Research Article

Advances in Mechanical Engineering

2016, Vol. 8(12) 1–10
Ó The Author(s) 2016
Dynamic analysis of a cable-stayed DOI: 10.1177/1687814016681721
aime.sagepub.com
bridge subjected to a continuous
sequence of moving forces

Mitao Song1, Dengqing Cao2 and Weidong Zhu2,3

Abstract
In this work, an eigenfunction expansion approach is used to study the dynamic response of a cable-stayed bridge excited
by a continuous sequence of identical, equally spaced moving forces. The nonlinear dynamic response of the cable-stayed
bridge is obtained by simultaneously solving nonlinear and linear partial differential equations that govern transverse and
longitudinal vibrations of stay cables and transverse vibrations of segments of the deck beam, respectively, along with
their boundary and matching conditions. Orthogonality conditions of exact mode shapes of the linearized cable-stayed
bridge model are employed to convert the coupled nonlinear partial differential equations of the original nonlinear
model to a set of ordinary differential equations by using the Galerkin method. The dynamic response of the cable-
stayed bridge is numerically solved. Convergence of the dynamic response from the Galerkin method is investigated.
Effects of close natural frequencies, mode localization, the distance between any two neighboring forces, and geometric
nonlinearities of stay cables on the forced dynamic response of the cable-stayed bridge are captured using a convergent
modal truncation.

Keywords
Cable-stayed bridge, continuous sequence of moving forces, Galerkin method, dynamic analysis, convergence

Date received: 25 October 2015; accepted: 26 October 2016

Academic Editor: Nikolaos Nikitas

Introduction focused on VBI problems of cable-stayed bridges since

this type of bridges can exhibit interesting dynamic
Dynamic vehicle–bridge interaction (VBI) problems behaviors, such as energy transfer between different
have attracted much attention of investigators for over vibration modes and the beating phenomenon.5 With
a century.1–4 Researchers focused on the development the advent of high-speed computers, computationally
of models of vehicle–bridge systems and computational efficient numerical techniques have been developed to
methods, which were mainly dependent on the power model and analyze VBI problems of complex cable-
of computers at that time. Most earlier studies associ-
ated with VBI problems focused on very basic and sim- 1
Faculty of Civil Engineering and Mechanics, Jiangsu University, Zhenjiang,
ple cases, where moving force or moving mass models China
were considered with basic beam models.1–4 The basic 2
School of Astronautics, Harbin Institute of Technology, Harbin, China
beam models were described by partial differential 3
Department of Mechanical Engineering, University of Maryland,
equations, and their analytical or closed-form solutions Baltimore, MD, USA
are available.
Corresponding author:
In the past decades, cable-stayed bridges have been Mitao Song, Faculty of Civil Engineering and Mechanics, Jiangsu University,
widely used all over the world due to their economic Zhenjiang 212013, Jiangsu, China.
advantages and aesthetic qualities. Many studies have Email: [email protected]

Creative Commons CC-BY: This article is distributed under the terms of the Creative Commons Attribution 3.0 License
(http://www.creativecommons.org/licenses/by/3.0/) which permits any use, reproduction and distribution of the work without
further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/
open-access-at-sage).
Downloaded from ade.sagepub.com by guest on December 10, 2016
2 Advances in Mechanical Engineering

stayed bridges. The finite element method is one of the knowledge, effects of closed natural frequencies and
most versatile numerical methods used by most mode localization on the vehicle-induced vibration of
researchers. However, the finite element method is the cable-stayed bridge have not been investigated.
time-consuming and inconvenient for parameter studies Dynamic analysis of a beam excited by a sequence of
due to large numbers of elements and amount of input moving mass loads has attracted many researchers21–24
data. Hence, the modal superposition method was used since it has many practical applications such as a train
to study dynamic behaviors of VBI problems formu- traveling on a railroad track.
lated by finite element models, which truncated higher In this work, an eigenfunction expansion approach
vibration modes and considerably reduced numbers of is used to study the dynamic response of a cable-stayed
equations.4,6–9 For instance, Madrazo-Aguirre et al.9 bridge subjected to a continuous sequence of identical,
employed this method to conduct dynamic analysis equally spaced vehicles traveling at a constant speed
of an under-deck cable-stayed bridge with steel- across the bridge deck. For simplicity, the vehicles are
concrete composite decks subjected to a moving load. modeled as moving forces. Geometric nonlinearities of
On the other hand, due to much computational time stay cables are taken into consideration. Nonlinear and
and model complexity, new strategies such as a sub- linear partial differential equations that govern trans-
structure method were developed to model complex verse and longitudinal vibrations of the stay cables and
bridge structures.10–14 transverse vibrations of segments of the deck beam,
Close natural frequencies and mode localization respectively, are derived, along with their boundary
often occur in complex structures that are composed of and matching conditions. The exact natural frequencies
several components. Pierre15 found that mode shapes and mode shapes of the linearized undamped cable-
are strongly localized in nearly periodic structures with stayed bridge model are used to spatially discretize par-
weak coupling. Wei and Pierre16,17 found the same phe- tial differential equations of the original nonlinear
nomenon in shrouded blade assemblies with cyclic sym- cable-stayed bridge model with damping. The dynamic
metry. Natsiavas18 used a perturbation method to response of the cable-stayed bridge is obtained by sol-
investigate the mode localization phenomenon in a 2 ving the resulting nonlinear ordinary differential equa-
degree-of-freedom damped linear oscillator. Further tions (ODEs) via a numerical method. Convergence of
investigations of similar systems were performed in the Galerkin method for the dynamic response of the
Balmes.19 Cao et al.20 studied the free vibration of a cable-stayed bridge subjected to a continuous sequence
cable-stayed bridge that consists of a simply supported of moving forces is investigated. Effects of closed natu-
four-cable-stayed deck beam and two towers for vari- ral frequencies, mode localization, the distance between
ous symmetrical and non-symmetrical bridge cases with any two neighboring forces, and geometric nonlineari-
regard to sizes of components of the bridge and initial ties of stay cables on the forced dynamic response of
sags of cables and found that there are very close natu- the cable-stayed bridge are captured using a convergent
ral frequencies when the bridge model is symmetrical modal truncation.
and/or partially symmetrical, and mode shapes tend to
be more localized when the bridge model is less
Problem formulation
symmetrical.
As mentioned by Wei and Pierre,17 neglecting the As shown in Figure 1, a cable-stayed bridge consists of
localization phenomenon caused by small mistuning a simply supported four-cable-stayed deck beam and
may lead to qualitative errors in the forced response of two towers subjected to a continuous sequence of mov-
a periodic structure. However, to the authors’ best ing concentrated forces. The deck beam consists of

Figure 1. Schematic of a cable-stayed bridge that consists of a simply supported four-cable-stayed deck beam and two rigid towers,
subjected to a continuous sequence of identical, equally spaced moving forces.

Downloaded from ade.sagepub.com by guest on December 10, 2016

Song et al. 3

seven segments separated by its junctions with the stay where L = minfLb1 , Lb2 , . . . , Lb7 g, e = (1=L2b1 )
cables and towers. The following assumptions are made pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
(Eb1 Ib1 )=mb1 , and Dst = (5F1 L3b1 )=(48Eb1 Ib1 ).
in the formulation of the vibration problem of the
Moreover, the following nondimensional parameters
cable-stayed bridge model subjected to a continuous
need to be introduced to furnish a complete definition
sequence of moving forces: (1) the cable-stayed bridge
of elastodynamic properties of the cable-stayed bridge
is modeled as a planar system. (2) Both the cables and
model
the deck beam are considered as a homogeneous one-
dimensional continuum having linear elastic behaviors. Eci Aci Eb Ib L2 Hci Dst
(3) The two towers, to which the stay cables are mci = , hbj = j j , xci = ,k= ð2Þ
Hci Eb1 Ib1 Eb1 Ib1 L
attached, are built on a hard rock foundation and can
be assumed to be rigid;20,25 they are connected to the where Hci is the tension in the stay cable ci on which its
deck beam through roller supports. (4) The concen- initial sag is dependent, that is, Dci = (mci gL2ci
trated forces are equally spaced and have the same cos ui )=(8Hci ), in which g is the acceleration of gravity.
amplitude, moving from the left end of the deck beam Transverse damping per unit length of the stay cable ci
to its right end; the initial displacement and velocity of and segment bj of the deck beam are denoted by Cci
the stay cables and deck beam are all zeros. and Cbj , respectively, and their nondimensional para-
The four stay cables are anchored to the deck beam meters are defined by
at junctions S1, S3, S4, and S6, and the two towers are
connected to the deck beam at junctions S2 and S5. The Cci eL2 Cbj eL4
junctions S1, S2, ., S6 divide the deck beam into seven jci = , j bj = ð3Þ
Hci Ebj Ibj
segments b1, b2, ., b7. The length, mass per unit
length, elastic modulus, and cross-sectional area of the respectively.
stay cable ci are denoted by Lci , mci , Eci , and Aci , respec- The Newtonian method is used to derive both the
tively. The length, mass per unit length, elastic modu- nonlinear equations of motion of the cable-stayed
lus, and area moment of inertia of the segment bj of the bridge model and a full set of boundary and matching
deck beam are denoted by 2Lbj , mbj , Ebj , and Ibj , conditions. Assuming that cable longitudinal inertial
respectively. € ci ) are negligible in the prevalent low-
forces (mci U
Let (Xci , Yci ) be local coordinates of the stay cable ci frequency transverse vibration of the cable-stayed
in the vertical plane, with the origin located at the point bridge, the longitudinal cable displacement Uci can be
C for the stay cables c1 and c2 and at the point D for statically condensed, leading to coupled nonlinear
the stay cables c3 and c4. Let (Xbj , Ybj ) be local coordi- dynamic equations in terms of only the transverse cable
nates of the segment bj of the deck beam in the vertical and deck beam displacements Vci and Vbj , respectively.
plane, with the origin located in the middle of the seg- The equations of motion governing the transverse
ment bj of the deck beam. Under the assumption of a vibration of the cable-stayed bridge model read20
small ratio of the sag Dci to the length Lci , the static
 
equilibrium of the stay cable ci can be described
b2ci €vci + jci v_ ci  v00ci  mci eci (t) v00ci + y00ci = 0, xci 2 ½0, lci 
through a parabolic function Yci (Xci ) = 4Dci
½Xci =Lci  (Xci =Lci )2  in its domain, while the static ð4Þ
deflection of the deck beam is assumed to be negligible.
X
‘
With respect to the above equilibrium configuration, b4bj €vbj + jbj v_ bj + v0000 fbja (t)d(xbj  xbja ), xbj 2 ½lbj , lbj 
bj =
the dynamic configuration of the cable-stayed bridge a=1
model is completely described by longitudinal and ð5Þ
transverse displacements of the stay cables Uci (Xci , t)
and Vci (Xci , t), respectively, and transverse displace- where a prime and dot denote differentiation with
ments of the segments of the deck beam Vbj (Xbj , t). respect to nondimensional local abscissae xci and xbj
The following nondimensional variables are and the time t, respectively; d is Dirac delta function;
introduced eci (t) represents the uniform dynamic elongation
8 given by
> Xci Xb Lc Lb Yc
>
< xci = , xbj = j , lci = i , lbj = j , yci = i
L L L L Dst lc i
ð 
> V bj k k2 1 2
>
: uci =
Uc i
, vc =
Vci
, vb = , dc =
Dc i
, t = et eci (t) = vci (lci , t) tan ui + y0ci v0ci + v9ci dxci
Dst i Dst j Dst i Dst lci lci 2
0
ð1Þ ð6Þ

Downloaded from ade.sagepub.com by guest on December 10, 2016

4 Advances in Mechanical Engineering

 1=2  1=4 the nonlinear forced vibration of the cable-stayed

mci m b j e2
bci = Le , bbj = L , bridge. The equations governing the small amplitude
Hci Ebj Ibj
ð7Þ free vibration of the cable-stayed bridge can be
L3 Fa obtained by linearizing equations (4)–(19) and neglect-
fbja = , xbja = Xbja =L
Ebj Ibj Dst ing the structural damping and external loadings in the
neighborhood of the equilibrium configuration. The
in which
rth undamped natural frequency vr and its correspond-
X
j ing eigenfunction of the linearized cable-stayed bridge
Xbja = VF t  (a  1)LF  2 Lbj + Lbj , j = 1, 2, . . . , 7 model can be numerically solved using the methodol-
j = 1 ogy in Cao et al.20 The rth undamped eigenfunction of
ð8Þ the transverse vibration of the linearized cable-stayed
bridge model can be expressed by
h iT
The functions vci and vbj satisfy the following geo- (r) (r) (r)
f(r) = f(r) (r)
c1    fc4 fb1    fb7 , where f(r)
ci and fbj are
metric boundary conditions for any value of t
the rth undamped eigenfunction of the transverse
A : vb1 (  lb1 ) = 0, v00b1 (  lb1 ) = 0 ð9Þ vibration of the stay cable ci and the segment bj of the
deck beam, respectively.
B : vb7 (lb7 ) = 0, v00b7 (lb7 ) = 0 ð10Þ
C : vc1 (0) = 0, vc2 (0) = 0 ð11Þ Solution method
D : vc3 (0) = 0, vc4 (0) = 0 ð12Þ
The Galerkin method is used to analyze the vibration
The matching conditions at the junction Sk , where of the cable-stayed bridge. The dynamic response of the
k = 1, 3, 4, 6, which involve the stay cable ci , are stay cable ci and segment bj of deck beam are expressed
by
vbk (lbk ) = vbk + 1 (  lbk + 1 ) ð13Þ
X
N
v0bk (lbk ) = v0bk + 1 (  l bk + 1 ) ð14Þ vci (t, xci ) = qr (t)f(r)
ci (xci ) ð20Þ
r=1
hk v00bk (lbk ) = hk + 1 v00bk + 1 (  l bk + 1 ) ð15Þ
X
N

hbk v000  hbk + 1 v000

x c mc
 lbk + 1 ) = i i eci sin ui vbj (t, xbj ) = qr (t)f(r)
bj (xbj ) ð21Þ
bk (lbk ) bk + 1 (
k r=1
+ x ci ½v0ci (lci ) + mci eci (t)(v0ci (lci ) + y0ci (lci )) cos ui
where qr (t) are generalized coordinates. Substituting
ð16Þ equation (20) into equation (4) yields
! ! !
X
N X
N X
N 00
b2ci €qr f(r)
ci + jci q_ r f(r)
ci  qr f(r)
ci
r=1 r=1 r=1
2 3
! lc i
ð ! lci
ð !2
mci k 6 X
N X
N 0 k X
N 0
7
 4tan ui qr f(r)
ci (lci ) + k y0ci qr f(r)
ci dxci + qr f(r)
ci dxci 5 ð22Þ
lci r=1 r=1
2 r=1
0 0
" ! #
X
N 00
3 qr f(r)
ci + y00ci = 0
r=1

The matching conditions at the junction Sk , where Substituting equation (21) into equation (5) and
k = 2, 5, with a roller support are recalling equations (1) and (7)–(8), one has
! ! !
vbk (lbk ) = 0, vbk + 1 (  lbk + 1 ) = 0 ð17Þ X
N X
N X
N 0000
b4bj €qr f(r)
bj + jbj q_ r f(r)
bj + qr f(r)
bj
r=1 r=1 r=1
v0bk (lbk ) = v0bk + 1 (  lbk + 1 ) ð18Þ
X
‘
= fbja d(xbj  xbja ) ð23Þ
hbk v00bk (lbk ) = hbk + 1 v00bk + 1 (  lbk + 1 ) ð19Þ
a=1

Equations (4) and (5) together with the boundary Multiplying equation (22) by x ci f(s)
ci (xci ) and integrat-
and matching conditions in equations (9)–(19) describe ing the resulting equation with respect to xci from 0 to

Downloaded from ade.sagepub.com by guest on December 10, 2016

Song et al. 5

lci , multiplying equation (23) by hbj f(s)bj (xbj ) and inte-

X
4 X
N X
N
NrQ = x ci mci qm qn
grating the resulting equation with respect to xbj from
i=1 m=1 n=1
lbj to lbj , adding all the resulting equations, and using 0 1
lc i lc i lc i
ð ð ð
the following orthogonality relations of eigenfunctions B dci k
2 0 0 00
C
of the linearized cable-stayed bridge model20 @4 3 f(m) (n)
ci fci dxci f(r) e(m)
ci dxci  ^ci f(r) (n)
ci fci dxci A
lci
0 0 0
lci
X
4 ð ð30Þ
x ci b2ci f(r) (s)
ci (xci )fci (xci )dxci
i=1 X
4 X
N X
N X
N
x c mc
0 NrC =  k2 i i
qm qn qo
ð24Þ i=1 m=1 n=1 o=1
2lci
lbj
X
7 ð lci lci ð31Þ
ð ð
+ hbj b4bj f(r) (s)
bj (xbj )fbj (xbj )dxbj = Ms dsr
0 0 00

j=1
f(m) (n)
ci fci dxci f(r) (o)
ci fci dxci
lbj
0 0
2 3
lc i lci
X
4 ð ð lbj
6 00
(s) dci 7 X
‘ X
7 ð
 xci 4 fci (xci )fci (xci )dxci  8mci ^eci 2 f(r)
(r) (s)
ci (xci )dxci 5
i=1
l ci FrMF = hbj fbja f(r)
bj d(xbj  
xbja )dxbj ð32Þ
0 0
a=1 j=1
lbj
lb j
X
7 ð 0000
+ hbj f(r) (s) 2
bj (xbj )fbj (xbj )dxbj = vs Ms dsr respectively. Equation (26) is a set of nonlinear, coupled
j=1
lbj second-order ODEs. If one ignores the geometric nonli-
ð25Þ nearities of the stay cables, then equations (30) and (31)
Ð lci become NrQ = 0 and NrC = 0, respectively, and equation
f(s)
ci (lci )0 (s)
where ^e(s) 2 1
ci = k lci tan ui + k lci 0 yci f9ci dxci , Ms are (26) consequently becomes a set of uncoupled ODEs.
positive constants, and dsr is Kronecker delta, one
obtains spatially discretized equations of the cable-
stayed bridge Numerical results and discussion
Geometric and physical parameters of a cable-stayed
½M€ _ + ½Kq(t) + NQ (t) + NC (t) = FMF (t)
q(t) + ½Cq(t)
bridge are listed in Table 1. The first 20 natural frequen-
ð26Þ cies of the cable-stayed bridge and their corresponding
mode shapes are calculated using the method in Cao
where q(t) = ½q1 (t), q2 (t), . . . , qN (t)T ; ½M, ½C, and ½K
et al.20 and listed in Figure 2. It can be seen that symme-
are N 3 N real symmetric mass, damping, and stiffness
trical and anti-symmetrical mode shapes alternatively
matrices, respectively; and NQ and NC are N-dimen-
occur since the bridge model is symmetrical. An index
sional vectors whose entries are quadratic and cubic
G r is introduced as in Cao et al.20 to measure the degree
nonlinear terms, respectively; and FMF is an N-dimen-
of closeness of the rth natural frequency to its neighbor-
sional modal force vector. Entries of the matrices ½M,
½C, and ½K, and the vectors NQ , NC , and FMF are ing natural frequencies
8 vr + 1  vr
>
Mrs = Ms dsr ð27Þ < Gr =
>
v
3 100% if r = 1
r 
lc i
ð lbj
ð >
> vr  vr1 vr + 1  vr
X
4 X
7 : r G = min , 3 100% if r 6¼ 1
Crs = x ci jci f(s) (r)
+ f(s) (r) vr1 vr
ci fci dxci hb j j b j bj fbj dxbj
i=1
0
j=1
lbj ð33Þ
ð28Þ An index Lri 20,26 is introduced to measure the
degree of localization of the rth mode shape of the stay
Krs = v2s Mrs ð29Þ
cable ci

Ð lc i 2 2
mc i (f(r)
0
(r)
ci (xci ) + uci (xci ))dxi
Lri = 3 100% ð34Þ
P
4 Ð lci (r) 2 (r) 2 P
7 Ð lbj (r) 2
mci 0 (fci (xci ) + uci (xci ))dxci + mbj lb fbj (xbj )dxbj
j
i=1 j=1

Downloaded from ade.sagepub.com by guest on December 10, 2016

6 Advances in Mechanical Engineering

Table 1. Geometric and physical parameters of the cable-stayed bridge.

Parameter Unit Value

Deck beam Mass per unit length of the deck beam, mbj kg/m 16,940
Elastic modulus of the deck beam, Ebj N/m2 2.0 3 1011
Area moment of inertia of the deck beam, Ibj m4 1.20
Length of the segment b1 of the deck beam, 2Lb1 m 35
Length of the segment b2 of the deck beam, 2Lb2 m 40
Length of the segment b3 of the deck beam, 2Lb3 m 50
Length of the segment b4 of the deck beam, 2Lb4 m 50
Length of the segment b5 of the deck beam, 2Lb5 m 50
Length of the segment b6 of the deck beam, 2Lb6 m 40
Length of the segment b7 of the deck beam, 2Lb7 m 35
Stay cables Mass per unit length of the stay cables, mci kg/m 286
Elastic modulus of the stay cables, Eci N/m2 2.0 3 1011
Cross-sectional area of the stay cables, Aci m2 0.0362
Length of the stay cable c1, Lc1 m 52
Length of the stay cable c2, Lc2 m 60
Length of the stay cable c3, Lc3 m 60
Length of the stay cable c4, Lc4 m 52
Sag-to-span ratios of the stay cables, dci = Dci =Lci 0.01

Table 2. The degrees of localization of the mode shapes of different components and the closeness degrees of the natural
frequencies.
P
4
Mode Lr1 (%) Lr2 (%) Lr3 (%) Lr4 (%) 1 Lri (%) G r (%)
i=1

1 1.2438 21.6019 21.6019 1.2438 54.3086 47.9402

2 19.8862 15.6383 15.6383 19.8862 28.9510 10.1223
3 37.8115 1.9823 1.9823 37.8115 20.4124 6.3825
4 18.6717 25.5319 25.5319 18.6717 11.5928 6.3825
5 0.0024 49.9431 49.9431 0.0024 0.1090 0.0226
6 0.0021 49.9885 49.9885 0.0021 0.0188 0.0226
7 0.6643 27.3293 27.3293 0.6643 44.0128 10.4698
8 6.5361 10.7282 10.7282 6.5361 65.4714 13.0723
9 31.8058 0.4892 0.4892 31.8058 35.4100 2.0451
10 18.9553 25.6454 25.6454 18.9553 10.7986 1.3942
11 0.3224 48.6503 48.6503 0.3224 2.0546 1.3942
12 21.8307 21.4331 21.4331 21.8307 13.4724 1.8651
13 27.4705 0.0143 0.0143 27.4705 45.0304 6.2182
14 14.3402 1.0894 1.0894 14.3402 69.1408 6.2182
15 0.0002 49.9778 49.9778 0.0002 0.0440 0.0111
16 0.0042 49.9307 49.9307 0.0042 0.1302 0.0111
17 49.8283 0.0003 0.0003 49.8283 0.3428 0.0251
18 49.9181 0.0010 0.0010 49.9181 0.1618 0.0251
19 1.0456 0.0559 0.0559 1.0456 97.7970 5.4377
20 0.0003 49.9839 49.9839 0.0003 0.0316 0.0011

The degree of localization of the rth modeP shape of are close natural frequencies, but the converse may not
the deck beam can be measured by 1  4i = 1 Lri . The be true (mode localization occurs for mode 11, but the
closeness degrees of the natural frequencies and the natural frequencies are not close). With zero initial con-
degree of localization of the mode shapes of the stay ditions, the dynamic response of the cable-stayed bridge
cables and the deck beam are shown in Table 2. A natu- can be calculated from equation (26) using MATLAB’s
ral frequency is said to be close to its neighboring natu- hybrid fourth/fifth-order variable-step Runge–Kutta
ral frequencies if its closeness degree is larger than zero solver, ode45. It should be noted that Fa = 0:6mb g,
and less than 0.05%, and a mode shape is said to be and the modal damping ratio Crs in equation (28) is
localized at stay cables when the degree of localization always equal to 0:1drs for all the following calculation.
of the deck beam is less than 5.00%. It is observed that When the distance between any two neighboring
mode localization occurs on the stay cables when there forces (LF) is 30 m, the transverse displacement of the

Downloaded from ade.sagepub.com by guest on December 10, 2016

Song et al. 7

Figure 2. The first 20 natural frequencies of the cable-stayed bridge and their corresponding mode shapes.

Figure 4. Amplitude of the steady-state response of the

mid-point of the deck beam versus the velocity of the moving
forces; LF = 30 m. A vertical dashed line indicates that resonance
occurs when the velocity of the moving forces is that shown in
the expression near it there.

is set to 10, 20, and 50. The results presented in

Figure 3 and other results for a large number of cases
that are not reported here for the sake of brevity show
that Galerkin truncation with 20 terms yields rather
accurate results for the dynamic response of the cable-
stayed bridge subjected to a continuous sequence of
Figure 3. Dynamic response of the mid-point of the deck moving forces when the value of LF is larger than 30
beam for different numbers of Galerkin truncation terms when and the value of VF is smaller than 100. In the follow-
LF = 30 m: (a) VF = 17 m/s and (b) VF = 100 m/s. ing numerical examples, the first 20 modes of the line-
arized undamped cable-stayed bridge model are used in
mid-point of the deck beam with two different veloci- the Galerkin method.
ties (VF) of the moving forces, namely, VF = 17 m/s For 1 m=s  VF  100 m=s, Figure 4 plots the ampli-
and VF = 100 m/s are shown in Figure 3(a) and (b), tude of the steady-state response of the mid-point of the
respectively. The number of Galerkin truncation terms deck beam versus the velocity of the moving forces with

Downloaded from ade.sagepub.com by guest on December 10, 2016

8 Advances in Mechanical Engineering

Table 3. Values of the first 20 resonant velocities when LF = 30 m.

LF vr
Vvr = 2p Vv1 = 16.94 m/s Vv2 = 25.06 m/s Vv3 = 27.60 m/s Vv4 = 29.36 m/s
Vv5 = 39.11 m/s Vv6 = 39.12 m/s Vv7 = 43.22 m/s Vv8 = 51.34 m/s
Vv9 = 58.05 m/s Vv10 = 59.23 m/s Vv11 = 60.06 m/s Vv12 = 61.18 m/s
Vv13 = 65.90 m/s Vv14 = 70.00 m/s Vv15 = 78.27 m/s Vv16 = 78.28 m/s
Vv17 = 80.77 m/s Vv18 = 80.79 m/s Vv19 = 85.18 m/s Vv20 = 98.02 m/s

LF = 30 m as an example in both cases of considering facts: (1) the mid-point of the deck beam happens to be
and neglecting the geometric nonlinearities (the linear a node of the mode shapes for modes 2k (k = 1, 2, 3,
and nonlinear solutions). Paying attention to the fact .) that are anti-symmetric, which results in a zero
from equation (32) dynamic displacement for the mid-point of the deck
beam when the velocity of the moving forces is the
lbj
ð 2kth resonant velocities. (2) The mode shapes are loca-
X
‘ X
7
FrMF = fbja f(r)
bj d(xbj  
xbja (t))dxbj
a=1 j=1
lbj

lbj 0 0 11
X
‘ X
7 ð X j
@ V t
@ F  (a  1)lF  2
= fbja f(r)
bj d xbj  lbj + lbj AAdxbj
a=1 j=1
Le j = 1
lbj

lbj 0 0 11
X
‘ X
7 ð X j
@ V t
@ F  (a  2)lF  2
= fbja f(r)
bj d xbj  lbj + lbj AAdxbj ð35Þ
a=1 j=1
Le j = 1
lbj

lbj 0 0 11
X
‘ X
7 ð   X j
@ V eL
@ F t + F  (a  1)lF  2
= fbja f(r)
bj d xbj  lbj + lbj AAdxbj
a=1 j=1
Le VF j = 1
lbj

lbj
X
‘ X
7 ð   
eLF
= fbja f(r)
bj d xbj  xbja t+ dxbj
a=1 j=1
VF
lbj

a continuous sequence of identical, equally spaced lized on the stay cables for modes 5, 11, 15, and 17;
moving forces can be regarded as a periodic force with hence, the moving forces cannot excite the vibration of
a nondimensional period T that equals to eLF/VF. the cable-stayed bridge when their velocity is the 5th,
Hence, resonance may occur if the moving velocity of 11th, 15th, and 17th resonant velocities, which can be
the forces is Vvr = (LF vr )=2p or is close Vvr by linear seen from equation (32).
or nonlinear analysis. The velocity Vvr is called the rth Resonance can also occur when the velocity of the
resonant velocity in this work. The first 20 resonant moving forces is the product of a unit fraction and a
velocities when LF = 30 m are listed in Table 3. resonant velocity, which can be seen from the peaks
However, as shown from the linear solutions in when the velocity of the moving forces is Vv1/2, Vv1/3,
Figure 4, there are not peaks on the curve when the Vv13/2, and Vv19/4 on the linear solutions in Figure 4.
moving velocity of the forces is a resonant velocity The reason for this lies in the fact that there are higher
except the 1st, 3rd, 7th, 9th, 13th, and 19th resonant harmonics with frequencies equal to integer multiples
velocities. The reasons for this lie in the following two of the fundamental frequency 2p/T in the Fourier series
2 3
Tð=2 X
‘  Tð=2  
1 2 6 2p 2p 7
FrMF (t) = FrMF dt + 4cos n t FrMF cos n t dt5
T T n=1 T T
T=2 T =2
2 3 ð36Þ
X‘   Tð=2  
2 6 2p 2p 7
+ 4sin n t FrMF sin n t dt5
T n=1 T T
T =2

Downloaded from ade.sagepub.com by guest on December 10, 2016

Song et al. 9

moving forces with a resonant velocity corresponding

to a mode of the cable-stayed bridge, whose natural fre-
quency is close to those of its neighboring modes and
whose mode shape is localized on the stay cables, can-
not excite the vibration of that mode. This means that
modes with close natural frequencies, whose mode
shapes are localized on stay cables, play a minor role in
the dynamic response of the cable-stayed bridge. (2)
The geometric nonlinearities of stay cables can gener-
ally have some influence on the dynamic response of
the cable-stayed bridge, but have almost no influence
on that when the velocity of the moving forces is
around the resonant velocity corresponding to a mode
Figure 5. Amplitude of the dynamic response of the mid-point
whose mode shape is localized on the deck beam. (3)
of the deck beam versus the distance between any two
neighboring forces.
Resonance can also occur when the velocity of the mov-
ing forces is the product of a unit fraction and a reso-
for the periodic force FrMF in equation (32). In addition, nant velocity. (4) There is a critical distance between
comparing the linear and nonlinear solutions in any two neighboring forces below which transverse dis-
Figure 4, one can see that the geometric nonlinearities placements of the cable-stayed bridge generally increase
of the stay cables have some influence on the dynamic with the distance.
response of the cable-stayed bridge. Generally, the
peaks of the curves do not exactly occur at the Declaration of conflicting interests
undamped natural frequencies of the linearized model The author(s) declared no potential conflicts of interest with
of the cable-stayed bridge. However, the linear and respect to the research, authorship, and/or publication of this
nonlinear solutions almost coincide with each other article.
when the velocity of the moving forces is around the
19th resonant velocity. The reason for this lies in the
Funding
fact that the deck beam, which is a linear component,
has a major contribution to the dynamic response of The author(s) disclosed receipt of the following financial sup-
port for the research, authorship, and/or publication of this
the cable-stayed bridge since the mode shape for mode
article: This work was supported by the National Natural
19 is localized on the deck beam.
Science Foundation of China under Grant Nos 11302087 and
Figure 5 shows the amplitude of the steady-state 11442006, Natural Science Foundation of Jiangsu Province
response of the mid-point of the deck beam versus the under Grant No. BK20130479, Research Foundation for
distance between any two neighboring forces when the Advanced Talents of Jiangsu University under Grant No.
velocity of moving forces is the first resonant velocity. 13JDG068, and National Science Foundation under Grant
It can be seen that there are a critical distance No. CMMI-1000830.
LF = 193 m. In the region LF \ 193 m, the amplitude
of the steady-state response of the mid-point of the References
deck beam generally increases with LF and in the region
1. Timoshenko SP. On the forced vibrations of bridges.
LF . 193 m decreases with LF. However, there are fluc-
Philos Mag 1922; 43: 1018–1019.
tuations at some LF. It can also be seen from Figure 5
2. Jeffcott HH. On the vibration of beams under the action
that the geometric nonlinearities of the stay cables have of moving loads. Philos Mag 1929; 8: 66–97.
some influence on the dynamic response of the cable- 3. Ting EC, Genin J and Ginsberg JH. A general algorithm
stayed bridge. for moving mass problem. J Sound Vib 1974; 33: 49–58.
4. Blejwas TE, Feng CC and Ayre RS. Dynamic interaction
of moving vehicles and structures. J Sound Vib 1979; 67:
Conclusion 513–521.
The dynamic behavior of a cable-stayed bridge that 5. Kang Z, Xu K and Luo Z. A numerical study on non-
linear vibration of an inclined cable coupled with the
consists of a simply supported four-cable-stayed deck
deck in cable-stayed bridges. J Vib Control 2011; 18:
beam and two rigid towers subjected to a continuous 404–416.
sequence of identical, equally spaced forces moving at a 6. Olsson M. Finite element modal co-ordinate analysis of
constant velocity has been investigated. The dynamic structures subjected to moving loads. J Sound Vib 1985;
response of the cable-stayed bridge is calculated using 99: 1–12.
the Galerkin method via MATLAB. Some conclusions 7. Timoshenko S, Young DH and Weaver W. Vibration
from the numerical simulation are as follows: (1) the problems in engineering. 4th ed. New York: Wiley, 1974.

Downloaded from ade.sagepub.com by guest on December 10, 2016

10 Advances in Mechanical Engineering

8. Guo WH and Xu YL. Fully computerized approach to 18. Natsiavas S. Mode localization and frequency veering in
study cable-stayed bridge-vehicle interaction. J Sound a non-conservative mechanical system with dissimilar
Vib 2001; 248: 745–761. components. J Sound Vib 1993; 165: 137–147.
9. Madrazo-Aguirre F, Ruiz-Teran AM and Wadee MA. 19. Balmes E. High modal density, curve veering, localiza-
Dynamic behaviour of steel-concrete composite under- tion: a different perspective on the structural response. J
deck cable-stayed bridges under the action of moving Sound Vib 1993; 161: 359–363.
loads. Eng Struct 2015; 103: 260–274. 20. Cao DQ, Song MT, Zhu WD, et al. Modeling and analy-
10. Meirovitch L and Hale AL. On the substructure synthesis sis of the in-plane vibration of a complex cable-stayed
method. AIAA J 1981; 19: 940–947. bridge. J Sound Vib 2012; 331: 5685–5714.
11. Meirovitch L and Hale AL. A general dynamic synthesis 21. Nelson HD and Conover RA. Dynamic stability of a
for structures with discrete substructures. J Sound Vib beam carrying moving masses. J Appl Mech 1971; 38:
1982; 85: 445–457. 1003–1006.
12. Leung AYT. A simple dynamic substructure method. 22. Benedetti GA. Dynamic stability of a beam loaded by a
Earthquake Eng Struc Dyn 1988; 16: 827–837. sequence of moving mass particles. J Appl Mech 1974; 41:
13. Leung AYT. Multilevel dynamic substructures. Int J 1069–1071.
Numer Meth Eng 1989; 28: 181–191. 23. Mackertich S. Dynamic stability of a beam excited by a
14. Kong X, Wu DJ, Cai CS, et al. New strategy of substruc- sequence of moving mass particles. J Acoust Soc Am
ture method to model long-span hybrid cable-stayed 2004; 115: 1416–1419.
bridges under vehicle-induced vibration. Eng Struct 2012; 24. Pirmoradian M, Keshmiri M and Karimpour H. On the
34: 421–435. parametric excitation of a Timoshenko beam due to inter-
15. Pierre C. Mode localization and eigenvalue loci veering mittent passage of moving masses: instability and reso-
phenomena in disordered structures. J Sound Vib 1988; nance analysis. Acta Mech 2015; 226: 1241–1253.
126: 485–502. 25. Song MT, Cao DQ, Zhu WD, et al. Dynamic response of
16. Wei ST and Pierre C. Localization phenomena in mis- a cable-stayed bridge subjected to a moving vehicle load.
tuned assemblies with cyclic symmetry part I: free vibra- Acta Mech 2016; 227: 2925–2945.
tions. J Vib Acoust 1988; 110: 429–438. 26. Gattulli V and Lepidi M. Localization and veering in the
17. Wei ST and Pierre C. Localization phenomena in mis- dynamics of cable-stayed bridges. Comput Struct 2007;
tuned assemblies with cyclic symmetry part II: forced 85: 1661–1678.
vibrations. J Vib Acoust 1988; 110: 439–449.

Downloaded from ade.sagepub.com by guest on December 10, 2016