011
11-15 December 2011, Cancún, México


- 5 -

Figure 3. The first six frequencies and mode shapes of Stonecutters Bridge
development of a model that relates the bridge behavior and conditions to the
measurement at limited locations from the SHMS becomes an imperative task. To
develop such a structural performance relationship model, a sophisticated 3D full-scale
FEM was established and calibrated for structural health diagnosis and prognosis
purpose. Without such a dedicate FEM, it is very difficult, if not impossible, to identify
the critical locations and components and to quantify the corresponding levels of
vulnerability under various loading conditions (such as dead load, live load, temperature
load, wind load, seismic load, accidental load, and so on).
Figure 2 shows the 3D FEM of Stonecutters Bridge built using ANSYS software, in
which all the major elements, connections and boundary conditions, including stay
cables, twin-box deck, composite tower, side-span piers, prestress steel tendon,
hydraulic buffers, horizontal bearings, etc., were properly modeled. The final bridge
model was composed of over 4,700 elements. The initial model was established based
on the design drawing, and subsequently the static and dynamic model updating was
carried out according to in-situ measured results, including deck profile, cable forces
and modal frequencies. Selected structural parameters were optimized in order to make
computed results closely match the measured results. In general, a good agreement was
achieved after model updating. For example, the maximum upwards and downwards
differences of the bridge deck profile are 2.43 cm and -2.79 cm respectively in the main
span, and 3.69 cm and -2.59 cm respectively in the side spans, all of which are minimal
in comparison with the 1018-m major span; the maximum difference of the first ten
frequencies is less than 4.6%, and the average error is only 2.1%. Figure 3 shows the
first six natural frequencies and mode shapes obtained from the FEM. Table 1 presents
the comparison of the cable forces in a total of 224 stay cables. All the differences of
cable forces fall below 10% after the updating of FEM. Therefore, the FEM is
considered as a good representation of the real Stonecutters Bridge, and it, together with
the advanced SHMS, would serve as a convenient and powerful tool for the health
monitoring and safety evaluation of Stonecutters Bridge.
Table 1. Difference of cable forces between calculation and measurement
Relative
error ε
Number
of cables
Cable locations
ε < 5% 157
5% < ε <
7%
45
7% < ε <
10%
22
104N,325N,305N,302N,418N,419N,424N,128S,127S,
125S,120S,
115S,202S,228S,321S,305S,303S,415S,416S,420S,421
S,426S
ε > 10% 0
Total 224
5th International Conference on
Structural Health Monitoring of Intelligent Infrastructure (SHMII-5) 2011
11-15 December 2011, Cancún, México


- 6 -
3 SEISMIC ANALYSIS UNDER SPATIALLY VARYING EARTHQUAKES
The extremely long span of many cable-supported bridges requires a special
consideration of spatially varying ground motions in seismic analyses. Particularly
when a bridge is located in a complex terrain, the spatial variation is not only affected
by wave passage and incoherence but also by different local soil conditions. As a result,
long-span bridges are often subjected non-uniform seismic ground motions at multiple
supports, which may cause quantitative and qualitative difference in bridge responses as
compared with those induced by synchronous motions (Abdelghaffar 1982;
Abdelghaffar & Rubin, 1983). Although seismometers in the SHMS of Stonecutters
Bridge could offer useful measurement of site-specific earthquake records, the limited
number makes it still difficult to capture the spatial variability existing in a large area
accurately. This task took advantage of the seismic ground motions recorded at the
selected points, and simulated non-uniform ground motions at the multiple supports of
Stonecutters Bridge using the conditional simulation method. Subsequently, the seismic
responses of Stonecutters Bridge were analyzed accordingly (Zheng et al. 2011).
The conditional simulation method developed by Vanmarcke et al. (1993) was adopted
in this task to generate properly-correlated seismic ground motions at an arbitrary set of
closely-spaced points based on available seismograms. In this method, non-ergodic and
homogeneous time processes of ground motion Z
i
(t) at the point i can be expressed as
1
( ) [ cos( ) sin( )
K
i ik k ik k
k
Z t A t B t ω ω
=
= +

(1)
where coefficients A
ik
and B
ik
are zero-mean random variables, and ω
k
is the kth
frequency component. The covariance between the coefficients at points i and j, i.e.
( ) ) )
ij k ik jk ik jk
C E A A E B B ω = = ( ( , can be computed by (Haredia-Zavoni 1993)
{ }
2
2
1
( ) ( ) , for 1
2
1
( ) ( ) ( ) ( ) ( ) , for 2, , / 2
4
( ) ( ) , for / 2 1
k
k K k
k
ij k
ij k ij k ij K k
ij k
G k
C G G k K
G k K
ω
ω ω
ω
ρ γ ω ω
ω ρ γ ω ρ γ ω ω
ρ γ ω ω
− +
− +
¦
∆ =
¦
¦
¦
= + ∆ = ⋅⋅⋅
´
¦
∆ = + ¦
¦
¹
(2)
where γ
ij
= x
i
- x
j
is the relative position vector, G(ω) is the one-side point spectral
density function, and ρ
ωk

ij
) is the frequency-dependent spatial correlation function. In
this study, the following function proposed by Harichandran & Vanmarcke (1986) was
adopted.
{ } { }
1/ 2
2.78
( ) 0.736exp 5.063 / 0.264exp 0.744 /
5210 1
2.17
k
ij ij ij ω ω ω
ω
ρ γ γ θ γ θ
ω
θ
π

= − + −
¦ ¹
¦ ¦ | |
= +
´ `
|
\ ¹
¦ ¦
¹ )
(3)
where θ
ω
is the frequency-dependent scale of fluctuation (Vanmarcke 1983). The
random coefficients A
jk
and B
jk
at an arbitrary location j can be generated according to
the correlation with point i where the ground motion was recorded.
5th International Conference on
Structural Health Monitoring of Intelligent Infrastructure (SHMII-5) 2011
11-15 December 2011, Cancún, México


- 7 -
The seismogram recorded at the site of Ting Kau Bridge that is close to Stonecutters
Bridge was used as the seed because the SHMS installed on Stonecutters Bridge has not
recorded any seismic-induced ground motion data yet. The seismogram was adjusted in
both frequency domain and time domain in order to remove the noise contamination and
amplify the amplitude. It was assumed that the ground acceleration at Pier 1 of
Stonecutters Bridge (as shown in Figure 4(a)), and the ground motions at the other
seven piers and two towers were generated using aforementioned method. The
simulated ground motions at two tower bases are also shown in Figure 4.
In dynamic analysis, different absolute displacement time histories were applied at the
bases of the bridge piers and towers. Meanwhile, the bridge responses under uniform
ground motion were also analyzed for the comparison purpose. Figure 5 shows the
typical seismic responses under uniform and non-uniform ground motions, including the
displacement time history at the top of West Tower and stress time history in Stay Cable
228S connecting West Tower and mid-span of the bridge deck. Both indicate that
neglecting spatial variation in ground motions may considerably underestimate seismic
demand at certain locations. Apparent differences between uniform and non-uniform
ground motions could be observed in the tower displacements, bending moments at
tower bases, stress in girders and stay cables, etc. All these results imply that the spatial
variation in seismic ground motions due to wave passage, local soil conditions and
incoherence effect needs to be appropriately taken into account in the seismic analysis
of long-span cable-stayed bridges such as Stonecutters Bridge. More information about
the methodology and results in this task can be found in Zheng et al. (2011).
0 2 4 6 8 10 12 14 16 18
-1.6
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
1.6
A
c
c
e
l
e
r
a
t
i
o
n
(
m
2
/
s
)
Time(s)
0 2 4 6 8 10 12 14 16 18
-1.6
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
1.6
A
c
c
e
l
e
r
a
t
i
o
n
(
m
2
/
s
)
Time(s)
0 2 4 6 8 10 12 14 16 18
-1.6
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
1.6
A
c
c
e
l
e
r
a
t
i
o
n
(
m
2
/
s
)
Time(s)

(a) at pier 1 (seed) (b) at west tower (c) at east tower
Figure 4. Seed ground acceleration and simulated ground accelerations
0 2 4 6 8 10 12 14 16 18
-0.04
-0.03
-0.02
-0.01
0.00
0.01
0.02
0.03
0.04
D
i
s
p
l
a
c
e
m
e
n
t
(
m
)
Time(s)
Uniform
Non-uniform

0 2 4 6 8 10 12 14 16 18
-8
-6
-4
-2
0
2
4
6
8
S
t
r
e
s
s
t
(
M
P
a
)
Time(s)
Uniform
Non-uniform

(a) Displacement at the top of west tower (b) Stress in stab cable No. 228S
Figure 5. Comparison of seismic responses under uniform and non-uniform ground
motions
5th International Conference on
Structural Health Monitoring of Intelligent Infrastructure (SHMII-5) 2011
11-15 December 2011, Cancún, México


- 8 -
4 BUFFETING ANALYSIS UNDER NONSTATIONARY TYPHOON
Long-span bridges are often vulnerable to excessive buffeting responses that may lead
to serious problems of fatigue, comfortableness, and vehicle safety. Although most of
previous buffeting studies considered stationary wind field, it is noted that a typhoon
usually possesses time-varying mean wind and fluctuating wind, especially when it
passes over a complex terrain. This task performed a fully-coupled buffeting response
analysis of Stonecutters Bridge under nonstationary typhoon wind. In October 2002, a
field measurement system for wind turbulence parameters (FMS-WTP) consisting of a
50m high guyed mast and a sensing system was installed at the bridge site. The one
hour nonstationary wind time history was measured during typhoon Dujuan. By
decomposing the wind speed into mean wind and fluctuating wind, the nonstationarity
of the typhoon can be described by the time-varying mean wind speed and the
evolutionary power spectra density (EPSD) for fluctuating wind speed. Figure 6 shows
the nonstationary properties of longitudinal wind at the measurement point. Then,
nonstationary typhoon can be derived for the whole wind field based on the results at
the measurement point. The cross EPSD of the fluctuating wind was obtained according
to the coherence functions with decay factor. Using the typhoon wind field simulation
procedure for a site surrounded by the complex terrain (Xu and Huang, 2011), the
normalized mean wind speed profile at the bridge site corresponding with the direction
of this one hour nonstationary wind time history can be obtained. Consequently, the
time-varying mean wind speed vector of longitudinal wind was obtained for the whole
typhoon wind field based on the measurement results. Next, aeroelastic forces and
buffeting forces acting on the bridge were computed. In particular, the buffeting forces
were also nonstationary due to the time-varying features of mean wind speed and
evolutionary PSD.
Based on the FEM and the simulated nonstationary typhoon load, a fully-coupled
buffeting response analysis was carried out to study the behavior and safety of
Stonecutters Bridge under a nonstationary typhoon. The pseudo-excitation method, a
frequency-domain method, was used in the dynamic analysis. Figure 7 shows the RMS
torsional displacement at the middle point of the main span. Time-varying feature of the
buffeting-induced response could be clearly observed in this one hour. Similar
observations can also be made to the dynamic response at other locations, including
vertical displacement and lateral displacement. Furthermore, a comparative study
indicated that the EPSD has larger effect on the nonstationarity of buffeting response
than the time-varying mean wind.
0 400 800 1200 1600 2000 2400 2800 3200 3600
8
12
16
20
24
Time (s)
W
i
n
d

s
p
e
e
d

(
m
/
s
)

(a) Time-varying mean wind speed (b) Evolutionary power spectrum density
Figure 6. Time-varying characteristics of non-stationary typhoon (longitudinal wind)
5th International Conference on
Structural Health Monitoring of Intelligent Infrastructure (SHMII-5) 2011
11-15 December 2011, Cancún, México


- 9 -
0 400 800 1200 1600 2000 2400 2800 3200 3600
0.004
0.005
0.006
0.007

R
M
S

v
a
l
u
e

(
m
)
Time (s)
Torsional displacement

Figure 7. Time-varying buffeting-induced torsional displacement at the middle point of
main span (Solid line: time-varying value, dashed line: mean value)
5 CONCLUSIONS
This paper presents the work on the safety evaluation of Stonecutters Bridge under
extreme events based on the installed SHMS comprising 1571 sensors of different
types. A SHM-oriented FEM was first established and updated according to the field
measurement data including cable forces, bridge deck profile and natural frequency. A
good agreement between the computed results and the measured data demonstrates that
the 3D FEM is a good numerical representation of Stonecutters Bridge. The
sophisticated FEM substantially enhances the ability of the existing SHMS to conduct
bridge health diagnosis and prognosis in the sense that: (1) it provides a relationship
between unobserved responses/members and the field measurement data; (2) it provides
an opportunity to assess bridge safety under various loading conditions including
extreme events such as earthquakes, typhoons and so on.
The FEM was used to predict the seismic behavior of Stonecutters Bridge when
subjected to spatially varying ground motions. Using a conditional simulation method,
non-uniform ground motions at the multiple supports were simulated based on the
measured ground acceleration by a seismometer. The wave passage and incoherence
effect was considered in the simulation. Seismic responses of Stonecutters Bridge under
uniform and non-uniform ground motions showed apparent differences, and some
comparison results demonstrated that neglecting spatial variability in ground motions
may underestimate seismic demand at some locations of Stonecutters Bridge.
A 3D FEM-based fully-coupled buffeting analysis of Stonecutters Bridge was carried
out under typhoon load. In particular, nonstationary characteristics of typhoon,
including time-varying mean wind speed and EPSD of fluctuating wind speed, were
appropriately considered in the analysis. The nonstationary wind field was simulated
based on measurement data during Typhoon Dujuan. The wind forces acting on the
bridge were computed accordingly. The RMS responses of Stonecutters Bridge at all
key locations indicate a clear time-varying feature. In this case study, the EPSD
contributed more to the nonstationarity of the buffeting response than the time-varying
wind speed. These results indicated that the nonstationary characteristics in wind load
and bridge response cannot be overlooked in the buffeting analysis when a typhoon is
considered.
6 ACKNOWLEDGEMENTS
The work described in this paper was financially supported by The Hong Kong
Polytechnic University through its niche area program on performance-based structural
health monitoring of large civil engineering structures, and the Hong Kong Highways
Department through a contract research on the development of structural health
5th International Conference on
Structural Health Monitoring of Intelligent Infrastructure (SHMII-5) 2011
11-15 December 2011, Cancún, México


- 10 -
prognosis tools for Stonecutters Bridge under in-service conditions. All views expressed
in this paper are entirely those of the authors.
7 REFERENCES
Abdelghaffar, AM. 1982. Suspension bridge vibration: continuum formulation. Journal
of the Engineering Mechanics Division, ASCE, 108(6): 1215-1232.
Abdelghaffar, AM and Rubin, LI. 1983. Lateral earthquake response of suspension
bridges. Journal of Structural Engineering, ASCE, 109(3): 664-675.
Heredia-Zavoni, E. 1993. On the Response of Multi-support MDOF Systems to
Spatially Varying Earthquake Ground Motion, Ph.D. Dissertation, Department of
Civil Engineering and Operations Research, Princeton University.
Harichandran, R and Vanmarcke, E. 1986. Stochastic variation of earthquake ground
motion in space and time. Journal of Engineering Mechanics, 112(2): 154-174.
Vanmarcke, E. 1983. Random Fields: Analysis and Synthesis. Cambridge: The MIT
Press.
Wong, KY. 2010. Structural health monitoring and safety evaluation of Stonecutters
Bridge under the in-service condition. Bridge Maintenance, Safety, Management and
Life-cycle Optimization, Frangopol, Sause & Kusko (eds), London: Taylor & Francis
Group.
Xu, YL, Huang, WF, and Liu, HJ 2011. Prediction of typhoon wind speed and profile
over complex terrain, Journal of Wind Engineering and Industrial Aerodynamics (to
be submitted)
Zheng, Y, Xu, YL and Hu, L. 2011. Seismic response analysis of long-span cable-
stayed bridges subjected to multi-support excitation. Proc., The Fifth International
Symposium on Environmental Vibration, Chengdu, China, October 20-22.