Keywords: high speed trains, braking performance, deceleration variability, degraded conditions.

The operation of traffic on a railway network requires effective control of the stopping behaviour of the trains using a signaling system. For example the European Signaling System (ETCS or ERTMS transmits to the train driver or to the on-board automatic train control system the information that permits the control of train speed with brake applications. The need to optimize the traffic density and the new possibilities of technology have led to the design of on-board automatic train control systems that calculate, with data transmitted from the ground, the exact distance to prevent the passing of a danger point.

Train speed supervision systems, namely automatic train protection (ATP) system principles use a comparison between actual train speed and reference values evaluated as a function of train position and are usually referred to as braking curves. Their definition requires a preliminary knowledge of braking models that allows the conversion of the general parameters affecting the braking performance of a train into a deceleration profile as a function of time and speed.

The guaranteed deceleration values used to calculate braking curves are obtained by applying a proper safety factor to the nominal deceleration depending on train characteristics. Train actual deceleration may differ from the nominal one arising from the variability of several parameters: the safety factor has to take into account such variability and to guarantee that the train can comply with the imposed speed constraints with an appropriate safety margin. This study enables the evaluation of the probability that the real deceleration is smaller than the one used in the basic braking model and consequently that the real stopping distance is longer than the one calculated using the braking model. This model can then be used to tune the value of the safety margin in order to obtain a certain reliability, defined in probabilistic terms.

The numerical procedure used to simulate the braking performance is based on the Monte Carlo method that is a method for iteratively evaluating a deterministic model using sets of random numbers as inputs, often used when the model is complex, nonlinear, or involves several parameters. This paper starts from previous studies developed for some types of passenger and freight trains and extends the proposed model to high-speed trains. The paper presents the main features and the implementation of a deterministic model used to describe high-speed train deceleration, then the probability distributions of input parameters affecting the braking distance are presented and discussed. The results of this model are then compared with some experimental data.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £110 +P&P)