Research Article

Transportation Research Record

2023, Vol. 2677(2) 443–461
Ó National Academy of Sciences:
Assessment of Running Safety of Transportation Research Board 2022
Article reuse guidelines:
Developed Train–Bridge–Wave Model [Link]/journals-permissions
DOI: 10.1177/03611981221106787

Under One-Direction and Two [Link]/home/trr

Bidirectional High-Speed Train Passing

Subjected to Rail Irregularities

Amin Razzaghi Kalajahi1, Morteza Esmaeili1 , and Jabbar Ali Zakeri1

Abstract
A study about the running safety of trains moving over bridges subjected to hydrodynamic loads under one-direction and
two bidirectional high-speed train passing is presented. Dynamic performance of the developed 3D train–bridge–wave system
(TBW model) subjected to different hydrodynamic loads and applying Association of American Railroads track irregularities
is established in this study. The dynamic responses of the bridge, which is under one-direction and two bidirectional high-
speed train passing and subjected to several hydrodynamic loads and rail irregularities, are analyzed. The dynamic responses
of the bridge and the running safety indices of the trains on the bridge under several types of sea states in one-direction and
two bidirectional train passing are analyzed. Results of developed TBW’s sensitivity analyses have shown the importance of
sea-state conditions for safe train running. Irregularities have an obvious effect on the dynamic responses of the bridge.
Between comparing one-direction and two bidirectional high-speed train passing modes, bidirectional train passing has a
greater effect on vertical acceleration and displacement than horizontal ones in the presence of a hydrodynamic load.
Therefore, the combination of wave forces, rail irregularities, and bidirectional train passing jeopardizes the running safety of
high-speed trains by increasing both lateral and vertical displacements and accelerations. Furthermore, the effect of wave
force on the running safety indices of the train is investigated by considering different train speeds. As expected, the tendency
observed in all the derailment criteria is similar, indicating that the risk of derailment increases with the increasing of the run-
ning speed and wave height.

Keyword
high-speed train, bridge, wave, running safety, one direction and bidirectional passing

The rapid development in recent decades of several high- The behavior of bridges is affected by different loads,
speed rail networks around the globe has made it neces- and special attention needs to be paid to this. Dynamic
sary to build new railway lines that meet the strict design bridge–train interaction is a topic that has attracted the
requirements of this type of transport. Among the impor- attention of railway engineers. Many studies have been
tant structures in railroads are bridges that are con- carried out to resolve vibration problems and ensure the
structed with different length and spans over the rail vehicle–bridge coupling system’s performance (1,3–5).
routes and make the passage of rail vehicles possible with Other studies about lateral dynamic behavior of railway
acceptable quality (1). Bridges are among the most
important railway structures that may need to be built in 1
challenging locations; one of these is the construction of School of Railway Engineering, Iran University of Science and Technology,
Tehran, Iran
bridges to safely cross waterways. The river/sea-crossing
bridge is a common infrastructure for the extension of Corresponding Author:
rail from the mainland to islands or coastal areas (2). Morteza Esmaeili, m_esmaeili@[Link]
444 Transportation Research Record 2677(2)

bridges (6) and train–bridge interaction models have various external loads have different properties, the
been done (7–11). A 3D finite element method (FEM), dynamic responses of the bridge and their effects on the
which is employed in Abaqus finite element software to running safety of high-speed trains might be different.
model and analyze the bridge and train while considering In this study, running safety means that acceleration
the interaction between them using the Hertz theory, has and displacement values are allowed with regard to
been presented (1). Chinese and European codes in both vertical and lat-
With the rapid expansion of the infrastructure net- eral directions, which are introduced in next section as
work in past decades, more crossings have been gener- railway bridge safety criteria. This research presents a
ated. Vessel, vehicle, and other collisions are the cause of developed 3D train–bridge–wave (TBW) interaction
many bridge collapse accidents (12, 13). Earthquakes model. A continuous railway bridge with
and strong crosswinds may jeopardize the running safety (32 + 48 + 32) m box girders is considered as a case
of the vehicle, especially when it crossing a viaduct (14). study. This bridge is located in China, and some
Train running safety is a major concern among railway research such as vessel and ice-floe collision has been
engineers, as a derailment may cause significant personal carried out which is appropriate for validation with the
and material damage. This problem becomes more TBW model (13,21). The train used in the validation is
important if the derailment occurs on bridges, especially selected based on the passing train in the field test
at high speeds, where the consequences may be even (1,31). In this paper, a two-lane bridge, high-speed
worse (15,16). Many studies have been carried out in train, and the interactions between them are modeled
relation to the dynamic response of a train–bridge cou- considering one-direction and two bidirectional trains
pling system subjected to crosswinds and earthquakes in passing. Moreover, in the 3D TBW model, hydrody-
the presence of levels of track irregularities (17–20). namic loads with several wave heights (Hs) and periods
Also, in accordance with the importance of trains run- (Ts) are considered according to the Morison equation
ning safely under vessel–bridge or ice–bridge collision, (29). When a hydrodynamic load acts on bridge piers it
numerical analyses and experimental tests have been car- may cause dislocation, uneven deformation, displace-
ried out (12,21,22). ment, or acceleration, which can affect the bridge’s
One topic that has not been considered in railway response to train passing. Several scenarios with vary-
bridge research is the consideration of coupled high- ing wave parameters such as wave height and wave
speed train and bridge system subjected to wave hydro- period (Hs, Ts) have been carried out in this paper.
dynamic load under one-direction and two bidirectional Finally, vertical/lateral acceleration (Ay, Az) and verti-
high-speed train passing. Many structural failures and cal/lateral displacement (Uy, Uz) values of railway
vehicle accidents resulting from extreme waves have been bridge deck are compared with the permissible values
reported in previous studies (23–25). In relation to the with regard to Chinese and European codes, and the
importance of hydrodynamic failures on coastal bridges, values of safe speed according to environmental condi-
experimental tests for a large-scale bridge superstructure tions (wave parameters) are determined in different
model have been done (26). Wave hydrodynamic lateral scenarios.
force components and vibration may jeopardize the vehi- Moreover, track irregularities are an important source
cles running on the deck. Also, when a train crosses over of excitation for both the bridge and the vehicle. The
a bridge, asymmetric loading is applied to the bridge irregularities are deviations of the track from the design
which can laterally move the structure. The moving train geometry, which is one of the most significant load
load on a two-lane bridge can cause lateral displacement amplification factors in railway track systems (32–34).
(27). These displacements and accelerations could be As mentioned, a running train excites the track through
increased with the change in wave load and train speed, the wheel–rail contact mechanisms. Under certain condi-
and may jeopardize the running safety according to tions of track, the load coming from the train could be
domestic or international codes’ criteria. Many studies higher than the static value as a result of irregularities. In
have been conducted to determine the amount of wave this study, the TBW model is developed by applying rail
force applied to marine structures (28–30). For a bridge, irregularities and additional sensitivity analyses are done
the external load considered in the design may be a vessel to evaluate the performance of the high-speed train and
collision, an ice-floe collision, a vehicle passing, a train bridge system under the excitation of wave hydrody-
passing, or hydrodynamic loads. However, little of the namic load and rail irregularities. The fourth quality
current research takes into consideration the wave hydro- level of the Association of American Railroads (34) is
dynamic load on railway bridges under train passing. applied to the TBW model to consider irregularities.
Moreover, sometimes more than one external load, like Also, assessment of the developed TBW model under
one-direction and two bidirectional train passing and two bidirectional high-speed train passing is carried out.
wave hydrodynamic load, may occur simultaneously. As Furthermore, the train running safety is evaluated based
Razzaghi Kalajahi et al 445

on safety criteria such as derailment factor, Prud’homme Table 1. Maximum Allowable Vertical Displacement of
criteria, and wheel offload factor. Finally, a running Bridge (40)
safety chart in the present work is extracted according to
Range of span
different vehicle speeds and wave heights.
This study builds on research originally developed by Designed speed L < 40 m 40 m\L < 80 m L . 80 m
Razzaghi Kalajahi et al., (35) and presents a developed
250 km/h L/1400 L/1400 L/1000
3D TBW interaction model. More details and analyses
300 km/h L/1500 L/1600 L/1100
are provided in this reseach. Some items which are con- 350 km/h L/1600 L/1900 L/1500
sidered in this study are listed as below:

- Several train speeds (200–400 km/h) and several sea

states are considered for one-direction passing iii. The maximum vertical and lateral acceleration of
- Two bidirectional trains running under several sea the deck (Ay, Az) are determined as 0.13 g and
states are analyzed (two trains are modeled) 0.1 g, respectively.
- More details are applied to the model to investigate
running safety indices such as: derailment factor, Also, for better review, other codes such as European
wheel unloading, transversely horizontal force of codes (EN) and ISO criteria have been used to study safe
wheelset (Prud’homme criteria) passing in this research.
- Finally, a running safety chart under several speeds
and sea states is presented in this study. European High-Speed Railway Bridge Criteria

Railway Bridge Safety Criteria i. According to Euro Codes, the maximum allow-
able vertical displacement of the deck (d) is deter-
Among the restrictive criteria for safe passage of high- mined by Figure 1.
speed railway bridges are vertical and lateral accelera- ii. According to Euro Codes, the maximum allow-
tions and displacements of decks, and derailment factors able lateral displacement of the deck (Uz) is
which are referenced in various sources such as Chinese determined by Equation 2.2 according to length
and European codes. So as mentioned, several scenarios of span (L) and maximum allowable horizontal
by varying wave parameters (Hs, Ts) and passing train rotation (r) as shown in Table 2.
with speed of 300 km/h in one-direction and two bidirec-
tional train passing have been addressed in this paper.
Finally, vertical/lateral acceleration (Ay, Az) and vertical/ L2
lateral displacement (Uy, Uz) values of railway bridge are Uz ł ð2:2Þ
8r
compared with the permissible values with regard to
Chinese and European codes (20,36–39), and the values
of safe speed according to sea states (wave parameters) iii. According to Euro Codes and ISO, the maxi-
are determined in these scenarios. mum vertical and lateral acceleration (Ay, Az) are
determined as 1.25 m/s2 for good safety of train
passing on bridges (36,37,39,41,42). Comparing
Chinese High-Speed Railway Bridge Criteria the two codes with each other, it can be seen that
China’s high-speed railway bridge criteria for displace- in general the Chinese code offers more restrictive
ment and acceleration, which have been used to study criteria than the European codes. However, both
safe passing in this research (40), are as follows: codes have been used to better investigate the
present research outcomes in the next sections.
i. The maximum allowable vertical displacement of
the deck is determined by Table 1 for different
speeds.
Running Safety Assessment of Train
ii. The maximum allowable lateral displacement of
the deck (Uz) is determined by Equation 2.1 The running safety assessment of trains is a topic of the
according to length of span (L). utmost importance in railway engineering. Therefore, it
is important to ensure that the criteria used to evaluate
the running safety of vehicles are realistic enough to
L avoid derailments. Railway vehicle derailments are the
Uz ł ð2:1Þ
4000 result of wheels running off the rails that provide the
446 Transportation Research Record 2677(2)

Wheel Unloading
When the vibrations experienced by the vehicle are such
that some wheels lose contact with the rail, a derailment
by wheel unloading may occur. This type of derailment
is caused by severe vibrations of the vehicle in both the
vertical and the lateral directions, caused by track irregu-
larities, earthquakes (20), damping system (of the vehi-
cle) malfunction, and hydrodynamic loads among other
sources. The wheel unloading limit for the ratio between
the reduction of wheel vertical load DQ and the static
load per wheel Q0 is given by:

DQ Q Q
= 0 ð3:1Þ
Figure 1. Maximum allowable vertical displacement of bridge Q0 Q0
(36, 37, 43).
where Q is the dynamic vertical load of the wheel. In a
limit situation, in which a wheel loses contact with the
rail, the dynamic load is null, and the DQ/Q0 ratio
Table 2. Maximum Allowable Lateral Rotation of Bridge
(36, 37, 43)
becomes 1.0. However, to avoid this extreme situation,
the limits to the wheel unloading ratio that can be found
Speed (km/h) Single deck Multi deck bridge in the literature are always less than 1.0 (20). There are
no specific norms in the European codes for the running
V < 120 1,700 3,500 safety assessment of trains during extreme wave condi-
120\V < 200 6,000 9,500
V . 200 14,000 17,500 tions. Therefore, the limit value of 0.9 (20) for the wheel
unloading criterion for the European code and 0.6 (13)
for the Chinese code has been adopted in the present
work.
necessary guidance to the train. According to their
causes, the derailments may be divided into the following
categories: (i) wheel flange climb (derailment factor); Transversely Horizontal Force of Wheelset
(ii) wheel unloading (offload factor); and (iii) transver- The Prud’homme criterion to limit the maximum total
sely horizontal force of wheelset (Prud’homme criteria), lateral force exerted by a wheelset on the track (Ymax) is
or a combination of the above factors. given by:

2Q0
Ymax ł 10 + ð3:2Þ
Derailment Factor 3
Wheel flange climb derailments are the result of excessive where Q0 is the static load per wheel in kilonewtons.
lateral vibrations, vertical vibrations, or both, of the
track that may cause the wheel to climb over the rail, Developed TBW Model
resulting in a reduction of lateral guidance provided by
the rail or a non-functioning damping system of the vehi- Train–Bridge Interaction Model
cle. This kind of derailment generally occurs when the The Chinese high-speed train and bridge models are used
wheel experiences a high level of lateral force combined in this study. A 3D FE model is employed in Abaqus
with a reduction in the vertical contact force on the flan- finite element software to model and analyze the bridge
ging wheel. Phenomena such as earthquakes, strong and train while considering the interaction between them
winds, or hydrodynamic loads combined with high levels using the Hertz theory (44). Next, the Abaqus model is
of track irregularities can contribute to these derail- validated by comparing the lateral acceleration of the
ments. The derailment criterion limits the ratio between bridge under vessel collision load, which was done by
the lateral Y and vertical Q contact force in each wheel Xia et al. (13). The 3D TBW model is then used to assess
to minimize the risk of derailment. The limit value for a sensitivity analysis according to wave height (Hs) varia-
the Y/Q ratio varies from country to country. In the tions (when train speed is considered 300 km/h) to moni-
present study, the limit value of 0.8 proposed by both tor train running safety according to the code’s criteria.
Chinese and European codes is adopted. The train model is composed of locomotive and wagon.
Razzaghi Kalajahi et al 447

Table 3. Train’s Mechanical Properties and Dimensions (13)

Description Name Unit Power car Passenger cars

Car body dimensions s1; s2; q1; q2; h1 m 5.73; 5.73; 6.8; 3.75; 0.75 9; 3.75; 3.75; 0.75
Mass of car body Mc ton 63.98 43.82
Car body inertia moments Jx; Jy; Jz ton.m2 59.4; 2505.3; 2485.4 23.2; 2100; 2080
Mass of bogie Mb ton 3.434 3.04
Bogie inertia moments Jx; Jy; Jz ton.m2 1.766; 2.453; 4.905 1.580; 2.344; 3.934
Secondary suspension stiffness Kz; Ky KN/m 297.2; 1245.87 176; 265
Secondary suspension damping Cz; Cy KNS/m 98.1; 98.1 39.2; 45.12
Secondary suspension dimensions b2; h2 m 1.23; 0.42 1.23; 0.42
Primary suspension stiffness Kz; Ky KN/m 2452.5; 1226.25 2350; 590
Primary suspension damping Cz; Cy KNS/m 98.10; 29.43 58.86; 19.62
Mass of wheel-axle Mw ton 1.776 1.776
Wheel-axle moment Jx; Jy; Jz ton.m2 1.138; 1.138; 0.00785 1.138; 1.138; 0.00785
Primary suspension & wheel b0; b1; h3; t; rw m 0.75; 1; 0.2; 1.25; 0.455 0.75; 1; 0.2; 1.25; 0.455

Note: ‘‘Mass of wheel-axle’’ comprises two wheels and one axle.

Each locomotive or wagon consists of a car body, two separately and is merged with the deck as an integrated
bogies, four wheel-sets, and the spring and damping con- model. Piers and deck are modeled as a 3D deformable
nections between the three components. The car body solid element considering concrete specification. Rails are
configuration and the train’s mechanical properties and modeled as a 3D deformable solid element considering
dimensions are presented according to China’s high- steel specification and tied on deck; also, for considering
speed train in Figure 2 and Table 3, respectively. Also, the interaction between wheels and rails the Hertz theory
the cross-section of the concrete box deck is illustrated in is used (as shown in Figures 5 and 6). Rails and springs-
Figure 3. dampers related to pad and fastener are modeled.
The following assumptions are made in modeling the
train vehicle: iv. The considered bridge is a two-lane continuous
with concrete box (32 + 48 + 32) m (as shown
i. The car body, bogies, and wheel-sets in each in Figure 6). The substructure of the bridge
vehicle are regarded as rigid components, neglect- includes four concrete solid piers with rectangu-
ing their elastic deformation during vibration as lar sections (8 3 4.5) m and piers are fixed at
shown in Figure 4. the seabed. Deck mounted on piers is fixed neo-
ii. The connections between car body, bogies, and prene pot bearings and the bearings are modeled
wheel-sets are represented by linear springs and according to the design of Xia et al. (13). For
viscous dashpots (as shown in Figures 5 and 6). the fixed bearings, the rotational angle about
iii. Each train body has five degrees-of-freedom the transverse axis z of the girder end is free,
(DoFs). They correspond to the lateral displa- whereas the other three translational displace-
cement, the roll displacement, the yaw displace- ments and two rotational angles are connected
ment, the vertical displacement, and the pitch through master-and-slave relations to the pier-
displacement. Each bogie on the vehicle has five top and the damping ratio of the bridge is taken
DoFs: the lateral displacement, the roll displace- as 2.5%. The heights of the piers are assumed to
ment, the yaw displacement, the vertical displa- be 20 m, and sea level, which is proposed as the
cement, and the pitch displacement. For each hydrodynamic force interaction with the piers, is
wheel under the bogie, three DoFs are consid- considered at level +15 m.
ered: the lateral displacement, the roll displace- v. The train passes the bridge with V = 300 km/h.
ment, and the vertical displacement. Thus the vi. Scenarios with and without rail irregularities are
train modeled in TBW for each vehicle with 2- considered in the developed TBW model.
bogies and 4-axles can be modeled by a 27-DoF vii. Single- and double-sided crossing scenarios are
dynamic system, as shown in Figure 6. taken into account.

In relation to other research (11), in longer spans A summary of the train modeling method and the def-
(L . 15 m) the dynamic response is not sensitive to the inition of its interaction with the bridge is provided in the
track stiffness value, so the slab track is not modeled following steps:
448 Transportation Research Record 2677(2)

Figure 2. Car body configuration: (a) side view, (b) top view, and (c) front view (13).

Figure 3. Cross-section of the concrete box deck (13).

Figure 5. Modeling wheel–rails interaction by Hertz spring.

i. Several 3D models were developed with real

dimensions of car body, bogies, and wheels
using 3D discrete rigid shells module in
Abaqus.
ii. Real mass and inertia were assigned for each
part of the wagon and locomotive according to
the car body’s specification (which are shown
Figure 4. Train 3D car body model (locomotive and wagon). in Table 3).
Razzaghi Kalajahi et al 449

Figure 7. Hydrodynamic force on pier.

the movement of fluid could be ignored. Therefore, the

hydrodynamic force (FWave) acting on the column
Figure 6. Train–bridge 3D model.
includes two components (as shown in Equations 4.1–
4.5): one is the inertial force (FI) and the other is the drag
iii. Several springs were modeled to connect differ- force (FD) on the column attributable to the effect of
ent parts of the car body to each other by using viscosity and swirl (as shown in Figure 7).
the interaction module of Abaqus. In this model The Morison equation is adopted to calculate the
several rigid parts were linked in all directions wave forces on the columns, of which the structure
by primary and secondary springs for consider- diameter or width D is smaller than 0.2 times wave
ing suspension stiffness and damping (as shown length L.
in Figures 3 and 4).
iv. All wheels were connected to rails using the FWave = FD + FI ð4:1Þ
interaction module of Abaqus and applying Cd
Hertz theory. FD = r A u2 ð4:2Þ
2
v. In addition to defining displacement and rota-
s

tion boundary conditions, velocity boundary FI = Cm rVu ð4:3Þ

condition in x direction was defined to model
Cd
rAu2 + Cm rVu
s
car body movement in Abaqus to perform FWave = ð4:4Þ
dynamic analysis. 2
vi. To define the developed TBW model, the fourth Here, r is the density of the fluid, V is the volume of
quality level of the Association of American the submerged structure, A is the area of the column sec-
Railroads was applied on rails to consider tion, u and ů are the absolute velocity and acceleration of
irregularities. the fluid, respectively, Cm is the inertia coefficient and Cd
vii. In the two bidirectional train passing mode, two is the drag force coefficient of the fluid. The total hydro-
trains (two locomotives and two wagons) are dynamic force on the unit length (ds) of the column along
modeled and passed in opposite direction. the Z-axis direction can be expressed as:
 2
Sea Hydrodynamic Wave Load F Cd 2 pD s

FHydrodynamic = = rDu + Cm r u ð4:5Þ

ds 2 4
Based on the Morison theory (29), it is assumed that the
effect of fluid on the structure is caused by the accelera- The velocity of the fluid is calculated for several sea
tion field and velocity field, and the effect of structure on states by Equation 4.6. Here, H, T, and L are wave
450 Transportation Research Record 2677(2)

Table 4. Wave Estimation According to Wind Conditions (45)

Wind conditions Wave size

Wind speed Fetch Wind duration Wave height (HS) Wave length (L) Period (TS) Sea states

56 km/h 518 km 23 h 4.1 m 76.5 m 8.6 s I

74 km/h 1,313 km 42 h 8.5 m 136 m 11.4 s II
92 km/h 2,627 km 69 h 14.8 m 212.2 m 14.3 s III

Figure 9. Train–bridge–wave (TBW) 3D model for one-direction

train passing.

Figure 8. Pier lateral movement caused by wave force and first

train passing. train passing jeopardizes the running safety of high-speed
trains by increasing both lateral and vertical displace-
ments and accelerations.
characteristics, d is depth of water, t is time, z is wave Train, bridge, and hydrodynamic load modeling are
crest location along Z-axis and x is wave crest location performed using the FEM in Abaqus software. Sea states
along X-axis. I, II, and III are applied to assess sensitivity analysis. As
mentioned, the substructure of the bridge includes four
!  
cosh 2p concrete solid piers with rectangular sections (8 3 4.5) m
pH L ð d + zÞ 2p 2p
u= cos x  t ð4:6Þ and piers are fixed at the seabed. The height of piers is
T sinh 2pdL
L T
assumed as 20 m and sea level, which is proposed for
As shown in Table 4, three sea states, sea states I, II, hydrodynamic force interaction with piers, is considered
and III, are considered based on three different mean at level +15 m. The wave direction (Z-axis) is assumed
wind speeds UWind of 56, 74, and 92 km/h for the China to be perpendicular to the train passing direction (X-
Sea area. The wave hydrodynamic force is calculated and axis). Also, hydrodynamic load is applied to bridge piers
applied to the Abaqus model as harmonic forces accord- simultaneously and the train passes the bridge with speed
ing to the marine conditions as mentioned in Table 4 and of 300 km/h for the sensitivity analysis. Eight scenarios
Equation 4.5. The assumed values for the coefficients Cd with wave and track irregularities variation have been
and Cm are 1 and 2, respectively (28–30). applied in this study to consider the effect of the presence
or absence of wave and track irregularities for train run-
ning on bridge. The full 3D TBW model is shown in
Developed TBW Model Figure 9. As mentioned, China’s high-speed railway code
When a train crosses over a bridge, asymmetric loading and the European code are used to assess running safety.
is applied to the bridge which can move the structure (as To make a good judgment with respect to the importance
shown in Figure 8). The wave hydrodynamic force is cal- of the time the wave reaches the piers, as the worst-case
culated and applied to the Abaqus TBW model accord- scenario each maximum value of hydrodynamic load is
ing to the marine conditions mentioned in Table 4 and applied on the piers at the time when the train arrives
using Equation 4.5. The combination of wave forces and at the middle of the deck. To study the effect of
Razzaghi Kalajahi et al 451

Figure 10. Pier lateral movement caused by wave force and two
bidirectional train passing.

irregularities on the TBW model and safety of the bridge

to train passing, the fourth quality level of the
Association of American Railroads (34) is applied to the
TBW model. Moreover, four other scenarios with wave
and two bidirectional high-speed trains passing are
applied in this study to consider the effect of two bidirec-
tional trains running on the bridge and compared with
one-direction train passing. The bridge and two bidirec-
tional passing train FE models (developed TBW model)
are shown in Figures 10 and 11.
The developed TBW modeling is presented in a flow- Figure 11. Train–bridge–wave (TBW) 3D model for two
chart as shown in Figure 12 to show the solving metho- bidirectional train passing.
dology for one-direction and two bidirectional train
passing with and without irregularities.
The equations of motion for the train–bridge system
subjected to a collision load can be expressed as:
Model Validation    
 
MVV 0 €xV CVV CVB x€V
The 3D finite element model of train–bridge has already +
0 MBB €xB CBV CBB €xB
been validated with experimental works (1). Moreover,       
the developed 3D finite element model of TBW is vali- KVV KVB xV FVB 0
+ = + ð4:7Þ
dated in this study according to numerical simulation KBV KBB xB FBV FC
results of the train–bridge model (13) considering 10 MN
vessel collision load which is applied on pier 2 at the level where M, C, and K are mass, damping, and stiffness
+10 m above the seabed. The vessel load is taken from matrices of the train–bridge system, and x, x_ and €x are
Xia et al. (13) and represents the collision history of a displacement, velocity, and acceleration vectors, respec-
ship, which is a wide continuous pulse with a total dura- tively; FVB and FBV are interaction forces between vehicle
tion of 1.8 s, as shown in Figure 13. The numerical analy- and bridge, and the subscripts V and B represent vehicle
sis model and the solution method are already verified and bridge, respectively. The components of these
through the comparison between the calculated results matrices and vectors can be found elsewhere (13). FC is
and the in situ measured data (28). Moreover, this the generalized vector of the collision load applied on the
numerical simulation has already been validated with bridge. In the validation stage, the train consists of
several experimental works for railway bridges with and wagons and locomotives considering vessel collision load
without collision loads (12,13,21,31). modeled as shown in Figure 14, and the comparison of
452 Transportation Research Record 2677(2)

Figure 12. Developed train–bridge–wave (TBW) running safety solving methodology.

Table 5. Comparing Lateral Mid-Span Acceleration and

Displacement (Maximum Value)

Mid-span Xia (13) Present study

Lateral acceleration (m/s2) 0.84 0.80

Lateral displacement (mm) 0.54 0.52

of vessel and pier 2, as shown in Figure 14. Mesh size is

one of the effective parameters in FEM analyses. Mesh
configuration can be selected with regard to several con-
ditions. Although smaller elements lead to a higher preci-
sion of the model, they require more time to analyze. In
this research, several analyses were carried out with vari-
ous mesh sizes on the bridge model to determine the opti-
Figure 13. Time histories of vessel collision load (13). mum meshing dimensions. Results did not experience
any considerable changes by decreasing the mesh sizes
smaller than 0.5 m. Consequently, in this study optimum
lateral acceleration time histories at the top of pier 2 mesh size is selected 0.5 m as shown in Figure 14 to
under vessel collision load for a train speed of 200 km/h obtain appropriate train, rail, and deck interactions.
is presented in Figure 15. In the sensitivity analysis, the Moreover, choosing this mesh size is helpful in relation
train consists of one locomotive and one wagon. The axle to including rail and deck connection points. The com-
loads of locomotive and wagon are 19.5 and 14.25 tons, parison with the numerical study demonstrated that the
respectively. Dimensions and mechanical properties of results of the present model were close to the collision
the train car (27-DoFs dynamic system) are described in test results. The accuracy of the results of the two models
Figure 2 and Table 3. Vessel impact is modeled as con- is higher than 92% for maximum values and 87% for
centrated loads which are distributed on the contact level root mean square (RMS), as shown in Figure 15. These
Razzaghi Kalajahi et al 453

8% and 13% differences for maximum and RMS values

could be the result of some of the simplifications consid-
ered in the train–bridge interaction model, input para-
meters and modeling errors, calculation methods, and so
forth. In this study, the time step is taken as 0.005 s. Xia
et al. (13) considered this parameter as 0.0001 s in their
numerical analysis. This is one of the main reasons for
13% difference between the results of the present study
and Xia et al.’s (13) results. Moreover, comparison of lat-
eral mid-span acceleration and displacement (maximum
value) between the two models is presented in Table 5.

Sensitivity Analyses of Developed TBW

Model Under Several Conditions
The effects of some important parameters on the running
Figure 14. 3D validation model for considering vessel collision safety of a high-speed train on a bridge subjected to
load. hydrodynamic load can be assessed using this 3D vali-
dated model. First, eight scenarios as shown in Table 6
are applied according to assuming various conditions for
sea states and track irregularities. These sea states are
considered based on the China Sea area and track irregu-
larities are considered based on fourth quality level of
the Association of American Railroads (as shown in
Figure 16). Moreover, four other scenarios with hydro-
dynamic load and two bidirectional high-speed trains
passing have been applied in this study for considering
the effect of two bidirectional trains running on the
bridge.
In all cases with hydrodynamic force, the time history
of the sea states is applied on piers at the level +15 m
above the seabed. The time histories of several sea states
Figure 15. Comparing lateral acceleration time histories at the are obtained from Equation 4.6.
top of pier 2 under vessel collision load for a train speed of This section investigates the dynamic response of the
200 km/h. bridge during the one-direction and two bidirectional

Table 6. Scenarios for Wave and Irregularities Variation in One-Directional and Bidirectional Passing Modes

Scenario Irregularities Hs (m) Ts (s) Train passing mode

1 NA 0 0 One directional
2 NA 4.1 8.6 One directional
3 NA 8.5 11.4 One directional
4 NA 14.8 14.3 One directional
5 4th quality level of AAR 0 0 One directional
6 4th quality level of AAR 4.1 8.6 One directional
7 4th quality level of AAR 8.5 11.4 One directional
8 4th quality level of AAR 14.8 14.3 One directional
9 NA 0 0 Bidirectional
10 NA 4.1 8.6 Bidirectional
11 NA 8.5 11.4 Bidirectional
12 NA 14.8 14.3 Bidirectional

Note: AAR = Association of American Railroads.

454 Transportation Research Record 2677(2)

displacement under train passing (which is less than

1 mm) is less than the permissible value. In the following,
by applying hydrodynamic loads on piers and increasing
the height of the wave, the lateral displacement of the
deck increases and exceeds the permissible values of the
European and Chinese codes. Lateral displacement
response of the bridge subjected to stormy hydrody-
namic load (H = 14.8 m) could be up to 26 times higher
than that without hydrodynamic load. As shown in
Figure 16. The track profile irregularities for line level 4 (34). Figure 18, by applying hydrodynamic loads on piers and
increasing the height of the wave, the vertical displace-
ment of the deck increases but does not exceeds the per-
missible values of the codes. However, because of the
increasing trend of displacement with increasing wave
height, the displacement will approach critical values (a
negative value indicates downward vertical displace-
ment). The vertical displacement response of the bridge
subjected to stormy hydrodynamic load (H = 14.8 m) is
40% higher than that without hydrodynamic load.
When the train is passing on the bridge without hydro-
dynamic force (wave height = 0), the amount of lateral
acceleration under train passing is less than the permissi-
ble value (which is negligible). By applying hydrody-
namic loads on the piers and increasing the height of the
wave, the lateral acceleration of the deck increases and
exceeds the permissible values of the European and
Chinese codes (as shown in Figure 19). As shown in
Figure 20, when the train is passing on the bridge with-
out hydrodynamic force (wave height = 0), the amount
of vertical acceleration under train passing is 0.8 m/s2,
which is less than the permissible value. In the follow-
Figure 17. Lateral displacement of the bridge deck with and ing, by applying hydrodynamic loads on the piers and
without irregularities against several sea states. increasing the height of the wave, the vertical accelera-
tion of the deck increases and exceeds the permissible
train passing with speed of 300 km/h through the bridge values when the wave height exceeds 8 m. Also, the
under several sea states by time history analyses. The struc- intensity of the acceleration has increased after the 4.1-
tural dynamic response discussed in this study includes the meter wave, which indicates that after this wave height,
lateral displacement (Uz), vertical displacement (Uy), and the bridge will be more sensitive to changes.
lateral and vertical acceleration (Az, Ay) response at the
middle point of the span. The performance of the bridge
structure and the train (developed TBW model) under dif- Effect of Irregularities on TBW Model
ferent sea states is carefully discussed. As mentioned, track irregularities are an important
source of excitation for both the bridge and the vehicle.
To study the effect of irregularities on the TBW model
Effect of Hydrodynamic Load on Railway Bridge (One-
and safety of the bridge to train passing, the fourth qual-
Direction Train Passing) ity level of Association of American Railroads (34) is
First, to study the effect of the waves on the safety of the applied on the TBW model (as shown in Figure 16). The
bridge to one-direction train passing, analyses are simu- bridge’s behavior, such as vertical and lateral displace-
lated with and without hydrodynamic loads, and the ments and accelerations, is compared when the train
bridge’s behavior, such as vertical and lateral displace- passes at a speed of 300 km/h.
ments and accelerations, is compared when the train As shown in Figures 17 to 20, when the train is passing
passes at a speed of 300 km/h. As shown in Figure 17, on the bridge, the amount of lateral and vertical displace-
when the train is passing on the bridge without hydrody- ments and accelerations under train passing are increased
namic force (wave height = 0), the amount of lateral by applying track profile irregularities.
Razzaghi Kalajahi et al 455

Figure 18. Vertical displacement of the bridge deck with and

without irregularities against several sea states.
Figure 20. Vertical acceleration (m/s2) of the bridge deck with
and without irregularities against several sea states.

impact than track irregularities in these cases. Variation

of vertical acceleration and displacement is increased
considerable by applying irregularities (about 10% for
acceleration and 5% for displacement), but the amount
of variation is decreased when the hydrodynamic load is
increased. This is because the hydrodynamic load has a
greater impact than track irregularities. Generally, irre-
gularities have a greater effect on vertical acceleration
and displacement than horizontal ones in the presence of
hydrodynamic load. The combination of these two phe-
nomena (wave and irregularities) jeopardizes the running
safety of the train when it is crossing the bridge.

Results of Two Bidirectional Train Passing Mode on

Figure 19. Lateral acceleration (m/s2) of the bridge deck with
and without irregularities against several sea states. Developed TBW Model
In the two bidirectional passing mode with opposite
direction and the same speed, the maximum accelerations
Figure 21 shows variation of the TBW’s responses and displacements occur at the bridge mid-span. The
when irregularities are applied on the model. By applying worst-case scenario occurs when both trains are near the
irregularities on the model without hydrodynamic load mid-span; this is obtained from several analysis results.
(wave height = 0), and the train is passing on the bridge, In this section, the results of accelerations and displace-
the amount of lateral acceleration increases more than ments at the middle of the deck subjected to parallel train
30% (about 32% for RMS and 35% for maximum passing in opposite directions are presented (as shown in
value). Generally RMS is an appropriate method for Figures 22–25).
variable quantities, but maximum value could be useful Figures 22 to 25 show variation of the developed
for running safety assessment. By applying irregularities TBW’s responses in two bidirectional train passing
on the model with application of hydrodynamic load against one-direction train passing. As shown in Figures
(wave height = 4.1, 8.5, 14.8 m), variation of lateral 22 to 25, by applying two bidirectional train passing on
acceleration and displacement are not increased consid- the model with and without applying hydrodynamic load
erable. This is because hydrodynamic load has a greater (wave height = 0, 4.1, 8.5, 14.8 m), lateral displacements
456 Transportation Research Record 2677(2)

Figure 21. Variation of train–bridge–wave’s responses when excited by irregularities: (a) wave height = 0, (b) wave height = 4.1 m,
(c) wave height = 8.5 m, and (d) wave height = 14.8 m.
Note: RMS = root mean square; MAX = maximum.

Figure 22. Assessment of lateral displacement under one- Figure 23. Assessment of vertical displacement under one-
direction and two bidirectional high-speed train passing. direction and two bidirectional high-speed train passing.
Razzaghi Kalajahi et al 457

load. The combination of wave forces and two bidirec-

tional train passing jeopardizes the running safety of
high-speed trains when they are crossing the bridge.

Train Running Safety Analysis

Assessing Train Running Safety Under Several Sea
States (Vtrain = 300 Km/H)
To study the effect of the waves on the train running
safety, analyses were simulated with and without hydro-
dynamic loads and the running safety factors, such as
derailment factor, unloading factor, and Prud’homme
criteria were compared when the train passes at a speed
of 300 km/h. Figure 26 shown the derailment factors, off-
load factors, and lateral wheel/rail forces of the car when
the train travels on the bridge at V = 300 km/h, without
hydrodynamic load and under three sea states. It can be
Figure 24. Assessment of lateral acceleration under one- seen from the figure that the running safety indices of the
direction and two bidirectional high-speed train passing. train are affected by the wave loads. Under the hydrody-
namic loads, the maximum values for the running safety
indices are amplified. The derailment factor increases
and exceeds the permissible values of the European and
Chinese codes in sea states II and III. The offload factor
increases and exceeds the permissible values of the
Chinese code in all sea states and exceeds the permissible
values of the European code in sea states II and III.
When the bridge is excited by any of the three hydrody-
namic loads, the lateral wheel/rail force (Prud’homme
criteria) of the train running at V = 300 km/h are within
the corresponding safety allowances in the Chinese and
European codes.

Assessing Train Running Safety Under Several Sea

States and Speeds
In this section, with the same bridge and car bodies, the
effect of wave force on the running safety of train is
investigated by considering different train speeds. From
Figure 27 the increasing tendency of the train running
safety indices with the train speed is obvious. Generally,
Figure 25. Assessment of vertical acceleration under one- the higher the train speed, the bigger the running safety
direction and two bidirectional high-speed train passing. indices. When V = 400 km/h, the derailment factor and
offload factor under all sea states exceed the allowable
values, and when the train is running at V = 200 km/h
are decreased but vertical displacements are increased these factors are within the corresponding safety allow-
considerably. These variations happen because two bidir- ances in the both Chinese and European codes. Also,
ectional high-speed train passing generates antisym- when the bridge is excited by any of the three hydrody-
metric loads in these cases. Also, by applying two namic loads, the lateral wheel/rail force (Prud’homme
bidirectional train passing on the model, in all cases lat- criteria) for most conditions is within the corresponding
eral accelerations are decreased but vertical accelerations safety allowances in the both Chinese and European
are increased considerably, as shown in Figures 24 and codes (except for V = 400 km/h under sea state III).
25. Generally, two bidirectional train passing has a The global envelope of each of the analyzed safety cri-
greater effect on vertical acceleration and displacement teria, as function of the running speed of the vehicle and
than horizontal ones in the presence of a hydrodynamic of the wave height, considered in the present work, is
458 Transportation Research Record 2677(2)

Figure 26. Assessment of train running safety factors under V = 300 km/h: (a) derailment factor, (b) offload factor, and (c) Prud’homme
criteria.

plotted in Figure 28. As expected, the tendency observed taking a continuous sea-crossing bridge as a case study,
in all the criteria is similar, indicating that the risk of the dynamic responses of the bridge at the mid-span of
derailment increases with the increasing of the running bridge which is under high-speed train passing and sub-
speed and wave height. Derailment factor and wheel jected to several hydrodynamic loads were analyzed. An
unloading criteria define the safety zone, as may be seen assessment procedure for the running safety of one-
in Figure 28. Therefore, the vehicle remains safe at every direction and two bidirectional high-speed trains on a
speed if no wave occurs, whereas for the other cases, the bridge subjected to hydrodynamic load is proposed. The
derailment risk is predicted with the wheel unloading cri- following conclusions can be drawn from the sensitivity
terion and with the derailment factor. analyses of the developed TBW model under several sea
states and track irregularities in both one-direction and
two bidirectional train passing modes.
Conclusion As shown, sea wave has an obvious effect on the
A study in relation to the running safety assessment of dynamic responses of the railway bridge. Both lateral
trains moving over bridges subjected to hydrodynamic and vertical displacements and accelerations responses of
loads is presented. This research was conducted to inves- the bridge subjected to hydrodynamic load are much
tigate the response of a two-lane sea-crossing bridge sub- greater than those without hydrodynamic load. The lat-
jected to high-speed trains in different passing modes eral displacement and acceleration of the bridge are more
such as one-direction passing, two bidirectional passing influenced by hydrodynamic load, and vertical displace-
in opposite direction and same speeds. The dynamic ment and acceleration of the bridge are more influenced
behavior of the 3D train–bridge system subjected to dif- by track irregularities. By applying hydrodynamic loads
ferent hydrodynamic loads (TBW model) was carried out on piers and increasing the height of the wave, the lateral
under one-direction and two bidirectional train passing. displacement and acceleration values of the deck increase
Additional analyses were done by applying Association and exceed the permissible values of the European and
of American Railroads irregularities to the model. By Chinese codes.
Razzaghi Kalajahi et al 459

Figure 27. Assessment of train running safety factors under several speeds and sea states: (a) derailment factor, (b) offload factor, and
(c) Prud’homme criteria.

The running safety of the train is affected by both the

type of sea state and track irregularities. Strong waves
may threaten the running safety of high-speed trains. The
effect of irregularities on the dynamic responses of the
bridge is obvious. By applying irregularities on the model
without hydrodynamic load (wave height = 0), and train
passing on bridge, the amount of lateral acceleration is
increased more than 30%. Also, by applying irregulari-
ties to the model with hydrodynamic load, variation of
lateral acceleration and displacement are not increased
considerably, because the hydrodynamic load has a
greater impact than track irregularities in these cases.
Therefore, it can be concluded that irregularities have a
greater effect on vertical acceleration and displacement
than horizontal ones in the presence of a hydrodynamic
load, and the combination of these two phenomena
(wave and irregularities) jeopardizes the running safety of
train when it is crossing the bridge.
As another conclusion between one-direction and two
bidirectional train passing modes, two bidirectional train
Figure 28. Running safety charts obtained for several conditions. passing has greater effect on vertical acceleration and
460 Transportation Research Record 2677(2)

displacement than horizontal ones in presence of hydro- Mechanics, Vol. 64, No. 1, 2017, pp. 33–44. [Link]
dynamic load. Therefore, the combination of wave forces org/10.12989/sem.2017.64.1.033.
and two bidirectional train passing jeopardizes the run- 2. Fang, C., Y. Li, K. Wei, J. Zhang, and C. Liang. Vehicle–
ning safety of high-speed trains by increasing the both Bridge Coupling Dynamic Response of Sea-Crossing Rail-
lateral and vertical displacements and accelerations of way Bridge Under Correlated Wind and Wave Conditions.
Advances in Structural Engineering, Vol. 22, No. 4, 2019,
the deck and exceeding the permissible values of the
pp. 893–906. [Link]
European and Chinese codes. Furthermore, the effect of
3. Frýba, L. Dynamics of Railway Bridges. Thomas Telford
wave force on the running safety of trains is investigated Publishing, London, 1996.
by considering different train speeds. Risk of derailment 4. Liu, X. C., B. L. Wang, L. C. Yang, and Y. Jia. Finite Ele-
increases with the increasing of the running speed and ment Analysis of Pier Top Horizontal Displacement
wave height. Running safety indices of the train are Caused by Newly-Built Road Under High-Speed Railway
affected by the wave loads. Under the hydrodynamic Bridge. Applied Mechanics and Materials, Vol. 353–356,
loads, the maximum values for the running safety indices 2013, pp. 823–827. [Link]
such as derailment factor, offload factor, and net/AMM.353-356.823.
Prud’homme criteria are amplified. Derailment factor 5. Museros, P., M. Romero, A. Poy, and E. Alarcón.
and wheel unloading criteria define the safety zone in the Advances in the Analysis of Short Span Railway Bridges
for High-Speed Lines. Computers & Structures, Vol. 80,
running safety chart, and the vehicle remains safe at
No. 27–30, 2002, pp. 2121–2132.
every speed if no wave occurs, whereas for the other
6. Dias, R. F. da S. A Study of the Lateral Dynamic Beha-
cases the derailment risk predicted with the wheel viour of High Speed Railway Viaducts and its Effect on
unloading criterion and with the derailment factor. Vehicle Ride Comfort and Stability. 2008. [Link]
10.1201/9781439828434.ch82
Acknowledgments 7. Liu, K., G. De Roeck, and G. Lombaert. The Effect of
Dynamic Train-Bridge Interaction on the Bridge Response
We would like to show our gratitude to Professor He Xia and During a Train Passage. Journal of Sound and Vibration,
Professor Nan Zhang from Beijing Jiao tong University for Vol. 325, No. 1–2, 2009, pp. 240–251.
sharing their data for China’s High-Speed Railway Bridge spec- 8. Hughes, S. A Critical Analyis of the Øresund Bridge, Con-
ifications with us during the course of this research, and we necting Copenhagen to Malmö. Transportation Research
thank them for their insights.
Record: Journal of the Transportation Research Board,
2010. 1654: 133–140.
Author Contributions 9. Bargi, K., and A. Aghabozorgi. Dynamic Analysis of High
Speed Train Passing on Bridge. Tehran University, Civil
The authors confirm contribution to the paper as follows: study
Department, Iran, 2010.
conception and design: A. Razzaghi Kalajahi, M. Esmaeili, J.A.
10. Varandas, J., P. Hölscher, and M. Silva. Dynamic Beha-
Zakeri; data collection: A. Razzaghi Kalajahi, M. Esmaeili;
viour of Railway Tracks on Transition Zones. Computers
analysis and interpretation of results: A. Razzaghi Kalajahi, M.
& Structures, Vol. 89, No. 13–14, 2011, pp. 1468–1479.
Esmaeili; draft manuscript preparation: A. Razzaghi Kalajahi.
11. Rashid, S. Parametric Study of Bridge Response to High
All authors reviewed the results and approved the final version
Speed Trains. PHD thesis. KTH Royal Institute of Tech-
of the manuscript.
nology, 2011, p. 104.
12. Xia, C. Y., J. Q. Lei, N. Zhang, H. Xia, and G. De Roeck.
Declaration of Conflicting Interests Dynamic Analysis of a Coupled High-Speed Train and
The author(s) declared no potential conflicts of interest with respect Bridge System Subjected to Collision Load. Journal of
to the research, authorship, and/or publication of this article. Sound and Vibration, Vol. 331, No. 10, 2012,
pp. 2334–2347. [Link]
13. Xia, C. Y., H. Xia, and G. De Roeck. Dynamic Response
Funding of a Train-Bridge System Under Collision Loads and Run-
The author(s) received no financial support for the research, ning Safety Evaluation of High-Speed Trains. Computers
authorship, and/or publication of this article. and Structures, Vol. 140, 2014, pp. 23–38. [Link]
10.1016/[Link].2014.04.010.
14. Arvidsson, T., A. Andersson, and R. Karoumi. Train Run-
ORCID iD ning Safety on Non-Ballasted Bridges. International Journal
Morteza Esmaeili [Link] of Rail Transportation, Vol. 7, No. 1, 2019, pp. 1–22. https://
[Link]/[Link]
15. Montenegro, P. A., R. Calcxada, N. Vila Pouca, and M.
References Tanabe. Running Safety Assessment of Trains Moving
1. Jahangiri, M., and J. A. Zakeri. Dynamic Analysis of Over Bridges Subjected to Moderate Earthquakes. Earth-
Train-Bridge System Under One-Way and Two-Way quake Engineering & Structural Dynamics, Vol. 45, No. 3,
High-Speed Train Passing. Structural Engineering and 2016, pp. 483–504. [Link]
Razzaghi Kalajahi et al 461

16. Montenegro, P. A., H. Carvalho, D. Ribeiro, R. Calcxada, Earthquakes University of Science and Technology Beijing
M. Tokunaga, M. Tanabe, and W. M. Zhai. Assessment of Main Contents. PhD thesis. School of Civil and Environ-
Train Running Safety on Bridges: A Literature Review. mental Engineering, University of Science and Technology,
Engineering Structures, Vol. 241, 2021, P. 112425. https:// Beijing, 2010.
[Link]/10.1016/[Link].2021.112425. 31. Xia, H., and N. Zhang. Dynamic Analysis of Railway
17. Xu, Y. L., N. Zhang, and H. Xia. Vibration of Coupled Bridge Under High-Speed Trains. Computers and Struc-
Train and Cable-Stayed Bridge Systems in Cross Winds. tures, Vol. 83, No. 23–24, 2005, pp. 1891–1901. [Link]
Engineering Structures, Vol. 26, No. 10, 2004, pp. 1389–1406. org/10.1016/[Link].2005.02.014.
18. Li, Y., S. Qiang, H. Liao, and Y. Xu. Dynamics of Wind– 32. Jenks, C. W. Rail Corrugation Mitigation in Transit. Transit
Rail Vehicle–Bridge Systems. Journal of Wind Engineering Cooperative Research Program. Sponsored by the Federal
and Industrial Aerodynamics, Vol. 93, No. 6, 2005, Transit Administration, 1998. [Link]
pp. 483–507. tions/tcrp/tcrp_rrd_26.pdf.
19. Alexander, N. Multi-Support Excitation of Single Span 33. Grassie, S. L. Rail Corrugation: Characteristics, Causes,
Bridges, Using Real Seismic Ground Motion Recorded at and Treatments. Proceedings of the Institution of Mechani-
the SMART-1 Array. Computers & Structures, Vol. 86, No. cal Engineers, Part F: Journal of Rail and Rapid Transit,
1–2, 2008, pp. 88–103. Vol. 223, No. 6, 2009, pp. 581–596. [Link]
20. Montenegro e P. Almeida A Methodology for the Assessment 09544097JRRT264.
of the Train Running Safety on Bridges. PhD thesis. Faculty 34. Lei, X. High Speed Railway Track Dynamics. Models, Algo-
of Engineering of the University of Porto, 2015, p. 292. rithms and Applications. Science Press, Beijing/Springer
21. Xia, C., J. Ma, and H. Xia. Dynamic Analysis of a Train- Nature Singapore Pte. Ltd., 2017.
Bridge System to Vessel Collision and Running Safety of 35. Razzaghi Kalajahi, A., M. Esmaeili, and J. A. Zakeri.
High-Speed Trains. Vibroengineering Procedia, Vol. 5, Dynamic Stability Analysis of High-Speed Trains Moving
2015, pp. 509–514. Over Sea-Cross Bridge Subjected to Rail Irregularities.
22. Li, Y., J. Deng, B. Wang, and C. Yu. Running Safety of International Journal of Coastal and Offshore Engineering,
Trains Under Vessel-Bridge Collision. Shock and Vibration, Vol. 5, No. 2, 2021, pp. 56–69.
Vol. 2015, 2015, pp. 1–11. [Link] 36. CEN. EN 1990-Eurocode - Basis of Structural Design. 2005.
252574. 37. CEN. EN 1991- Eurocode: Actions on Structures. 2005.
23. Ataei, N., M. Stearns, and J. E. Padgett. Response Sensi- 38. CEN. EN 1998- General Rules, Seismic Actions and Rules
tivity for Probabilistic Damage Assessment of Coastal for Buildings. 2005.
Bridges Under Surge and Wave Loading. Transportation 39. Jeon, B. G., N. S. Kim, and S. Kim, Il. Estimation of the
Research Record: Journal of the Transportation Research Vibration Serviceability Deflection Limit of a High-Speed
Board, 2010. 2202: 93–101. Railway Bridge Considering the Bridge-Train Interaction
24. Kitada, T. Considerations on Recent Trends in, and Future and Travel Speed. KSCE Journal of Civil Engineering, Vol.
Prospects of, Steel Bridge Construction in Japan. Journal 20, No. 2, 2016, pp. 747–761. [Link]
of Constructional Steel Research, Vol. 62, No. 11, 2006, s12205-015-0565-z.
pp. 1192–1198. 40. Code for Design of High-Speed Railway (China). Ministry
25. Robertson, I. N., H. R. Riggs, S. C. Yim, and Y. L. Young. of Railways of the People’s Republic of China, 2009.
Lessons From Hurricane Katrina Storm Surge on Bridges 41. ISO. ISO 2631- Mechanical Vibration and Shock- Evalua-
and Buildings. Journal of Waterway, Port, Coastal, and tion of Human Exposure to Whole-Body Vibration. 2002.
Ocean Engineering, Vol. 133, No. 6, 2007, pp. 463–483. 42. Jiang, Y., B. K. Chen, and C. Thompson. A Comparison
26. Bradner, C., T. Schumacher, A. M. Asce, D. Cox, C. Hig- Study of Ride Comfort Indices Between Sperling’s Method
gins, and M. Asce. Experimental Setup for a Large-Scale and EN 12299. International Journal of Rail Transportation,
Bridge Superstructure Model Subjected to Waves. Journal Vol. 7, No. 4, 2019, pp. 279–296. [Link]
of Waterway, Port, and Coastal Engineering, Vol. 137, No. 23248378.2019.1616329.
1, 2011, pp. 3–11. [Link] WW. 43. Bouassida, Y., E. Bouchon, P. Crespo, P. Croce, L.
1943-5460.0000059. Davaine, S. Denton, M. Feldmann, et al. Bridge Design to
27. Cuadrado Sanguino, M., P. Gonz ález Requejo, J. M. Goi- Eurocodes. EUR 25193 EN. Publications Office of the Eur-
colea Ruigómez, J. Nasarre, and R. Dias. Analysis of Lat- opean Union, Luxembourg, 2012.
eral Displacements in Large Railway Viaducts Under 44. Bhaskar, A., K. L. Johnson, G. D. Wood, and J. Wood-
Traffic Loads. Impact on Ride Safety and Passenger Com- house. Wheel-Rail Dynamics With Closely Conformal
fort. Proc., 8th World Congress on Railway Research, Seoul, Contact, Part 1: Dynamic Modelling and Stability Analy-
Korea, 2008, pp. 1–12. sis. Proceedings of the Institution of Mechanical Engineers,
28. Chakrabarti, S. Discussion of ‘‘Nondeterministic Analysis Part F: Journal of Rail and Rapid Transit, Vol. 211, No. 1,
of Offshore Structures’’. Journal of the Engineering 1997, pp. 11–26.
Mechanics Division, Vol. 97, No. 3, 1971, pp. 1028–1029. 45. Cao, H. J., F. Yi, and W. B. Feng. Estimate of the Extreme
29. Sorensen, R. M. Basic Wave Mechanics for Ocean Engi- Wave Height in the South China Sea Using GPD Method.
neering. John Wiley, New York, NY, 1993. IOP Conference Series: Earth and Environmental Science,
30. Song, P. B. Study on the Calculation Method for Hydrody- Vol. 189, No. 5, 2018, P. 052035. [Link]
namic Pressure of Bridge Piers in Deep Water Under 1755-1315/189/5/052035.