Journal of Marine Science and Technology, Vol. 18, No. 3, pp.

395-404 (2010) 395

THE INFLUENCE OF BROKEN CABLES ON THE
STRUCTURAL BEHAVIOR OF LONG-SPAN
CABLE-STAYED BRIDGES


Chin-Sheng Kao* and Chang-Huan Kou**


Key words: cable-stayed bridge, broken cable, structural behavior,
ultimate load-bearing capacity.
ABSTRACT
The influence of broken cables on the structural behavior of
long-span cable-stayed bridges is examined considering all the
nonlinear characteristics of a long-span cable-stayed bridge.
The results show the influences of the broken cable on the
structural behavior, such as the internal force, displacement,
and the ultimate load-bearing capacity, are clear when an
outside cable breaks. The results of the present study should
assist in deciding when to replace cables in long-span cable-
stayed bridges.
I. INTRODUCTION
In the last half-century, many cable-stayed bridges have
been built or are under construction all over the world. The
rapid progress of this type of bridge is mainly due to the de-
velopment of computer technology, high-strength materials
and construction technology [1, 2, 5, 9]. Because of its aes-
thetic appeal, economics and ease of erection, the cable-stayed
bridge is considered to be most suitable for medium- to
long-span bridges with spans ranging from 200 to about 1000
m [2, 5].
The cables sustaining the cable-stayed bridge may break
due to a collision resulting from a car accident, lack of main-
tenance over a long period of time, or excessive corrosion of
the connection. It is also possible similar behavior may occur
due to loosening a cable before replacing it. Therefore, it is
worth studying whether a slender long-span cable-stayed
bridge would cause any safety problems for these reasons;
especially when there have been real cases of structural
damage resulting from cables breaking due to corrosion [14].
Further study of this issue has therefore been undertaken.
In fact, the longer the span length, the more flexible is the
bridge structural system. Because of their huge size and
complicated nonlinear structural behaviors, the analysis of
cable-stayed bridges is much more complicated than conven-
tional bridges, such as truss and girder bridges. The sources of
nonlinearity in cable-stayed bridges mainly include: the cable
sag, beamcolumn and large deflection effects. Many studies
on this type of bridge have been carried out in the last
half-century. Tang suggested using the iteration method to
analyze the nonlinear effects on a cable-stayed bridge [12].
Lazar et al used slanted prestressed cables to resist the external
load, and perform stress analysis with the balanced load
method [8]. Tang used a methodology proposed by Ernst [3],
based on the equivalent Youngs modulus (Eeq) of cables
and on considering the beamcolumn effect, to carry out a
static analysis for a cable-stayed bridge [13]. Podolny and
Scalzi proposed a stiffness method as the basis for the static
analysis of cable-stayed bridges. They found most nonlinear
behaviors of cable-stayed bridges were caused by the sagging
effect in the cable, and this nonlinear effect must be considered
and incorporated into the initial dead load analysis [11].
Fleming derived a stable function under the influence of the
beam element to modify the axial stiffness, the lateral stiffness
and the bending stiffness, to establish a structural analysis
model of cable-stayed bridges using the finite element analysis
concept [4]. Hegab analyzed the structure of a three- dimen-
sional double-cable plane cable-stayed bridge using the energy
method with an incremental iteration approach, and also con-
sidering the torsion effect [6, 7]. Nazmy and Abdel-Ghaffar
conducted a nonlinear static analysis of a three-dimensional
cable-stayed bridge by applying the finite element approach.
They also performed the analysis using incremental iteration
of the loading and the tangent stiffness matrix of the structure,
considering all the potential nonlinear behaviors. Many stud-
ies of this type of bridge have been carried out, but only a few
papers have analyzed the effect of broken cables on cable-
stayed bridges [10].
In the present study, the following approaches or method-
ologies were employed: the updated Lagrangian formulation
was used to treat for the nonlinear large displacement effect
produced by the entire cable-stayed bridge; the curved cable
element was considered for the sagging cable in the cable-
Paper submitted 05/01/09; revised 06/15/09; accepted 06/23/09. Author for
correspondence: Chin-Sheng Kao (e-mail: csk@mail.tku.edu.tw).
*Department of Civil Engineering, Tamkang University, Taiwan, R.O.C.
**Department of Civil Engineering & Engineering lnfomatica, Chung-Hua
University, Taiwan, R.O.C.
396 Journal of Marine Science and Technology, Vol. 18, No. 3 (2010)

stayed bridge; the enclosed box girder element was investi-
gated by considering the coupling of the axial force with the
bending moment for directly assessing the beamcolumn
effect of the main girder and the tower; the PrandtlReuss
incremental theory and the Von Mises yielding criterion were
used to deal with the material nonlinear effects before ultimate
failure of the cable-stayed bridge was reached; and lastly,
displacement control was used to solve the equilibrium equa-
tion of the nonlinear system.
II. THEORIES AND METHODS
1. Basic Assumptions
(a) Let the stress in the direction of the slab thickness be zero
and the cross-section be plane and remain plane before
and after yielding.
(b) Localized bucking of steel plates is not considered.
(c) Both the normal stress caused by warping deformation
and the shearing stress caused by bending deformation are
ignored.
(d) The material is an ideal elasto-plastic material.
(e) The PrandtlReuss incremental theory and the Von-Mises
yielding criterion hold in the plastic state.
(f) The shear flow in the elasto-plastic state in the enclosed
cross-section remains constant.
2. Incremental Stiffness Matrix of the BeamColumn
Element
Based on the Von Mises yielding criterion, the equivalent
stress can be expressed as:

2 2
3
y
= + = (1)
where
y
stands for the axial yielding stress.
Based on the PrandtlReuss incremental theory, the com-
position rule of normal stress increment and shear stress
increment of an ideal elasto-plastic element can be ex-
pressed as:

1 3
3 2
L L
z z
ep
D D
D
D D


(
( = =
` ` `
(


) ) )
(2)
In the elastic zone:
[ ] [ ]
1 3
3 2
0
0
e
D D E
D D
D D G

= = =

(3)
and in the plastic zone:

[ ] [ ]
2
1 3
2
3 2
9 / 3 /
3 / /
P
D D EG B EG B
D D
D D EG B EG B



= = =

(4)
O
Y
X
Z
4
3
2
1 o = Internal joint

Fig. 1 Cable element.


Where B = E
2
+ 9G
2
; the linear axial strain increment
L
can be expressed from the updated Lagrangian formula-
tion; the shear strain increment can be calculated by the
thin-wall beam theory; E is Youngs modulus; G is the shear
modulus and the elasto-plastic matrix of the incremental the-
ory can be expressed as:
[ ] [ ] [ ]
ep e p
D D D = (5)
By applying the updated Largrangian formulation and the
principle of virtual work, the equilibrium equation becomes:
[ ]
( )
{ } { }
e
e e e
ep G
K K u P + =

(6)
Where [K
ep
]
e
and [K
G
]
e
are the elasto-plastic stiffness and
geometrical stiffness matrix respectively. They can be ex-
pressed as:
[ ] [ ]
e
T
ep L ep L
V
K B D B dv =

(7)
[ ] [ ] [ ]
0
e T
G
v
K G G dv =

(8)
where [B
L
] and [G] are the relative matrixes between strain
and displacement, and [
0
] is the initial stress matrix.
3. Incremental Stiffness Matrix of the Cable Element
For a cable element, the present study employed a curve
element with m nodes (see Fig. 1). Its incremental displace-
ment and the coordinate vector of the internal element can be
expressed as:

( ) ( ) [ ] [ ]
( ) ( ) [ ] [ ]
1
1
, ,
, ,
T
e
m
T
e
m
u s N s u u N u
X s N s X X N X

= =

= =

(9)
where X(s) and X
e
are the coordinate vectors of the internal
element and the internal node respectively; u
e
is the incre-
ment of the nodal displacement vector, and [N] is a shape
function matrix [11].
After considering curvature, the linear strain increment
L

C.-S. Kao and C.-H. Kou: The Influence of Broken Cables on the Structural Behavior of Long-Span Cable-Stayed Bridges 397

of the flexible cable and its nonlinear strain increment
N
can
be expressed as:

L s t
u u
s R

(10)
2 2 2
1
2
N n t s s t
u u u u u
s s R s R

(
| | | | | |
= + + + (
| | |

\ \ \
(

(11)
Thus the [B
L
] and [G] matrixes are respectively:
[ ] [ ]
[ ]
2
2
1
T
T
L S
N
X
B N e
J s

= +



( ) ( ) ( )
[ ]
2 2 2
2 2 2
2 2 2
T
n
X Y Z
e N
s s s

+ +

(12)
[ ] [ ]
[ ]
* 1 1
1
1
, ,
T
T
e N N
G e N e
J s r

= + +

[ ]
[ ]
*
1
T
T
n n
n
e N N
e N e
J s r

+ +

(13)
where R is the radius of curvature of the cable; J is the Jaco-
bian matrix; N
1
and N
n
are the complementary function of the
Lagrangian; is the natural coordinate of the cable element;
and r is a position vector of the cable.

[ ] [ ]
11 12 13
21 22 23
31 32 33
n t s
e e e
e e e e
e e e


= =



e e e (14)

13 23 33
*
11 21 31
0 0 0
e e e
e
e e e


=



(15)
The following formula can then be used to attain the in-
cremental stiffness matrix of the cable element:
[ ] [ ] [ ]
T
e L c c L
l
K B E A B ds =

(16)
[ ] [ ]
0
T
g
l
K G T G ds =

(17)
where E
C
A
C
is the axial rigidity of the cable, and T
0
is the
initial cable force.
4. Solving the Nonlinear Equation
The NewtonRaphson iteration method and arch-length
method were employed to solve the equilibrium equation of
the system.
(a) Girder cross-section
(b) Shape of tower
(d) Side view of a cable-stayed bridge
(c) Cross-section of tower
4@5.0 m = 20.0 m
0.02 m
0.015 m
0.02 m
3 m
6.0 m
52 m
116 m
64 m
7.0 m
5.0 m
20 m 24 m
34.0 m
52 m
64 m
24 m
40 m
600 m
14@20 m = 280 m
14@20 m = 280 m
14@20 m = 280 m

Fig. 2. Shape and dimension of cable-stayed bridge.

III. STRUCTURAL ANALYSIS
1. Basic Data
The present analysis was performed for a symmetrical,
fan-shaped, double-cable-plane, double-tower three-span
cable-stayed bridge. The bridge has a total span of 1160 m,
including a 600 m center-span and two side-spans of 280 m
each. The A-shape tower situated in the bridge is 140 m high,
with lateral diaphragm beams provided at the very top of the
tower, and at 52 m and 120 m respectively from the top of the
tower. The upper part of the tower, from the top of the tower to
the 52 m diaphragm beams, is the anchoring zone of the cables.
112 cables are attached to the bridge; 14 pairs of cables are
allocated on each side of the two towers. The spacing of the
cable anchors on the tower is 4m; and the cable spacing on the
girder is 20m. The constraint condition of the tower founda-
tion is set as fixed joints, while the joints between the cable
and the girder are hinged. The intersections of the tower and
the girder are simulated with hinge joints. Figure 2 and Table
1 show the analyzed model and the cross-sectional parameters
398 Journal of Marine Science and Technology, Vol. 18, No. 3 (2010)

Table 1. Cross-sectional parameters of cable-stayed bridge.
A (m
2
) I
X
(m
4
) I
Y
(m
4
) J (m
4
)*
Girder 1.025 1.9688 37.917 3.888
Tower 0.960 4.3333 7.1867 6.125
Cable 0.020699 (m
2
) / Single cable
*Neglecting the longitudinal stiffener


C
L
28 15 1 14

Fig. 3. Cable numbering and location.


of the girder, tower and cable. In the simulation, the girder and
tower were made of SM490Y (JIS) steel, and the cable wires
were made of ST1570 (JIS) steel.
The object being considered is a two-tower, symmetrical-
span, cable-stayed bridge. Broken cables on only one tower of
the bridge were analyzed. Figure 3 shows the cable pairs 114
on the side-span and the cable pairs 1528 on the center-span
were considered to break one by one each time for analysis
and comparison.
The static mechanical behavior of the cable-stayed bridge
was investigated, both in the initial state and in the completed
bridge state, and the mechanical behavior before and after the
cable broke was also compared.
2. Analysis Models
In the initial state analysis, the anchor end on the girder of
the stayed-cable was assumed to be a rigid support, as shown
in Fig. 4. The vertical reaction of each rigid support resulting
from dead load W
G
of the superstructure was then calculated
according to the continuous beam theory. This reacting force
was taken as the vertical component force of the initial cable
force. Using this rigid-support theory, the initial cable force in
the center-span can be calculated. The cable force in the side-
span can be derived from the equilibrium condition of its
horizontal components on both sides of the tower.
In the completed bridge state analysis, the 3-D frame in-
corporating a structural model of the curved cable was used.
Besides the dead load W
G
derived by the following equation, a
live load equivalent to the dead load was included to analyze
the load-bearing capacity.
( ) 1.4 70.0
G s s
W A = + (18)
Where 1.4 is the multiplier for considering the weights of other
components, A
s
is the cross-section area of the girder;
s
is the
W
G
R
j, support
Cable anchorage
Girder

Fig. 4. Structural model of continuous beam with rigid supports.


W
G

Fig. 5. Schematic diagram of loading.


unit weight of the steel, and 70.0 (kN/m) is the extra dead load
including the slab, pavement, rails, curbs, etc.
The ultimate strength analysis used the same structural
model as that used for the completed bridge state analysis. A
live load, which was ( 1) times the dead load was used (see
Fig. 5). Here, the value of is gradually increased from zero
during the analysis, using the incremental displacement ap-
proach or arc-length approach for the nonlinear convergence
calculation. The axial displacement of the girder end was used
as the increment parameter for making the judgment.
3. Results of Analysis
(a) The cable forces increased significantly in the cables near
the 1st and 15th pairs after the 1st pair of cables in the
completed bridge state had broken (see Fig. 6). A slight
decrease in the axial force in the left-hand side girder and
a slight increase of axial force in the right-hand side girder
were observed (see Fig. 7). The bending moment in the
left-hand side-span was significantly diminished, be-
coming a negative moment, while the bending moment of
the center-span girder was significantly increased (see Fig.
8). There was a slight decrease in the axial force in the
tower (see Fig. 9). The bending moment in the tower in-
creased significantly (see Fig. 10). The side-span girder
was appreciably lifted (see Fig. 11). A significant increase
in tower deformation was observed (see Fig. 12).
(b) In the ultimate state, the plastic hinge of the girder occurs
within the range of joints where the 6th and 10th pairs of
cables are anchored and where the 14th and 28th pairs of
cables are anchored (see Fig. 13).
(c) In the completed bridge state, the sag at the center of the
girder reached its maximum value of 62.43% when the 1st
pair of the outermost cables broke (see Table 2).
(d) The uplift deformation of the girder at different positions
may be attributed to the redistribution of cable forces re-
sulting from the cables breaking at different positions, as
shown in Table 3.
C.-S. Kao and C.-H. Kou: The Influence of Broken Cables on the Structural Behavior of Long-Span Cable-Stayed Bridges 399

2500
2000
1500
1000
500
0
C
a
b
l
e

f
o
r
c
e

(
t
)
0 580 1160
Location in direction of bridge axis (m)
First pair of broken cable
Before cable breakage

Fig. 6. Distribution of cable force in completed bridge state.


2500
2000
1500
1000
500
0
-5000
0 580 1160
Location in direction of bridge axis (m)
First pair of broken cable
Before cable breakage
A
x
i
a
l

f
o
r
c
e

(
t
)

Fig. 7. Distribution of girder axial force in completed bridge state.


0 580 1160
Location in direction of bridge axis (m)
First pair of broken cable
Before cable breakage
30000
20000
10000
0
-10000
-20000
M
o
m
e
n
t

(
t
-
m
)

Fig. 8. Distribution of girder bending moment in completed bridge state.


First pair of broken cable
Before cable breakage
A
x
i
a
l

f
o
r
c
e

(
t
)
16000
14000
12000
1000
8000
6000
4000
2000
0
0 20 40 60 80 100 120 140
Tower lever (m)

Fig. 9. Distribution of tower axial force in completed bridge state.


(e) There was a 28% increase in deformation occurring in the
tower at the far end of the bridge axis when the 3rd pair of
cables broke (see Table 4). In Table 4, the - sign denotes
tower displacement towards the side span.
(f) There were 1040% increases in cable force for cables
located near a broken cable (see Tables 5 and 6).
60000
50000
40000
30000
20000
10000
0
M
o
m
e
n
t

(
t
-
m
)
0 20 40 60 80 100 120 140
Tower level (m)
First pair of broken cable
Before cable breakage

Fig. 10. Distribution of tower bending moment in completed bridge state.


0 145 290 435 580 725 870 1015 1160
Breakage of the first pair of cable
Before cable breakage
Location in direction of bridge axis (m)
2
0
-2
-4
-6
V
e
r
t
i
c
a
l

d
i
s
p
l
a
c
e
m
e
n
t

(
m
)

Fig. 11. Diagram of Girder Deformation.


0 0.5 1 1.5 2 2.5
Breakage of the first pair of cable
Before cable breakage
Location in direction of bridge axis (m)
H
e
i
g
h
t

o
f

M
a
t
i
o
n

T
o
w
e
r

(
m
)
160
120
80
40
0

Fig. 12. Diagram of Tower Deformation.


C
L
Where a plastic hinge occurs
1 2 3 4 5 6 7 8 10 9 11 12 13 14 28 27 26 25 24 23 22 21 19 20 18 17 16 15

Fig. 13. Location of ultimate plastic hinge of cable-stayed bridge.


(g) There was roughly a 58% increase in the total girder stress
when the 15th pair of cables broke (see Table 7). This
stress is calculated by [ ]*
x z
xx
M y P
A I
= + .
400 Journal of Marine Science and Technology, Vol. 18, No. 3 (2010)

Table 2. Deviation of main girder central sag resulting from broken cable at different locations.
Side span Center span
No. of broken
cable
Displacement, center of
girder of center span(m)
Deviation of the
displacement (%)
No. of broken
cable
Displacement, center of
girder of center span(m)
Deviation of the
displacement (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
5.8266
3.9085
3.8909
3.7270
3.6296
3.5742
3.5516
3.5481
3.5592
3.5722
3.5803
3.5858
3.5931
3.5869
62.43
8.96
8.46
3.90
1.18
-0.36
-0.99
-1.09
-0.78
-0.42
-0.19
-0.04
0.16
-0.01
15
16
17
18
19
20
21
22
23
24
25
26
27
28
5.1222
3.8991
3.9523
3.8004
3.6998
3.6227
3.5770
3.5516
3.5521
3.5594
3.5686
3.5757
3.5860
3.5858
42.79
8.69
10.18
5.94
3.14
0.99
-0.28
-0.99
-0.98
-0.78
-0.54
-0.32
-0.03
-0.04

Displacement, center of girder of
Center-span with no broken cable(m)
3.5872


Table 3. Uplifted deformation of girder resulted from broken cable at different locations.
No. of broken
cable
Joint number of girder with
increased deformation
No. of broken
cable
Joint number of girder with
increased deformation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
14
214
614
814
1014
1, 1014, 1525
1, 13, 14, 1527
1, 14, 1528
1, 1528
1, 2, 1528
15, 1528
27, 1520, 2228
28, 2428
29, 15, 16, 2428
15
16
17
18
19
20
21
22
23
24
25
26
27
28
1928
2028
2228
2328
24, 2528
111 ,2628
112, 27, 28
112, 15, 28
114, 1517
214, 1518
214, 1519
1014, 1521
1014, 1522
1014, 1523


Table 4. Deviation of top-of-tower displacement resulted from broken cable at different locations.
Side span Center span
No.
of broken cable
Top-of-tower
Displacement
(after breakage, m)
Influence of
top-of-tower
displacement (%)
No.
of broken cable
Top-of-tower
Displacement
(after breakage, m)
Influence of
top-of-tower
displacement (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
2.4551
1.0580
1.0423
0.9240
0.8552
0.8156
0.8004
0.8005
0.8090
0.8186
0.8250
0.8275
0.8272
0.8263
196.94
27.96
26.06
11.75
3.43
-1.35
-3.20
-3.18
-2.16
-0.99
-0.22
0.08
0.05
-0.06
15
16
17
18
19
20
21
22
23
24
25
26
27
28
0.3949
0.7586
0.7826
0.8225
0.8414
0.8463
0.8431
0.8352
0.8287
0.8242
0.8228
0.8244
0.8283
0.8291
-52.24
-8.25
-5.35
-0.52
1.77
2.36
1.97
1.01
0.23
-0.32
-0.49
-0.30
0.18
0.27

Top-of-tower displacement (m) with no
broken cable
0.8268
C.-S. Kao and C.-H. Kou: The Influence of Broken Cables on the Structural Behavior of Long-Span Cable-Stayed Bridges 401

Table 5. Deviation of cable force (%) resulted from broken cable at different locations of side span.
No. of broken cable (side span)
Span
No. of vertical
cable
Cable force before
breakage (tf)
1 2 3 4 5 6 7 8 9 10 11 12 13 14
1 1750 16.9 15.4 6.3 1.0 -2.1 -3.3 -3.3 -2.7 -1.9 -1.0 -0.4 -0.1 -0.2
2 723.5 36.3 19.6 12.1 6.7 2.4 -0.4 -1.8 -2.3 -2.0 -1.4 -0.7 -0.3 -0.2
3 1047 31.1 13.5 17.7 13.0 7.6 3.2 0.1 -1.5 -2.1 -1.7 -1.1 -0.5 -0.3
4 953.6 27.1 10.0 21.0 19.2 13.9 8.0 3.2 -0.0 -1.7 -1.9 -1.3 -0.6 -0.3
5 934.2 16.1 6.4 17.0 21.0 20.4 14.2 7.7 2.7 -0.5 -1.7 -1.5 -0.8 -0.4
6 890.2 1.2 3.2 11.9 17.2 22.8 21.1 14.0 7.3 2.1 -0.7 -1.4 -1.0 -0.5
7 841.7 -8.4 0.7 6.7 11.9 18.4 22.0 21.7 14.2 6.8 1.6 -0.7 -1.0 -0.5
8 787.4 -13.0 -1.0 2.2 6.3 12.2 19.3 21.5 21.5 14.4 6.3 1.2 -0.5 -0.5
9 728 -13.5 -1.8 -1.1 1.5 5.7 12.3 20.2 21.6 21.3 14.2 5.4 1.0 -0.3
10 662.8 -11.2 -2.0 -3.0 -2.0 0.3 5.0 12.3 21.5 22.3 21.9 13.0 4.4 0.5
11 587.7 -7.6 -1.6 -3.5 -3.9 -3.4 -0.9 4.5 12.8 23.7 24.8 24.2 11.0 2.2
12 496.3 -3.9 -1.0 -3.2 -4.4 -5.3 -4.7 -1.8 4.3 14.2 27.4 31.8 22.0 5.6
13 381 -1.1 -0.5 -2.3 -3.8 -5.4 -6.2 -5.4 -2.1 5.3 17.4 31.6 40.3 11.7
Side
Span
14 250.5 0.4 -0.1 -1.3 -2.4 -3.7 -4.9 -5.4 -4.3 -0.4 7.7 19.3 30.0 33.5
Tower
28 256.2 -0.2 0.0 0.3 0.6 1.0 1.4 1.7 1.5 0.2 -2.6 -7.0 -11.4 -13.5 -8.0
27 381.8 -1.7 -0.1 -0.0 -0.1 0.1 0.2 0.4 0.3 -0.5 -2.1 -4.6 -7.1 -8.2 -5.0
26 491.3 -1.8 -0.0 -0.2 -0.4 -0.6 -0.6 -0.5 -0.5 -0.9 -1.6 -2.6 -3.5 -3.9 -2.5
25 579.5 -0.8 0.1 -0.1 -0.5 -0.8 -0.9 -0.9 -1.0 -1.1 -1.3 -1.5 -1.6 -1.6 -1.0
24 654.5 0.9 0.4 0.1 -0.4 -0.9 -1.1 -1.2 -1.2 -1.2 -1.1 -1.0 -0.7 -0.5 -0.4
23 723.2 2.7 0.7 0.4 -0.3 -0.9 -1.2 -1.3 -1.3 -1.2 -1.1 -0.8 -0.4 -0.1 -0.2
22 789 4.4 0.9 0.6 -0.2 -0.9 -1.2 -1.4 -1.3 -1.3 -1.1 -0.7 -0.3 0.0 -0.1
21 852.5 5.2 1.1 0.7 -0.2 -0.9 -1.3 -1.4 -1.4 -1.3 -1.1 -0.7 -0.3 -0.0 -0.1
20 911.4 4.6 1.0 0.6 -0.3 -0.9 -1.3 -1.4 -1.3 -1.3 -1.1 -0.8 -0.3 -0.1 -0.1
19 965.5 2.1 0.6 0.3 -0.4 -1.0 -1.2 -1.3 -1.3 -1.2 -1.1 -0.8 -0.4 -0.1 -0.2
18 991.3 -2.9 -0.2 -0.5 -0.8 -1.1 -1.2 -1.2 -1.2 -1.1 -1.0 -0.8 -0.4 -0.2 -0.2
17 1087 -10.8 -1.3 -1.5 -1.2 -1.1 -1.1 -1.0 -0.9 -1.0 -0.9 -0.7 -0.4 -0.2 -0.2
16 743.9 -22.3 -3.0 -3.1 -1.9 -1.3 -1.0 -0.8 -0.8 -0.8 -0.9 -0.7 -0.4 -0.2 -0.2
Center
Span
15 1757 -37.4 -5.3 -5.1 -2.8 -1.5 -0.8 -0.5 -0.5 -0.6 -0.7 -0.6 -0.4 -0.2 -0.2


Table 6. Deviation of cable force (%) resulted from broken cable at different locations of center span.
No. of broken cable (side span)
Span
No. of vertical
cable
Cable force before
breakage (tf)
28 27 26 25 24 23 22 21 20 19 18 17 16 15
1 1750 -0.3 -0.2 -2.0 -1.5 -1.2 -1.1 -1.1 0.9 0.2 -0.9 -2.5 -5.2 -5.5 -31.1
2 723.5 -0.3 -0.2 -2.1 -1.6 -1.3 -1.2 -1.4 0.5 -0.1 -1.1 -2.3 -4.3 -4.2 -23.2
3 1047 -0.4 -0.3 -2.1 -1.6 -1.4 -1.4 -1.7 0.2 -0.5 -1.2 -2.2 -3.4 -3.0 -15.6
4 953.6 -0.3 -0.2 -2.1 -1.7 -1.4 -1.5 -1.9 0.0 -0.7 -1.4 -2.0 -2.7 -1.9 -9.0
5 934.2 -0.3 -0.1 -2.1 -1.7 -1.5 -1.6 -2.0 -0.2 -0.9 -1.5 -1.9 -2.1 -1.0 -4.1
6 890.2 -0.3 -0.1 -2.1 -1.7 -1.5 -1.7 -2.1 -0.3 -1.0 -1.6 -1.7 -1.7 -0.5 -0.8
7 841.7 -0.2 -0.0 -2.0 -1.6 -1.5 -1.7 -2.2 -0.4 -1.1 -1.6 -1.7 -1.4 -0.2 1.0
8 787.4 -0.2 -0.0 -2.0 -1.6 -1.5 -1.7 -2.2 -0.4 -1.1 -1.6 -1.7 -1.4 -0.1 1.5
9 728 -0.3 -0.1 -2.1 -1.7 -1.5 -1.7 -2.2 -0.3 -1.0 -1.5 -1.6 -1.4 -0.2 1.1
10 662.8 -0.6 -0.5 -2.4 -1.9 -1.6 -1.7 -2.1 -0.2 -0.9 -1.5 -1.6 -1.5 -0.3 0.2
11 587.7 -1.2 -1.6 -3.3 -2.3 -1.7 -1.6 -1.9 -0.0 -0.7 -1.3 -1.6 -1.6 -0.5 -0.8
12 496.3 -2.6 -3.9 -5.0 -3.2 -2.0 -1.4 -1.4 0.4 -0.4 -1.0 -1.4 -1.5 -0.6 -1.5
13 381 -5.3 -8.3 -8.1 -4.9 -2.3 -0.9 -0.5 1.2 0.4 -0.4 -0.9 -1.1 -0.5 -1.4
Side
Span
14 250.5 -8.5 -13.9 -11.8 -6.7 -2.6 0.1 1.2 2.1 1.5 0.8 0.2 -0.1 -0.1 0.2
Tower
28 256.2 32.8 27.8 16.6 6.7 -0.7 -4.4 -5.1 -4.7 -3.7 -2.5 -1.5 -0.2 0.0
27 381.8 11.9 37.6 27.8 15.6 4.5 -2.6 -4.7 -5.9 -5.7 -4.5 -3.5 -1.2 -2.8
26 491.3 5.7 22.3 32.9 25.2 13.1 3.4 -0.7 -4.3 -5.7 -5.4 -4.8 -1.9 -5.6
25 579.5 2.1 11.2 22.5 26.3 22.3 11.6 6.1 -0.4 -3.9 -5.1 -5.3 -2.6 -8.1
Center
Span
24 654.5 0.4 4.4 11.0 22.5 23.6 20.0 14.4 5.6 -0.2 -3.2 -4.8 -2.8 -9.6
402 Journal of Marine Science and Technology, Vol. 18, No. 3 (2010)

Table 6. (Continued)
No. of broken cable (side span)
Span
No. of vertical
cable
Cable force before
breakage (tf)
28 27 26 25 24 23 22 21 20 19 18 17 16 15
23 723.2 -0.4 1.0 3.5 12.0 22.0 22.1 22.0 12.7 5.2 0.1 -2.8 -2.4 -9.2
22 789 -0.6 -0.5 -0.5 4.7 12.8 21.0 21.5 19.5 11.4 4.8 0.6 -1.2 -6.1
21 852.5 -0.7 -1.0 -2.4 0.5 5.7 12.6 19.4 20.4 17.3 10.1 5.1 0.8 0.1
20 911.4 -0.6 -0.9 -3.1 -1.7 1.3 6.0 11.9 20.3 20.1 15.2 10.4 3.5 9.4
19 965.5 -0.5 -0.8 -3.2 -2.5 -1.1 1.7 5.9 14.8 19.4 18.9 15.6 6.8 20.5
18 991.3 -0.4 -0.6 -3.0 -2.7 -2.2 -0.9 1.6 8.5 13.2 17.7 19.5 10.2 22.5
17 1087 -0.4 -0.4 -2.7 -2.6 -2.5 -2.0 -0.9 4.1 7.7 12.3 16.6 13.2 26.0
16 743.9 -0.3 -0.3 -2.4 -2.3 -2.5 -2.6 -2.4 1.1 3.5 7.4 12.2 18.9 31.4
Center
Span
15 1757 -0.3 -0.2 -2.1 -2.0 -2.3 -2.7 -3.1 -0.8 0.6 3.5 8.1 15.5 14.6


Table 7. Deviation of total girder stress resulted from broken cable at different locations.
Side span Center span
No. of broken
cable
Total stress in girder
[] (tf/m
2
)
Deviation of
total girder stress (%)
No. of broken
cable
Total stress in girder
[] (tf/m
2
)
Deviation of
total girder stress (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
37555
30445
33022
32859
32180
30937
29824
29826
29816
30494
31843
33256
33980
32459
33.56
8.28
17.44
16.86
14.45
10.03
6.07
6.07
6.04
8.45
13.25
18.27
20.85
15.44
15
16
17
18
19
20
21
22
23
24
25
26
27
28
44492
30409
32640
30625
29687
29838
29983
29598
29690
30332
31437
32784
34942
32482
58.23
8.15
16.08
8.91
5.58
6.12
6.63
5.27
5.59
7.87
11.80
16.59
24.27
15.52

Total stress of girder with no broken
Cable []* (tf/m
2
)
28118


Table 8. Deviation of total tower stress resulting from broken cable at different locations.
Side span Center span
No. of broken
cable
Total stress in tower
[] (tf/m
2
)
Deviation of
total tower stress (%)
No. of broken
cable
Total stress in tower
[] (tf/m
2
)
Deviation of
total tower stress (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
41915
28789
28465
27136
26293
25780
25531
25477
25507
25620
25804
26019
26149
26199
60.03
9.92
8.68
3.61
0.39
-1.57
-2.52
-2.73
-2.61
-2.18
-1.48
-0.66
-0.16
0.03
15
16
17
18
19
20
21
22
23
24
25
26
27
28
22420
25295
24989
25335
25625
25945
26252
25647
25789
25810
25703
25492
26097
26081
14.40
3.42
4.59
3.27
2.16
0.94
0.23
-2.08
-1.54
-1.46
-1.86
-2.67
-0.36
-0.42
Total stress of tower with no broken
Cable []* (tf/m
2
)
26191

C.-S. Kao and C.-H. Kou: The Influence of Broken Cables on the Structural Behavior of Long-Span Cable-Stayed Bridges 403

Table 9. Deviation of ultimate load-bearing capacity resulted from broken cable at different locations.
Side-span Center-span
No. of broken
cable
Coefficient of ultimate
load-bearing capacity
(with a pair of broken cables)
Decrease of ultimate
load-bearing capacity (%)
No. of broken
cable
Coefficient of ultimate
load-bearing capacity
(with a pair of broken cables)
Decrease of ultimate
load-bearing capacity (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
2.374
2.639
2.596
2.429
2.350
2.304
2.293
2.296
2.296
2.277
2.299
2.334
2.438
2.601
11.00
1.03
2.65
8.91
11.89
13.60
14.04
13.90
13.92
14.61
13.82
12.47
8.58
2.46
15
16
17
18
19
20
21
22
23
24
25
26
27
28
2.404
2.628
2.506
2.473
2.400
2.355
2.290
2.307
2.321
2.290
2.329
2.358
2.455
2.624
9.87
1.47
6.03
7.28
10.02
11.70
14.14
13.51
12.99
14.23
12.67
11.58
7.94
1.60

Ultimate load-bearing capacity with
broken cable
2.667


(h) There was an about a 60% increase in total tower stress
when the 1st pair of cables broke (see Table 8). This stress
is calculated by [ ]*
y
x z
xx yy
M x
M y P
A I I
= + .
(i) The ultimate load-bearing capacity of the cable-stayed
bridge decreased by about 15% when the 10th pair of ca-
bles neared the center of the side-span broke (see Table 9).

IV. CONCLUSIONS
Based on the results of the analyses presented above, the
following conclusions can be drawn:

(a) When a cable in a cable-stayed bridge breaks, the adjacent
cables will experience a significant increase in cable
forces. As such, for future cable replacement of the ex-
isting cable-stayed bridges, it is crucial to assess the in-
crement of cable force that may occur in the adjacent
cables in order to prevent yielding failure in the adjacent
cables.
(b) The maximum effects on both the internal force and ul-
timate load-bearing capacity of a cable-stayed bridge
will occur when the outermost cable of the bridge breaks.
In addition, the maximum effects on the ultimate load-
bearing capacity of the bridge may occur when the cable
pairs near the center of the span break. Therefore, when
replacing these cables one must assess the effects on the
internal force and ultimate load-bearing capacity of the
bridge to ensure its safety.
(c) When the outermost cable of a cable-stayed bridge breaks,
the tower may undergo a significant horizontal displace-
ment, and the center of the girder may experience sig-
nificant sagging. It is therefore required that a thorough
assessment of the increased displacement be made in ad-
vance when replacing the outermost cables, together with
proper adjustment of the cable forces in the adjacent ca-
bles in order to minimize the increment of displacements
for ensuring the bridge safety.
(d) Varying uplifts of the girder of the cable-stayed bridge
may occur because of the cable breakage occurring at
different locations. This unique characteristic can be used
to determine which pair of cables in an existing cable-
stayed bridge have broken and caused an accompanying
uplift of the girder.
REFERENCES
1. Billington, D. and Hazmy, A., History and aesthetics of cable-stayed
bridges, Journal of the Structural Division, ASCE, Vol. 117, pp. 3101-
3134 (1990).
2. Chang, F. K. and Cohen, E., Long-span bridges: state-of-the-art,
Journal of the Structural Division, ASCE, Vol. 107, pp. 1145-1160 (1981).
3. Ernst, H. J., Der e-modul von seilen unter beruecksichtigung des
durchhanges, Der Bauingenieur, Vol. 40, No. 2, pp. 52-55 (1965).
4. Fleming, J. F., Nonlinear static analysis of cable-stayed bridges, In-
ternational Journal of Computers & Structures, Vol. 10, pp. 621-635
(1979).
5. Gimsing, N. J., Cable Supported Bridges Concept and Design, second
edition, John Wiley and Sons, New York (1997).
6. Hegab, H. I. A., Energy analysis of doubled-plane cable-stayed bridges,
Journal of the Structural Division, ASCE, Vol. 113, pp. 2176-2188 (1987).
7. Hegab, H. I. A., Parametric investigation of cable-stayed bridges,
Journal of the Structural Division, ASCE, Vol. 114, pp. 1917-1928 (1988).
8. Lazar, B. E., Troitsky, M. S., and Douglass, M. M., Load balancing
404 Journal of Marine Science and Technology, Vol. 18, No. 3 (2010)

analysis of cable-stayed bridges, Journal of the Structural Division,
ASCE, Vol. 98, pp. 1725-1740 (1972).
9. Leonhardt, F. and Zellner, W. Past., Present and future of cable-stayed
bridges, Proceedings of the Seminar on Cable-Stayed Bridge Recent
Developments and Their Future. Yokohama, Japan, pp. 1-33 (1991).
10. Nazmy, A. S. and Abdel-Ghaffar, A. M., Three dimensional nonlinear
static analysis of cable-stayed bridges, International Journal of Com-
puters & Structures, Vol. 34, pp. 257-271 (1990).
11. Podolny, W. J. and Scalzi, J. B., Construction Design of Cable-Stayed
Bridges, John Wiley and Sons (1976).
12. Tang, M. C., Analysis of cable-stayed girder bridges, Journal of the
Structural Division, ASCE, Vol. 97, pp. 1481-1496 (1971).
13. Tang, M. C., Design of cable-stayed girder bridges, Journal of the
Structural Division, ASCE, Vol. 98, pp. 1789-1802 (1972).
14. Wang, W. T., Cable Replacement of Cable-Stayed Bridges, China Com-
munications Press (1997).