Engineering Structures 100 (2015) 707–717

Contents lists available at ScienceDirect

Engineering Structures
journal homepage: www.elsevier.com/locate/engstruct

Review article

Recent development of design and construction of short span high-speed

railway bridges in China
Bin Yan 1, Gong-Lian Dai 1, Nan Hu ⇑,2
School of Civil Engineering, Central South University, Changsha, Hunan 410075, China

a r t i c l e i n f o a b s t r a c t

Article history: High-speed rail (HSR) has become a vital component of the national transportation network in China. A
Received 23 February 2014 large number of standard short-span bridges were constructed as part of the infrastructure associated
Revised 26 June 2015 with the HSR development. The main focus of this short review is to showcase how the fast HSR construc-
Accepted 29 June 2015
tion in China during the past decade was achieved by using typical simply supported bridges and contin-
Available online 10 July 2015
uous beam bridges. This paper provides a brief historical review of the HSR development in China along
with emerging design issues. Design concepts and structural dimensions of two typical spans are dis-
Keywords:
cussed, including the superstructure, substructure as well as the auxiliary facilities on the bridge deck.
High-speed rail
Bridges
Different construction methods were also discussed.
Short span Ó 2015 Elsevier Ltd. All rights reserved.
Design
Construction

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 707
2. Brief history . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 708
3. General design issues. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 710
3.1. Live load. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 710
3.2. Serviceability limits. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 711
3.3. Other issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 711
4. Typical superstructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 712
4.1. Simply supported beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 712
4.2. Continuous beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 713
4.3. Deck system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 715
5. Typical substructure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 715
6. Typical construction method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 716
7. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 716
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 716
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 716

1. Introduction

High-speed Rail (HSR) offers a fast and robust travel option that
⇑ Corresponding author at: College of Engineering, Michigan State University, MI
enhances the quality of life and supports economic growth. Japan
48824, USA. was the first country that built a passenger dedicated line (PDL)
E-mail addresses: binyan@csu.edu.cn (B. Yan), daigong@vip.sina.com (G.-L. Dai), for high-speed travel purpose, also known as Shinkansen. The first
hunan2@msu.edu (N. Hu). Shinkansen opened for the Tokyo Olympics in 1964 between Tokyo
1
Equal contribution. and Osaka. HSR in Europe first developed in several countries and
2
Former graduate student researcher at Central South University and currently, then has expanded into a regional service network. Over the past
postdoctoral research fellow at Dartmouth College.

http://dx.doi.org/10.1016/j.engstruct.2015.06.050
0141-0296/Ó 2015 Elsevier Ltd. All rights reserved.
708 B. Yan et al. / Engineering Structures 100 (2015) 707–717

few decades, a total of 13 countries have developed HSR network, Shanghai-Chengdu

mainly in Europe and East Asia. International examples from those
countries have proved that high-speed trains are capable of reach- Qingdao-Taiyuan
ing a speed over 250 km/h on a PDL, which can significantly reduce
Shanghai-Kunming
travel hours. A detailed historical review about the HSR develop-
ment in those countries can be found in papers by Taniguchi [1], Xuzhou-Lanzhou
Bouley [2], the European Commission [3], Gourvish [4], Zuber [5],
and Harrison et al. [6]. Hangzhou-Shenzhen

The worldwide HSR construction is still an ongoing process

Beijing-Harbin
which many countries have proposed future plans for developing
HSR networks, such as in USA and in Australia [7–10]. A survey Beijing-HongKong
data collected by the International Union of Railways (UIC) indi-
Beijing-Shanghai
cates that China is currently the leading country on HSR construc-
tion, as shown in Fig. 1. HSR in China is composed of upgraded 0% 20% 40% 60% 80%
existing lines with an average design speed of 250 km/h and new Bridges Proportion
lines with an average design speed of 350 km/h. Specifically, it
includes 9356 km of new built lines and 3209 km of upgraded Fig. 3. The percent of bridges in the infrastructure of major HSR lines of China.
lines. By 2020, the total length of HSR lines in China will reach
and local economic growth by providing an efficient means of
more than 20,000 km with a complete grid network that will con-
direct, high-speed transportation for individuals and businesses
nect major provincial capital cities as well as large cities with more
in China.
than five million population [11]. During the past decade, fast
Bridges are a major part of the HSR infrastructure, covering
expansion of the HSR network in China depended on the supports
more than 50% of total HSR mileage in China because the use of
of government policy, the increases in fiscal investment and the
bridges can avoid the interruption of existing lines and the occupa-
growths of expertise. The planning and construction of the HSR
tion of land [12]. Fig. 3 lists the proportions of bridge with respect
network in China was aimed to solve the overcrowding issue on
to the total length in major HSR lines of China [13]. Fast construc-
the regular railway system. HSR offers additional advantages, such
tion usually requires using a typical structural design, along with
as reduction in greenhouse gas emissions over other travel options,
construction details and procedures. Several key design features
as compared in Fig. 2. Thus, HSR revolutionized travel experience
are compared between typical short HSR bridges from six countries
in Table 1. A typical PDL in China usually includes a big portion of
Other simply supported beam bridges, a few continuous beam bridges
8.6% Japan(10.8%) and a few middle/long span bridges (usually single span more than
100 m) over existing lines or rivers. For example, the Beijing–
Shanghai PDL includes 90% simply supported beam along with
Spain(9.2%) 5% continuous beam and only 5% are medium/long spans [14]. In
the Chinese HSR bridge design specification, short span is defined
as a bridge with a single span smaller than 30 m. It can be seen
in Table 1 that two typical structural types in a HSR line of China
France(8.2%) are simply supported beam and continuous beam. The design
and construction of these two structural types has been improved
China over the years and eventually been considered as a mature indus-
50.9% Germany(5.4%) trial product. As a result, rapid design–construction process can be
achieved by using a full set of construction drawing, building steps
Italy(3.7%) and maintenance plan.
The objective of this paper is to present an up-to-date review on
Sweden(3%)
the emerging design and construction techniques about HSR
bridges in China, particularly on short spans (including standard
Fig. 1. HSR mileage by countries [11].
simply support beam and continuous beam). Another companion
paper [15] has been published in an earlier issue by Engineering
Structures that focused on the medium and large span HSR bridges.
This review will showcase how the use of typical design has con-
1
tributed in promoting HSR bridge to a highly industrialized prod-
Carbon dioxide emissions 4 uct. This paper includes a brief history of HSR in China followed
6 by an introduction of special issues that emerged in the design of
HSR bridges. Technical issues on the design and construction of
1 High-speed railway typical short span HSR bridges are discussed, including superstruc-
Occupied land 2 Motor coach ture, substructure and deck system. Information for this review is
Airplane
0 partially collected from design companies in China and partially
from the literatures.
1
Energy consumption 5.3 2. Brief history
5.6
The HSR development in China can be divided into five main
0 1 2 3 4 5 6
phases [16]: (1) from early 1980s to 1993, a decade-long feasible
Fig. 2. Dimensionless comparison of sustainability issues on three travel options study period by policy makers from the former Ministry of
[16]. Railways (MOR) of China; (2) from 1994 to 1997, a three-year
Table 1
Comparison of key features of high speed rail beam bridges among six countries.

Japan France Germany Spain Italy China

Beam type by Reinforced concrete, prestressed Reinforced concrete, Prestressed concrete Prestressed concrete, steel– Prestressed concrete, steel–concrete Prestressed concrete
material concrete, steel–concrete hybrid, steel prestressed concrete, concrete hybrid hybrid
steel–concrete hybrid,
steel

B. Yan et al. / Engineering Structures 100 (2015) 707–717

Typical cross-section T girder, box-girder Box-girder Box-girder T girder, box-girder Box-girder Box-girder
Cross-section
diagram

Typical span Simply supported: 24.2 m, 29.2 m, Simply Supported: 625 m Simply Supported: 25 m, Simply Supported: 26.6 m Simply Supported: 24 m, 33.6 m, Simply Supported:
34.2 m, 39.2 m and 44.2 m 44 m and 58 m 43.2 m and 55.0 m 32 m
Continuous: (3 + 3  6 + 3) m, (4– Continuous: 40–80 m Continuous: 45 m Continuous: 40–
5)  8 m, (4–10)  10 m, 5  15 m 100 m
Typical pier Rectangular wall or circular wall Rectangular wall or Rectangular wall Rectangular wall Rectangular wall Round ended wall
circular wall
Typical foundation Single pile, pile group and spread Pile group Single pile Single pile Single pile Pile group
footing Spread footing Pile group Pile group
Typical construction Precast Precast Cantilever Cantilever Cantilever Precast
method Cast-in-place Cantilever Cast-in-place Cast-in-place Cantilever
Cast-in-place Launching Launching

709
710 B. Yan et al. / Engineering Structures 100 (2015) 707–717

Fig. 4. HSR grid system in China by 2020.

period of extensive technical studies by major universities and considering the dynamic impact, which is close to the actual load
research agencies affiliated to the former MOR; (3) from 1998 to distribution of the train. Fig. 5a shows the major type of
2002, a series of experimental studies that carried out on several high-speed train used in China. Design live load was selected based
pilot HSR lines. The first HSR line, Qinhuangdao–Shenyang seg- on numerical and experimental investigations of the actual
ment was built in 1999 and had an average speed of 200 km/h; response on HSR bridges under different live load models. Fig. 5b
(4) from 2003 to 2007, a series of design specifications were
released based on lessons learned from nation-wide experience
and world-wide consulting; (5) from 2008 till now, milestone lines
have been completed and large-scale constructions are still in pro-
cess. For example, a 113.5 km long PDL connecting Beijing and
Tianjin was opened for the 2008 Olympic Games with an average
speed of 300 km/h. A 12,000 km grid system is expected to com-
plete by 2020. This grid system is composed of four North–South
corridors and four East–West corridors to meet the increasing need
in these regions, as shown in Fig. 4.

3. General design issues

Compared to a regular railway bridge, the design of a HSR bridge

requires a higher service limits due to the need of track smoothness
and train stability. Given that the live load on HSR bridges is about
(a) CRH380b/bl Train
45–65% lighter than a regular railway, the control of displacement
becomes the major design issue rather than the load-carrying
4×200kN 4×200kN
capacity [12,17]. In addition, bridges are expected to have high
64kN/m 64kN/m
durability, low maintenance, and pleasant appearance.

3.1. Live load

∞ 0.8m 1.6m 1.6m 1.6m 0.8m ∞ 1.6m 1.6m 1.6m
The use of the live load model for HSR bridge design varies from Standard Special
country to country. The live load model proposed by the (b) ZK live load
International Union of Railways (UIC) is widely adopted by
European countries, while Japan used single-track live load with Fig. 5. HSR train and design live load model adopted in China.
 B. Yan et al. / Engineering Structures 100 (2015) 707–717 711

presents so-called ‘‘ZK’’ live load model adopted in China. This load Table 3
mode includes a standard live load (80% of the UIC live load) and a Minimum requirements on longitudinal stiffness of supports for HSR simply
supported beam bridges (kN/m).
special live load [18]. Then, dynamic effects are simplified as a
quasi-static load by multiplying a dynamic factor (1 + l) on the Span (m) 612 16 20 24 32 40 48
static load. The coefficient l can be estimated by Pier 100 160 190 270 350 550 720
l = 1.44/(Lu0.5  0.2)  0.18, where Lu is equal to the span between (60) (100) (120) (170) (220) (340) (450)
supports for simple beam bridges and equal to the average span for Abutment 3000 (1500)

continuous bridges [19]. However, Dai et al. [20] mentioned that Note: The values in brackets are used for bridges with a single line.
the structural response under current ZK load is overestimated
and proposed to consider only 60% of the UIC load. Further inves-
tigations are still ongoing to determine an appropriate load for structure is smaller than the allowable values, a train-structure
future HSR line. coupled numerical analysis must be conducted to check other
parameters in Table 4 such that the safety of trains under high
3.2. Serviceability limits speed can be ensured [23]. The derailment coefficient is another
general parameter to measure the stability of the train wheels. A
The former MOR in China has issued a series of design specifica- wheel loading reducing rate is used to check the uneven force on
tions based on lessons learned from the International Union of each side wheel of a high-speed train. Sperling’s ride index is used
Railways (UIC) and the German Institute for Standardization to assess the effect of mechanical vibrations. All these factors are
(DIN), along with a large number of experimental studies by major used to evaluate the ride quality and comfort of railway vehicles.
universities in China [21–23]. The serviceability limits in designing Overall, three most important limits in designing of HSR bridges
of a HSR bridge are given in Table 2. The threshold limits on bridges are the vertical stiffness, beam end rotation and longitudinal stiff-
with a ballastless track bed are higher than bridges with ballast ness of piers.
track bed because it is difficult to adjust the smoothness of a bal-
lastless deck. 3.3. Other issues
Four key design parameters are among major concerns includ-
ing (1) vertical deflection of the beam that could lead to large Due to higher serviceability limits compared to the ones in con-
vibrations of the HSR train [24,25], (2) beam end rotation that ventional railway bridge design, other technical issues associated
could cause track instability [26,27], (3) long-term deflections that with dynamic response of HSR bridges have been investigated by
can result in an uneven track surface and threaten the operation of many previous research efforts, such as seismic performance
HSR trains [28,29], and (4) longitudinal deflection of the substruc- [33–38], track-structure interaction [39–46], and uneven settle-
ture that is used to control the nonlinear behavior between contin- ment [47].
uous track and the bridge superstructure and thus ensure the Track-structure interaction is a major concern to ensure the
stability of the track [30–32]. smoothness of track as well as the safety of the train. Tracks can
Table 3 lists the minimum required longitudinal stiffness of be regarded as additional constraints on the bridge superstructure.
piers for a simple beam bridge [21,23]. A series of dynamic criteria Yan [31] recommended that the stiffness of substructure and the
was established by design specifications to ensure the safety of the alignment of tracks should be estimated by considering the
train, as shown in Table 4. If the fundamental frequency of a track-structure interaction. Yan and Dai [32] found that the longi-
tudinal deflection of piers and the relative rotation at beam end
under earthquake load are smaller with the consideration of track-
Table 2 ing in the numerical model.
Serviceability criteria in HSR bridges.
The aerodynamic effects due to a high-speed train may result in
Contents Ballast Ballastless track instability of surrounding structures, such as noise barriers and
track overpasses [48]. Chen [50] carried out a case study on the aerody-
Design life 100 years 100 years namic effect of an overpass cable-stayed bridge. Numerical results
Live load ZK (0.8 ZK (0.8 UIC) showed that this effect could be ignored when the clearance is lar-
UIC)
Vertical deflection of beam 62.0 L/1400 (250 km/h)  L/1600
ger than 7.25 m and the train speed is lower than 500 km/h. Li [51]
(mm) (350 km/h), L 6 40 m also investigated the aerodynamic effect of different high speed
L/1400 (250 km/h)  L/1900 trains on nearby objectives and proposed a safety distance for var-
(350 km/h), 40 m < L 6 80 m ious trains.
L/1000 (250 km/h)  L/1500
For certain regions, sunlight-induced uneven temperature dis-
(350 km/h), L > 80 m
Relative beam end vertical 60.4 60.2 (cantilever beyond tribution can lead to severe local deformations on bridge deck
rotation in neighboring support 6 0.55 m)
beams (%) 60.3 (0.55 m < cantilever
beyond support 6 0.75 m) Table 4
Beam end horizontal rotation 60.1 60.1 Dynamic design criteria for HSR bridges.
(%)
Contents Criteria
Lateral deflection of beam (mm) L/1400 L/1400 (L 6 80 m)
(L 6 80 m) Fundamental frequency of simple >80/Lu (Lu 6 20 m)
L/1000 L/1000 (L > 80 m) beam (Hz) >23.58Lu0.592 (20 m 6 Lu 6 96 m)
(L > 80 m) Derailment coefficient 60.8
Vertical deflection of track/3 m 1.5 1.5 Wheel loading reducing rate 60.6
(mm) Peak acceleration (g) 0.35 (Ballast track)
Short-term substructure 30 20 0.5 (Ballastless track)
settlement (mm) Vehicle vertical acceleration (g) 60.13
Uneven settlement between 15 5 Vehicle transverse acceleration (g) 60.10
neighboring supports (mm) Sperling’s ride index 62.5 (very good)
Camber after laying track (mm) 20 10 (L 6 50 m) 2.5–2.75 (good)
Min. (L/5000, 20) (L 6 50 m) 2.75–3.0 (satisfactory)
712 B. Yan et al. / Engineering Structures 100 (2015) 707–717

since the ballastless track is attached directly on the bridge deck.
Dai et al. [49] carried out a numerical study to identify the temper-
ature effect on a beam bridge in a central region of China. The
results showed that deformation on the beam under an average
daily maximum temperature around 35 °C is within the allowable
Standard simply supported beam bridge
range but bridges in regions with higher temperature variation
may need additional checks on such temperature effect.
The success of addressing these issues in the past decade has
made great contributions to improve the design and construction
of HSR bridges in China and the train safety under a very high
speed. Overall, two biggest challenge among those issues are the
control of the service limits in practice and the consideration of
the track-structure interaction in analyses.

4. Typical superstructure
3 5 0k m /h 2 50 k m /h

The general design issues discussed above in Section 3 has

Ballastless track Ballasted track
resulted in two typical structure types. In the early 1990s, former
MOR of China launched several comprehensive projects on ten dif-
ferent bridge types [52], including simply supported beam (24 m
24 m 32 m 40 m and 32 m), continuous beams (2  24 m, 3  24 m, 2  32 m,
3  32 m, both formed from simply supported beams and
cast-in-place directly). A large number of experimental tests and
numerical analyses along with economic evaluations were carried
Straight Curve
out to identify the most cost-effective structural form. The results
proved that simply supported beam is the best candidate for
large-scale construction [53,12]. Later, 24 m and 32 m long simply
With noise barriers Without noise barriers supported beam were further experimental studied through a ser-
ies of full-scale tests on several pilot HSR lines [54]. Finally, such
(a) HSR simple beam category type of simply supported beam is widely accepted and heavily
used in the HSR construction. As mentioned in Section 1, over
90% of HSR bridges in China use simply supported beam. For other
short span bridges, continuous beam is used across existing lines
[12]. The design concepts, optimization process and material con-
sumptions of typical short span bridge (simply supported and con-
tinuous) are discussed as follows.

4.1. Simply supported beam

Simply supported beam can be further grouped according to

(b) 32m span HSR simple beam
Fig. 6. A simply supported beam bridge on HSR of China.
design speed, track bed type, main span and alignment, as shown
in Fig. 6a. Precast concrete simple beam with a main span of
32 m (Fig. 6b) is the most frequently used option for the design
of small span HSR bridge and have been proved as the most
cost-effective option among other types [52]. Two typical cross

T
T
700
t

w w
H

w
b b

B
B

(a) Single Cell (b) Double Cell

Fig. 7. Typical cross section of simply supported beam in HSR of China.

 B. Yan et al. / Engineering Structures 100 (2015) 707–717 713

Beam slope sections of simply support beams are shown in Fig. 7. Single-cell
box type is used for all PDLs and partially for freight-passenger
joint lines while double-cell box type is mostly used for the
inter-city rail. Extensive experimental and numerical studies have
been carried out to obtain this structural configuration. For exam-
ple, the cross section of such beam bridge was 13.4 m wide on sev-
eral 350 km/h pilot lines that completed prior to 2008, such as the
Beijing–Tianjin segment, and the Wuhan–Guangzhou segment. In
the year of 2008, the width was reduced to 12.0 m by removing
Fixed bearing Longitudinal movable bearing the maintenance lanes on the both side of the top flange because
high speed trains could cause a safety issue for maintenance engi-
Transverse movable bearing Movable bearing neers [55]. This change has resulted in a width reduction for all
cross-section types and saved construction material.
Fig. 8. Layout of bearings on simple beam.
The bearing layout for a simple beam is given in Fig. 8. The fixed
bearing id set on the downward side when the superstructure has a
slope. The extra cantilever extended beyond the bearings is 0.55 m
Table 5
long for a precast beam and 0.75 m long for a cast-in-place beam.
Uniform-depth continuous beam in HSR of China. The efficiency of a simple beam has been proved based on data
collected from the actual structural response and an economic
Average Span Top Bottom Top Bottom Web
speed (km/ (m) width width flange flange (cm)
evaluation. For example, the test results showed a 32 m precast
h) (m) (m) (cm) (cm) concrete simple beam with ballastless track bed has a fundamental
M S M S M S
frequency of 4.66 Hz, vertical deflection to span ratio of 1/5147,
beam end rotation of 0.075%, and camber for long-term deflection
350 2  24 13.4 6.06 30 45 50 110 30 65
3  24 13.4 6.06 30 45 50 110 30 65
of 5.9 mm [54]. The deflection targets can be achieved along with a
2  32 13.4 5.9 30 45 50 110 30 65 lower cost than other beam types. Economic evaluations showed
3  32 13.4 5.9 30 45 50 110 30 65 that the cost distribution of a simply supported beam is 51.3% for
2  40 13.4 5.74 30 45 50 110 30 65 the superstructure, 34.2% for the foundation, 8.7% for the deck
250 2  24 13.0 5.92 25 25 50 113 30 65 and 5.9% for piers [54]. The material quantities for such simply
3  24 13.0 5.92 25 25 50 113 30 65 supported beam are estimated as follows: 323.2 m3 of concrete,
2  32 13.0 5.86 25 25 50 113 30 65
62.8 t of reinforcing bars and 10 t of prestressing tendons. The
3  32 13.0 5.86 25 25 50 113 30 65
2  40 13.0 5.68 25 25 50 113 30 65 grade of concrete is C50 for the beam (f0 c = 50 MPa) and C40 for
the deck. The Young’s modulus of the reinforcing bar is
Note: M means the mid-span of the beam; S means the beam end of the side span.
2.0  105 MPa [53].

4.2. Continuous beam

T/2 T/2
Even though being a small portion compared to the simply sup-
t

ported beam, continuous beam bridge is another structural form

W
H

W
b
H

B/2
b

Fixed bearing Longitudinal movable bearing

B/2

(a) 1/2 Bearing section (b) 1/2 Mid-span section Transverse movable bearing Movable bearing

Fig. 9. Typical variable-depth continuous beams in HSR of China. Fig. 10. Layout of bearings on standard variable depth continuous beam.

Table 6
Variable-depth continuous beam in HSR of China.

Type Span (m) H (m) B1 (m) B2 (m) D1 (cm) D2 (cm) W (cm)

Mid Side Mid Side Mid Side Mid Side Mid Side
250 km/h ballast track bed 32 + 48 + 32 3.4 2.8 12.2 5.56 5.74 69 69 90 60 95 70
40 + 56 + 40 4.4 2.8 12.2 6.35 5.74 69 59 100 60 110 70
40 + 64 + 40 5.2 2.8 12.2 6.35 5.74 69 59 100 60 976 70
48 + 80 + 48 6.4 3.8 12.2 6.4 6.4 78 78 100 75 100 75
60 + 100 + 60 7.2 4.6 12.2 6.4 6.4 64 69 120 80 100 80
350 km/h ballastless track bed 40 + 56 + 40 4.35 3.05 12.0 7.7 6.7 40 65 80 60 80 60
40 + 64 + 40 6.05 3.05 12.0 7.7 6.7 40 65 80 60 80 60
48 + 80 + 48 6.65 3.85 12.0 7.7 6.7 40 65 100 80 90 60
60 + 100 + 60 7.85 4.85 12.0 7.9 6.7 40 65 120 80 100 80

Note: Mid means the mid-span of the beam; Side means the end of the side span.
714 B. Yan et al. / Engineering Structures 100 (2015) 707–717

for spanning the existing highways in HSR construction to meet The use of larger section depth and increase on the number of pre-
the goal of rapid construction. Continuous beam bridges could stressing tendons are two common ways to improve the cross sec-
have either uniform depth or variable depth. Uniform-depth con- tion stiffness in practice. Compared to a traditional railway bridge
tinuous beams are usually composed of two or three spans, as design, HSR continuous beam has a slightly larger depth. For a vari-
shown in Table 5, but this type is only suitable at sites that are less able depth continuous beam, prestressing tendons are set for both
sensitive to foundation settlement or for the linking segment lateral and vertical directions to control the stress in the cross sec-
between bridges and tunnels. Variable-depth continuous beams tion below the allowable value. Careful design of the bearing layout
(Fig. 9) were further investigated and a series of optimal bench- can reduce the deflection due to temperature and shrinkage, etc. A
mark design were established by main spans, including 48 m, typical bearing layout on a variable-depth continuous beam is
56 m, 64 m, 80 m, 100 m and 128 m [12]. A (32 m + 48 m + 32 m) shown in Fig. 10. Material consumption is given in Table 7.
span arrangement is the most frequently used one in the design
of a continuous beam. For example, such span arrangement is used
more than 30% of 291 continuous beams in the Beijing–Shanghai
A/2 A/2
segment. The side span to main span ratio for a typical continuous
beam varies from 0.6 to 0.7. Table 6 lists the structural configura-
tion of commonly used continuous spans.
The authors participated in a series of design projects on several
variable-depth continuous beam in the Beijing–Shanghai segment.
The biggest lesson learned from this experience is that both
short-term and long-term deflection at the top surface of bridge
deck must be low enough to maintain the smoothness of the track.

H
Table 7
Material requirement per meter on a variable-depth standard continuous beam in
China.

Type Span (m) Concrete Prestressing Reinforcing

(km/h) (m3) tendons (kg) bars (kg)
250 32 + 48 + 32 11.0 638 1886 A A
40 + 56 + 40 12.6 729 2020
40 + 64 + 40 13.4 780 2118
48 + 80 + 48 16.0 849 2397

B
B

60 + 100 + 60 18.7 1039 3044

R

R
350 32 + 48 + 32 11.8 491 1796
40 + 56 + 40 13.5 693 2380 A=800 cm B=300~480 cm R=B/2 A=680 cm B=300~400 cm R=20 cm
40 + 64 + 40 14.6 722 2422
48 + 80 + 48 16.2 822 2941 (a) Round ended (b) Rectangular
60 + 100 + 60 19.6 1027 3115
Fig. 12. Two commonly used piers in HSR of China.

Overhead line column Track structure Noise barrier

140 20 190 250 250 190 20 140

Cover

Water blocking
2% 6% 0% 2% 2% 0% 6% 2%
Cable conduits
Drainage pipe Signal Lines
Drainage pipe

Track centerline Cross-section centerline

(a) Ballastless track bed

140 20 220 250 250 220 20 140

2% 2% 2% 2%

Drainage pipe

Track centerline Cross-section centerline

(b) Ballast track bed

Fig. 11. Bridge deck facilities in HSR of China (Unit: cm).
 B. Yan et al. / Engineering Structures 100 (2015) 707–717 715

(a) Precast and erection

(b) Cast-in-place (c) Mobile erection machine

Fig. 13. Construction of HSR simple beam in China.

4.3. Deck system

Rapid construction of HSR in China is also relied on the use of

typical deck system. Ballastless track bed is normally used for
PDL with higher speed (i.e. faster than 300 km/h) and sometimes
for freight-passenger joint lines with an average speed of
250 km/h. In addition, the layout of bridge decks has slight differ-
ence, on which an open drainage system is adopted on the ballast
deck adopt while the ballastless deck uses a longitudinal open drai-
nage on deck surface and then runoff were conveyed by the drai-
nage system installed inside the structure. Deck surface must be
leveled before being set underneath the track bed. Reserved spaces
(a) Cast-in-place
are required for the installation of reinforcing bars and other aux-
iliary facilities, such as cable conduits, retaining wall (only for bal-
last track bed), noise barrier (areas sensitive to noise), handrail,
overhead line, integrated ground system, emergency access (every
3 km) and inspection equipment [23]. The detailed layout can be
found in Fig. 11.

5. Typical substructure

Typical piers for simple beams and continuous beams by cross

section are composed of round ended, round, rectangular and twin
rectangular types [56,57]. Two commonly used types are round
ended and rectangular piers. Zheng [58] recommended the round (b) Cantilever
ended pier in Fig. 12a for spanning rivers while the twin rectangu-
lar type in Fig. 12b is preferable for short piers (612 m). Portal
frames are sometimes adopted to span existing highway lines.
The selection of HSR bridge foundation depends primarily on
specific soil type at the construction site. Aside from the strength
and stability requirements, careful attentions should be paid for
uneven settlement of the foundation in which to meet the smooth-
ness requirements of HSR. Generally, drilled pile groups are used
with four typical diameters, including 1.0 m, 1.25 m, 1.5 m, and
2.0 m. For other special construction sites, smaller diameter is pre-
ferred for low cost, but the use of larger piles is required when the
length of pile is larger than 50 m. The spacing of piles is commonly (c) Rotation
two times the pile diameter. The thickness of pile caps is ranged
from 2.0 m to 3.5 m. Fig. 14. Common construction methods for HSR continuous beams.
716 B. Yan et al. / Engineering Structures 100 (2015) 707–717
6. Typical construction method
Various construction methods have been developed to address
the challenges on site constraints. A simply supported beam is typ-
ically precast in local factories. Such factories can be found every
28–40 km along the line to support a synchronous construction
process. Then, those beams are put in position by erection machi-
nes, as shown in Fig. 13a. The temperature on the beam surface
during the fabrication and curing should be lower than 60 °C and
the inner/outer temperature difference must be smaller than
15 °C. The position of bearings during the erection should be
leveled and the difference in height between bearings must be
lower than 2 mm. Otherwise, cracks could be triggered by the large
rotation on the beam ends. Other methods for simply beam are
shown in Fig. 13b and c.
Multiple choices are available for erecting HSR continuous
beams as shown in Fig. 14, including cantilever method,
cast-in-place method, launching method and rotation method.
Uniform depth beams are usually built by cast-in-place method
or launching method. Cast-in-place method is a cost-effective
approach that suited for bridges with low clearance in flat terrain
with good soil conditions while launching method saves construc-
tion duration by installing precast girders. The construction of such
structure starts by building multiple simply supported beams and
then connected into a continuous system. The cantilever method is
frequently used for variable depth beams because it offers advan-
tages over other methods on construction ease at sites across exist-
ing lines [59]. In special cases, rotation method is used for bridges
over valleys, rivers and wide existed lines.
7. Summary
China currently has the largest HSR system in the world. Such
efforts are made due to the vision of policy the makers, increased
fiscal investment and, most importantly, the use of an established
national database of typical bridge systems. This paper emphasized
that high serviceability limits are required to design HSR bridges
than a regular railway system in order to guarantee smoother track
surface and better passenger experience. HSR offers safe, afford-
able, green transportation for everyone. Especially in China, HSR
also relieved congestion on highways and runways with a more
convenient travel option with fewer hassles or delays.
This paper presents a brief review on studies in the past decade
on HSR design specifications, and typical structural design and
construction. It can be seen that the 32 m precast simple beam is
used in a majority of bridges built for HSR lines. Standard
cross-section and deck systems were designed and built in a simi-
lar way as an assembly line product. The design of continuous
beam, as a small portion with respect to the total length, is also
completed with a set of typical design such that this structural
form can be used in a versatile way.
Fast development of HSR in China during the past decade stands
right between the mature technologies that were learned from pio-
neers, and the promising future that is projected by newcomers.
Design specifications of HSR in China were updated on a year to
year basis that improved the design and construction practice.
Further investigations will focus on development of high perfor-
mance materials and improvement of maintenance. More opti-
mizations on structural design and understanding the structural
response will provide solutions to other countries developing
HSR, and spark the future design of HSR bridges.
Acknowledgements
The authors gratefully acknowledge the financial support and
first-hand design information provided by China Railway Siyuan
Survey and Design Group CO., LTD. The authors also would like
acknowledge Dr. William C. Taylor, a civil engineering emeritus
professor from Michigan State University for language checking.
References
[1] Taniguchi M. High-speed rail in Japan: a review and evaluation of the
Shinkansen Train. Working Paper no. 103. University of California
Transportation Center; 1992.
[2] Bouley J. Short history of ‘‘high-speed’’ railway in France before the TGV. Jpn
Rail Transport Rev 1994;3:49–51.
[3] European Commission. High-speed Europe, a sustainable link between
citizens. Luxembourg: Publications Office of the European Union, 22; 2010.
[4] Gourvish T. The high speed rail revolution: history and prospects. London: HS2
Ltd.; 2010.
[5] Zuber W. High speed rail in Europe – a three decade success story. Network
2011;73:8–11 [in Chinese].
[6] Harrison J, Dezarn S, Dobbins A, Issac L. Short history of HSR in the USA.
Network 2011;73:17–9.
[7] Amtrak. High-speed rail: a national perspective, high-speed rail experience in
the United States. Austin, Texas: Cambridge Systematics; 2008.
[8] Todorovich P, Schned D, Lane R. High-speed rail international lessons for U.S.
policy makers. Cambridge, USA: Lincoln Institute of Land Policy; 2011.
[9] Albalate D, Bel G. High-speed rail: lessons for policy makers from experiences
abroad. Public Admin Rev 2012;72:336–49.
[10] Australian Department of Infrastructure and Transport. High speed rail study
phase 2 report. Australian Government Report; 2013.
[11] UIC (International Union of Railways). High speed lines in the world. <http://
www.uic.org/spip.php?article573> [retrieved May 2013].
[12] Hu ST, Niu B, Ke ZT, Liu XG. Study on the optimization of standard span length
simply supported box girder for high-speed railway. China Rail Sci
2013;34:15–21 [in Chinese].
[13] Sun SL. Bridge engineering in Beijing–Shanghai high-speed railway. Rail Stand
Des 2008;6:1–4 [in Chinese].
[14] Sun SL. Design and practice of high speed railway bridge. China Railway Press;
2011.
[15] Hu N, Dai GL, Yan B, Liu K. Recent development of design and construction of
medium and long span high-speed railway bridges in China. Eng Struct
2014;74:233–41.
[16] Sun YF, Zhang GA, Wang MJ. Research on technological innovation dynamic
mechanism of railway engineering project. J China Rail Soc 2012;34:76–81 [in
Chinese].
[17] Hu N, Dai GL. The introduction of high-speed railway bridges in Wuhan–
Guangzhou passenger line. In: Proceeding of IABSE symposium 2009 Bangkok;
2009. p. 62.
[18] Xin XZ, Zhang YL, Dai FZ. Live load representative figure in railway design.
China Rail Sci 2006;27:31–6 [in Chinese].
[19] People’s Republic of China Ministry of Railway. Code for design of high speed
railway [in Chinese], TB 10621-2009/J 971-2009. China Railway Press; 2009
[in Chinese].
[20] Dai GL, Liu WS, Li LY. Study of the live load for small- and medium-span
bridges of high-speed railways. China Civil Eng J 2012;10:161–8 [in Chinese].
[21] People’s Republic of China Ministry of Railway. Fundamental code for design
on railway bridge and culvert, TB 10002.1-2005. China Railway Press; 2005 [in
Chinese].
[22] People’s Republic of China Ministry of Railway. Code for seismic design of
railway engineering, GB 50111-2006. China Planning Press; 2006 [in Chinese].
[23] People’s Republic of China Ministry of Railway. Code for design of railway
continuous welded rail. TB10015-2012, Beijing; 2013 [in Chinese].
[24] Han Y, Xia H. Dynamic responses of high-speed railway bridges and running
safety of vehicles during earthquakes. China Rail Sci 2006;27:136–7 [in
Chinese].
[25] Xu QY. Influence of short-wave random irregularity on the dynamic
characteristics of train-slab track-bridge system. China Civil Eng J
2011;44:132–7 [in Chinese].
[26] Zhao PR, Xiao JL, Liu XY. The influence of the displacement at bridge ends to
the fastening systems used on ballastless tracks. J China Rail Soc
2008;30:68–73 [in Chinese].
[27] Wei YH. Research on beam end displacement relevant problems for high-
speed railway ballastless track bridge. Ph.D. Dissertation. China Academy of
Railway Sciences; 2012 [in Chinese].
[28] Luo H. The impacts of creep deformation of long-span prestressed concrete
continuous beam bridges to coupled vibration analysis of train-bridge system
on high-speed railway. Ph.D. Dissertation. Central South University; 2011 [in
Chinese].
[29] He YB. Study on post creep deformation of long-span concrete continuous
bridge with ballastless tracks. J China Rail Soc 2008;30:120–4 [in Chinese].
[30] Bu YZ. Research on the transmission mechanism of longitudinal force for high-
speed railway bridges. Ph.D. Dissertation in Civil Engineering, Southwest
Jiaotong University; 1998 [in Chinese].
[31] Yan B, Dai GL, Zhang HP. Beam-track interaction of high-speed railway bridge
with ballast track. J Cent South Univ Technol 2012;19:1447–53.
 B. Yan et al. / Engineering Structures 100 (2015) 707–717
[32] Yan B, Dai GL. Seismic pounding and reductions of simply-supported beam
considering the interaction of continuously welded rail and bridges. Struct Eng
Int 2013;23:61–7 [in Chinese].
[33] Yang YB, Wu YS. Dynamic stability of trains moving over bridges shaken by
earthquakes. J Sound Vib 2002;258:65–94.
[34] Xia H, Han Y, Zhang N, Guo W. Dynamic analysis of train–bridge system
subjected to non-uniform seismic excitations. Earthquake Eng Struct Dynam
2006;35:1563–79.
[35] Yeh Y, Mo Y, Yang C. Seismic performance of rectangular hollow bridge
columns. J Struct Eng 2002;128:60–8.
[36] Ju SH. Nonlinear analysis of high-speed trains moving on bridges during
earthquakes. Nonlinear Dyn 2012;69:173–83.
[37] Li Y, Conte J. Modeling, simulation, and optimization of California high-speed
rail bridge under earthquakes. In: Catbas FN, Pakzad S, Racic V, Pavic A,
Reynolds P, editors. Topics in dynamics of civil structures, vol. 4. New
York: Springer; 2013. p. 555–69.
[38] Mo Y, Nien I. Seismic performance of hollow high-strength concrete bridge
columns. J Bridge Eng 2002;7:338–49.
[39] Zhang N, Xia H, Guo W. Vehicle–bridge interaction analysis under high-speed
trains. J Sound Vib 2008;309:407–25.
[40] Kwark JW, Choi ES, Kim YJ, Kim BS, Kim SI. Dynamic behavior of two-span
continuous concrete bridges under moving high-speed train. Comput Struct
2004;82:463–74.
[41] Frýba L. A rough assessment of railway bridges for high speed trains. Eng
Struct 2001;23:548–56.
[42] Xia H, Zhang N, De Roeck G. Dynamic analysis of high speed railway bridge
under articulated trains. Comput Struct 2003;81:2467–78.
[43] Xia H, Zhang N. Dynamic analysis of railway bridge under high-speed trains.
Comput Struct 2005;83:1891–901.
[44] Wu Y-S, Yang Y-B, Yau J-D. Three-dimensional analysis of train-rail-bridge
interaction problems. Vehicle Syst Dyn 2001;36:1–35.
[45] Dias R, Goicolea Ruigómez JM, Gabaldón Castillo F, Cuadrado Sanguino M,
Nasarre J, González Requejo P. A study of the lateral dynamic behaviour of high
717
speed railway viaducts and its effect on vehicle ride comfort and stability;
2008.
[46] Yau J-D, Yang Y-B, Kuo S-R. Impact response of high speed rail bridges and
riding comfort of rail cars. Eng Struct 1999;21:836–44.
[47] Yau JD. Response of a train moving on multi-span railway bridges undergoing
ground settlement. Eng Struct 2009;31:2115–22.
[48] Raghunathan RS, Kim HD, Setoguchi T. Aerodynamics of high-speed railway
train. Prog Aerosp Sci 2002;38:469–514.
[49] Dai GL, Zheng PF, Yan B, Xiao XN. Longitudinal force of CWR on box girder
under solar radiation. J Zhejiang Univ (Eng Sci) 2013;47:609–14 [in Chinese].
[50] Chen Y. Study of aerodynamic effect of high-speed passing train and its effect
on cable-stayed bridge crossing the line. M.S. Dissertation. Central South
University; 2013 [in Chinese].
[51] Li RX, Zhao J, Zhang S, Peng YM. Influence of the aerodynamic force to human
body near high-speed trains. China Rail Sci 2007;28:98–104 [in Chinese].
[52] Niu B. Summary of Chinese high-speed railway bridges. In: 18th Proceedings
of China bridges technology collection; 2008 [in Chinese].
[53] Liu JF, Liu CY. Summary of bridges on QinShen passenger dedicated railway
and pondering over bridge construction for high-speed railway. Rail Stand Des
2004;7:134–8 [in Chinese].
[54] Xu SP. Bridges design of Beijing–Shanghai high speed railway between Xuzhou
and Shanghai. Rail Stand Des 2010;7:41–4 [in Chinese].
[55] Dai GL, Hu N, Liu WS. The recent improvement of high-speed railway bridges
in China. In: Proceeding of IABSE 2010 34th symposium, September 22–24,
Venice, Italy; 2010.
[56] Zhu M, Yang YQ, Chen L, Bai QC, He TG. Pier design research on Wuhan–
Guangzhou high-speed railway. Rail Stand Des 2010;1:100–4 [in Chinese].
[57] Zhen JJ. Pier design of high-speed railway standard bridges. Rail Stand Des
2007;2:32–5 [in Chinese].
[58] Zheng J. High speed railway bridges in China. China Higher Education Press;
2008 [in Chinese].
[59] Liu WS, Dai GL, Hu N. Small and medium continuous beams in high-speed
railway of China. J Rail Sci Eng 2010;7:45–51 [in Chinese].