Design Research of Railway Bridges with Span Length Over 1000m in China

Gongyi XU, Daoming TIAN Xiaozhen LI

Dr., CEng, Professor, Professor Professor
National Engineering BRDI South-West Jiaotong
Design Master of China， Wuhan, P.R.China University,
BRDI, Wuhan, P. R. China, [email protected] Chengdu, P.R.China
[email protected] [email protected]
Gongyi XU, born 1963, received Daoming Tian, born 1963,
his PhD from South West received his bachelor’s degree Xiaozhen LI, born 1970, received
Jiaotong University, P.R. China from Wuhan University, P. R. his PhD from Southwest Jiaotong
China University, P. R. China

Summary
Based on Yangtze River crossing railway passage project connecting Shanghai and Nantong, the
mechanical performance of a cable-stayed bridge and a suspension bridge schemes with 1008m
spans are analyzed for design conception comparison. The double line railway is designed as per
Chinese standard for common use of both passenger trains and freight trains. Static deflection
analysis and dynamic simulation analysis of train-bridge coupling system are carried out.
Analysis results indicate that under load application of Chinese freight train at a speed of 80km/h
or high-speed passenger train at a speed of 200km/h, the static and dynamic deformation
performance of cable-stayed or suspension bridge with 1008m span meets safe running of trains
and the bridges have sufficient stiffness. The dynamic running stability and comfort index of
currently adopted Chinese trains can be evaluated as “satisfactory”.

Keywords: Railway bridge, spans over 1000m; cable-stayed bridge; suspension bridge,
mechanical performance

The South Bisan Seto Bridge in Japan (L=1100m)

(DAN M Frangopol et al,2004), and the Tagus
Bridge in Portugal (L=1013m) ( TERESA de
Lemos et al,2004) are only two suspension
bridges in operation with spans over 1000m. For
the cable-stayed bridge, Tianxingzhou Bridge in
China has the largest span (L=504m) (Qin et al,
2007).
The newly planned Hutong (Shanghai-Nantong)
Railway Project, will cross the Yangtze River at
8km downstream of the existing Sutong Highway
Yangtze River Bridge for alignment of general Fig.1 Bridge site location map
layout of the whole line, as shown in Fig.1 The
river is about 10km wide there with busy shipping traffic, so the Hutong Bridge should adopt same
navigation standard with the Sutong Bridge for the purpose of navigational requirements. So bridge
proposals with 1008m main spans are adopted for feasibility study for comparison with tunnel
proposals (Xu et al, 2007).
The railway design adopts Chinese National Grade I with double railway lines for common use of
passenger trains (Chinese Railway Standard Live load) and freight train (ZK live load), design train
speed is 160-200km/h. In case of combination of highway with railway, the design standard of
highway is Grade I with dual three-lane carriageway, the design vehicle speed is 100km/h. Net
dual-way navigational width is not less than 900m and that of single navigational channel not less
than 504m.

1. Introduction of Main Bridge Proposal

The general layout of the cable-stayed bridge 2156

and suspension bridge are shown in Fig.2 and 140 140 294 1008 294 140 140

Fig.3. The cable-stayed bridge proposal is

140m+140m+294m+1008m+294m+140m ＋
140m. The suspension bridge proposal adopts
twin suspension bridges with an Fig. 2 The cable-stayed proposal (Unit: m)
intermediated anchorage in between as 2×
(140+140+1008+140+140) m (China 3366
15 15
380 1008 140 140 140 140 1008 380
Railway Major Bridge Reconnaissance &
Design Institute, 2007).
Fig.3 the suspension bridge (unit: m)

1000 1500 1000 For the cable-stayed proposal, the main truss adopts
steel truss composite structure with triangle truss and
inverted trapezoid section, as shown in Fig.4. For
600

single built double line railway proposal, steel box

1600

structure is adopted for lower railway deck to

700

440

improve stiffness of the main girder. For combined

(a) Single built double line railway proposal

3200
850 900 900 850

2% 2%
1480

440
700

440

(b) combined highway and railway proposal Fig.5 Main truss section for combined railway and
Fig. 4 Main girder section (Unit: cm) highway suspension bridge proposal (Unit: cm)

highway and railway proposal, highway is arranged on the upper deck and railway on the lower.
The suspension bridge proposal for combined highway and railway lines is shown in Fig. 5. In order
to increase wind-resistance stability, a truss structure is adopted; the main truss is of Warren Type
with a panel length of 14m. The truss is 16m deep, with 15m wide lower chord and 32m upper chord.
2. Static Deformation Analysis Table 1 Displacement analysis results
In the static analysis, all the steel truss Bridge Type Mid-span Deflection/ Girder end
members are simulated as spatial beam deflection(m) span ratio rotation(Rad)
elements, and the orthotropic monolithic Cable-stayed
steel deck adopts grillage simulation. 1.499 1/672 0.94/1000
railway bridge
Fig. 6 is the sketch of the finite element Cable-stayed
model of the cable-stayed bridge and the rail-cum-road 1.611 1/626 1.43/1000
suspension bridge. The railway live load bridge
adopts Chinese live load with doubled Suspension
line live load, and that of the highway is 2.498 1/404 2.00/1000
bridge
Grade I live load with dual three-lane
carriageway. The analysis result in Table 1 indicates that the static deformation performance of the
two bridge types meets requirements and the bridges have satisfactory stiffness.

3. Train-bridge Dynamic Performance Analysis

For large span bridges, dynamic performance is of critical importance for running safety of trains. It
is the control factor of bridge design and of significant importance in suspension or cable-stayed
bridges with spans over 1000m. (Li, 1992, Lei et al, 2002 and HSU, 2007) In the past half decade,
analysis of dynamic responses under train loads has been developed from classic analytic equation
method to the train-bridge coupling dynamic method (Xia et al, 2002, Li et al, 1999 and Xiang,
2007).
Generally, according to D’Alembet Principle, the train system vibration equation is as following:

&& v + Cv u& v + K v u v = Pv
Mvu (1)

Where, Mv、Cv、Kv are mass, damping and stiffness matrix of vehicles respectively, uv 、 u& v 、 u&&v

stand for column vectors of displacement, speed, and acceleration speed of degrees of freedom of the
train, Pv is the load column vector applied to degrees of freedom in the vibration, which is normally
function of track irregularities and bridge vibration state.
When a train runs through the bridge with constant speed, the interaction between different trains can
be normally ignored because the vibration equation between trains are independent, and the dynamic
properties of coaches only couple with vibration of the bridge. In the study, the trains can be defined
as a multiple rigid bodies with multiple degrees of freedom composed of train bodies, front and back
bogies, wheel sets, and primary and secondary suspension system elastic elements. Where, the train
body and front and back bogies can be assumed as a rigid body with five degrees of freedoms, which
are, bouncing, swing, side rolling, pitching and yawing, and each wheel set are considered as two
degrees of freedom. The train body, front and back bogies and wheel sets are connected with primary
and secondary suspension systems. So there are totally 23 degrees of freedom for a train with four
axles, and 27 degrees of freedom of a six-axle locomotive.
Motion equation of the bridge system can be defined as following with matrix
&&b + Cb u& b + K b ub = Pb
Mb u (2)
Where Mb、Cb、Kb are mass, damping, and
stiffness matrix of the bridge, ub 、u& b 、u&&b stand

for column vectors of displacement, speed, and

acceleration speed of the degree of freedom of (a) Model of the cable-stayed bridge
the bridge. Pb is the load column vector of each
degree of freedom applied to the bridge from
the vehicles, which is function of vehicle
vibration motion on the bridge.
The dynamic Finite Element model of the (b) Model of the suspension bridge
cable-stayed bridge and suspension bridge are Fig. 6 Finite element model
shown in Fig.6.
The right item of Equation (1) and Equation Table 2 Train parameter
(2) are coupled, independent train with Train Type Train composition
bridge vibration equation are adopted for One DF4+50 C62 freight
F4 tracking Freight C62
numerical integration, which are iterative train
solution each other. The convergence China-Made “Pioneer” Power 5×（motor+ trailor+motor） ，
condition is the relative error of integration Distribution Train Set 15units
impact between wheel set with bridge track China-Made “China Star” Power 1 motor +9trailor+1motor，
should be smaller than the tolerant error, and Concentrated Train Set 11 units

Newmark-
integratio Table 3 Analysis results of Sperling Index

n method is applied in Train Type Bridge Type Motor Motor Trailer Trailer
each time step length Vertical Transversal Vertical Transversal
(Li et al, 1999). DF4 Cable-stayed 2.29 3.02 3.67 4.04
The design speed is （railway）
Locomotive (good) (good) (good) (acceptable)
160km/h (200km/h C62 Freight
plane condition Suspension 1.75 2.13 3.73 4.08
Train
preserved); the train bridge (good) (good) (good) (acceptable)
type (only dynamic
China-Made Cable-stayed 2.70 2.73 2.67 2.70
effect when single
“Pioneer” （railway） (good) (good) (good) (good)
track train running
Power
through is considered) Cable-stayed(Rail 2.74 2.71 2.69 2.69
Distribution
and its composition are -cum-Road) (good) (good) (good) (good)
Train Set
shown in table 2. The Suspension 2.71 2.74 2.69 2.71
track irregularities Bridge (good) (good) (good) (good)
applied are from field China-Made Cable-stayed 2.69 2.72 2.73 2.69
test on the “China Star” （railway） (good) (good) (good) (good)
Qinhuangdao-Shenyan Power Cable-stayed(Ra 2.72 2.72 2.74 2.68
g passenger railway Concentrated il-cum-Road) (good) (good) (good) (good)
line. Sperling index is Train Set Suspension 2.70 2.73 2.65 2.71
adopted to evaluate the bridge (good) (good) (good) (good)
riding comfort and
running stability in this paper (Garg V K et al, 1984).Table 3 and Table 4 show the dynamic
responses when the passenger train runs at 200km/h and the freight train at 80km/h. The results
indicate that the train comfort and running stability has achieved satisfactory requirement
(transversal comfort of the freight tractor is qualified). The vertical and transversal acceleration is
relatively small and the bridge has sufficient dynamic stiffness (Li et al, 2008).

4. Conclusion
Table 4 Dynamic responses of bridge
Mid-Main Span Deck Mid-Main Span Deck Mid-Main span
Maximum Vibration Maximum vibration Vertical
Train Type Bridge Type 2
Displacement (mm) acceleration (m/s ) Deflection
Vertical Transversal Vertical Transversal /span ratio
Cable-stayed
470.13 26.74 0.28 0.34 1/2144
DF4 Locomotive bridge（railway）
C62Freight Train Suspension
787.41 27.84 0.19 0.33 1/1280
bridge
Cable-stayed
156.24 10.17 0.17 0.21 1/6452
bridge（railway）
China-Made
Cable stayed
“Pioneer”
bridge 195.84 10.76 0.22 0.22 1/5147
Power Distribution
（rail-cum-road）
Train Set
Suspension
225.13 10.23 0.18 0.23 1/4477
bridge
Cable-stayed
China-Made 165.31 10.75 0.19 0.23 1/6098
bridge（railway）
“China Star”
Cable stayed
Power
bridge 207.34 11.31 0.23 0.24 1/4861
Concentrated Train
（rail-cum-road）
Set
Suspension bridge 235.44 10.59 0.19 0.27 1/4281

The analysis indicates that the proposed suspension bridge or cable-stayed bridge with 1008m span
can meet the requirements of passenger trains and freight trains at the design speed. So application of
cable-stayed bridge or suspension bridge with spans over 1000m in Shanghai to Nantong railway
passage project is technically feasible.
According to the large span bridges in operation in Japan and China, a deflection/span ration more
than 1/400 will not impact railway running safety, specific criteria should be established for long and
major bridges according to running stability of trains. The girder end rotation can be reduced by extra
end span to make the relieved maximum rotation angle be smaller than the restrict of 0.002 arc.
The above proposals are proposed at earlier study stage of the Shanghai-Nantong Yangtze River
Crossing Project, after multiple verification organized by China Ministry of Railways, the bridge
proposal is considered as more economical than tunnel proposal, so the bridge will be implemented.
Due to complicated water flow of this river section, the planned implementation location of the
bridge will be shifted 20km upstream, so the main bridge proposals will be adjusted accordingly.
References
[1] China Railway Major Bridge Reconnaissance & Design Institute. Feasibility report of project
crossing Yangzi River of railway line connecting Shanghai and Nantong [R].Wuhan,China
Railway Major Bridge Reconnaissance & Design Institute, 2007, in Chinese
[2] DAN M. Frangopol, KIYOHIRO Imai, Reliability of long span bridges based on design
experience with the Honshu–Shikoku bridges [J], Journal of Constructional Steel Research,
2004, V60(3): 373-392.
[3] Garg V K, Dukkipati R V. Dynamics of railway vehicle systems[M]. Toronto: Academic
Press, 1984
[4] Lei Junqing, Zheng Mingzhu, Xu Gongyi, Suspension Bridge Design[M], Beijing: China
Communications Press, 2002
[5] Li Guohao, Stability and vibration of bridge structures [M] , Beijing:. Railway Publishing
House of China, 1992.
[6] LI Xiaozhen, QIANG Shizhong, Coupling vibration analysis for long span steel cable stayed
bridge crossing over Yangtze river at Nanjing on Beijing Shanghai high speed railway,
Journal of Southwest Jiaotong University, 1999, V34(2): 153-157. in Chinese.
[7] LI Xiaozhen, QIANG Shizhong, Analysis Report for the bridges crossing over Yangtze river
at Nantong to Shanghai high speed railway[R]. Southwest Jiaotong University, 2008. in
Chinese
[8] HSU Gong-yi, SHU Si-li, YANG Jin, Application of suspension bridges stiffened by
prestressed concrete slabs in China [J]. Journal of Bridge Engineering, 2007, V12(2):
256-260.
[9] TERESA de Lemos, DAVID Eaton, MARTIN Betts, LUIS Tadeu de Almeida, Risk
management in the Lusoponte concession—a case study of the two bridges in Lisbon,
Portugal [J], International Journal of Project Management, 2004, V22(1): 63-73
[10] Xia He, Zhang Nan, Dynamic interaction of vehicles and structures, [M], Beijing: Science
Press, 2002
[11] XIANG Tianyu, ZHAO Renda, XU Tengfei, Reliability Evaluation of Vehicle-Bridge
Dynamic Interaction [J], Journal of Structure Engineering, ASCE, 2007, 133(8): 1092-1099.
[12] Xu gongyi, Mei xinyong, Zhang Yanfei. Feasibility report of project crossing Yangtze River
of railway line connecting Shanghai and Nantong Compare Bridge with Tunnel [R].