PERFORMANCE ASSESMENT SOLUTIONS FOR TRANSITION ZONES

EMBANKMENT-BRIDGE RAILWAYS THROUGH 3D NUMERICAL SIMULATIONS

IVO SEARA*
Supervisor: A. Gomes Correia

University of Minho
School of Engineering
Centre for Territory, Environment and Construction (C-TAC)

* ivoseara@civil.uminho.pt

Introduction
Transition zones are areas where a railway track exhibits abrupt
changes in vertical stiffness, such as bridge approach (fig. 1). These
abrupt changes in vertical stiffness in short distances can present
serious problems when subject dynamic loads which can generate
impacts that contribute to accelerated degradation of the track and
therefore its frequently maintenance and decrease of design life.
Problems at track transitions
can be divided into three
categories:
i) differential settlements;
ii) track stiffness case;
iii) track damping case.

The objective was to find the best ratio of stiffness between the two
inverted wedges, identifying the main differences between them.
Numerical simulations were carried out using a point moving load at
314km/h. Taking into account the assumptions made the principle of
superposition of effects was applicable to simulate the passage of the
Thalys train (fig. 2) .

Figure 2: Thalys train configuration

Figure 1: Variation of vertical stiffness

along the railway track
Based on an international survey several solutions have been
proposed for transition zones. These proposed solutions reported in
Seara and Gomes Correia (2008) have been implemented to promote
a smooth change in vertical stiffness between embankments and
bridges. However, it noticed that several of these solutions perform
below expectation. Thus, numerical simulations of the various
solutions could be a first step towards to a better understanding.

The railway track model incorporated rail, interface, sleeper, ballast,

sub-ballast, sub grade, embankment and technical block in transition
zone composed by two inverted wedges, soil and abutment. The
longitudinal length of the model was 69,9m and the transversal length
was 20,0m. With this geometry it was obtained a mesh composed by
91231 nodes and 95543 elements. A global schematic representation
of the railway track is presented in Figure 3, and its schematic
geometry is presented in Figure 4.

Objectives

Figure 3: Global view of numerical models

The work herewith is part of an ongoing Doctor thesis Study of the

Behavior of Transition Zones in High-Speed Railways.
The present study aims to understand the performance of one of the
conventional solutions for transition zones under a moving train from
the embankment section, through the transition zone and the bridge
deck.

Case Study 3D Numerical Models

The case study is a transition zone composed by two reversed
wedges (fig. 3 and 4). 3D numerical dynamic linear elastic models in
a finite element software DIANA were used for the numerical
simulations (train/track (wheel/rail) interaction wasnt considered).

3.3m

17.4m

45.6m

Sleeper
Ballast
2

2.85m

Wedge 1

14.4m

Embankment

1
1

Wedge 2

20.1m

2.06m

Embankment

/
Soil

Abutment

Abutment
6.19m

Soil

32.4m

a) Longitudinal
b) Transversal
9.27m
5.20m
Figure 4: Identification of the dimensions of the 5.53m
numerical
models
The necessary parameters to characterize the behaviour of the
materials were: Youngs modulus; Poissons ratio; density; dampings
ratio (Rayleigh damping was considered). In total three numerical
simulations were performed, whose main differences were:
- Model 1: E(Embankment) = 60MPa; rE1/E2 = E1 / E2 = 1,195
- Model 2: E(Embankment) = 60MPa; rE1/E2 = E1 / E2 = 2,0
- Model 3: E(Embankment) = 100MPa; rE1/E2 = E1 / E2 = 2,0

Results and Discussion

The results obtained for a point load with a value of 100kN moving at
314km/h, from the embankment to the bridge, is presented in figure 5.
Observing the results in this
figure it is possible to identify
important differences between
settlements on embankment
zone and on abutment/bridge
deck. Through these results it is
possible to conclude that when
the embankment has a Youngs
modulus with 100MPa, the track
exhibits a too large and
undesirable stiffness.

Figure 5: Settlements along the

track at each point of the
rail at the time of action
of the load

Analyzing the passage of the load from the embankment to the wedge
2, it is observed that the settlements start to decrease quickly,
approximately 5,1m before coming to the top of the wedge 2. This
means that the settlements only begins to occur when the load is felt at
a depth that reaches the material of the wedge 2. Thereafter, the
magnitude of the settlements hardly varies until the abutment, where
they decrease abruptly. Although no significant changes occur in
settlements value along the wedges 2 and 1; it is possible to verify that
the models 2 and 3 (where Youngs modulus ratio between wedge 1
and 2 is 2,0) present a better performance to make the transition to the
bridge.
When the passage of Thalys train was simulated, significant
differences between the settlements occurring along the embankment
and on the bridge were obtained. This means that it is really necessary
to improve existing solutions of transition zone between embankments
and bridges, which means to produce a decrease of settlements as
continuous as possible, and to avoid the bump at the end of the bridge.
REFERENCES
Seara, I. and A. Gomes Correia. 2008. Zonas de transio de
vias-frreas. A importncia de uma soluo geoestrutural.
In Proceedings of XI Congresso Nacional de Geotecnia.
Coimbra: Portugal. 245 252.
AKNOWLEDGEMENT
The authors wish to thank to Portuguese Foundation for Science and
Technology (FCT) the support given through the doctoral grant SFRH /
BD / 39055 / 2007.

Engenharia para a Qualidade de Vida: CIDADE Semana da Escola de Engenharia -11 a 16 de Outubro de 2010