Probabilistic simulation for the certification of railway vehicles

Abstract

The present dynamic certification process, built thanks to experts' experience is essentially based on experiments. The introduction of the simulation in this process would be of great interest. However an accurate simulation of complex, non-linear systems is complicated, in particular when rare events (unstable behaviour for example) are considered. After having analysed the system and the richness of the present procedure, this paper proposes a method to achieve, in some particular cases, a numerical certification. It focuses on the need for precise and representative excitations (running conditions) and on their variable nature. A probabilistic approach is therefore proposed and illustrated by an example.First the paper presents a short description of the vehicle / track system and of the experimental procedure. The proposed numerical process is then described. The necessity of analysing a set of running conditions at least as large as the one tested experimentally is moreover explained. In the third section a sensitivity analysis of the system is reported, to determine the most influential parameters. Finally the proposed method is summarized and an application is given.

Accuracy of the experimental assessment of running dynamics characteristics quantified through an uncertainty framework

“…A well-known propagation technique is the Monte-Carlo method which, along with many other methods having the same goal, is considered here as a type of a priori technique. For details on the applications of a priori methods to running dynamics problems, see Funfschilling et al [5][6][7] A priori methods were not considered sufficient for the railway-related problem addressed in DynoTRAIN. In fact, for railway vehicle assessment, an important part of representing uncertainty information is comparing the results of different assessment tools (e.g.…”

“…The above description summarises the process to obtain the EN14363 assessment quantities (refer to the standard itself, or to Funfschilling et al [5][6][7] for more details). The uncertainties associated with this process are the objects of analysis, particularly the uncertainty of the assessment quantities, but also the uncertainties associated with the variables that have an influence on the uncertainty of the assessment quantities.…”

Unified Framework and Survey for Model Verification, Validation and Uncertainty Quantification

“…Today, cars are in a transition to automated vehicles (AVs). AVs consist of multiple modules including environment (68)…”

“…They show how the variability of the input conditions such as the track geometry can be modeled and how to propagate it by the simulation model. In [68] they perform a sensitivity analysis of the inputs and in [66] they present a survey on uncertainty quantification methods in rail vehicle dynamics. Lestoille et al [114], Lestoille [113] create a stochastic model considering uncertainties and use advanced stochastic methods such as polynomial chaos expansion to calibrate it.…”

Analysis of the dynamic response in the railway vehicles to the track vertical irregularities. Part II: The numerical analysis

“…The numerical simulations are used as basic tools in research and the designing stage for estimating the dynamic behaviour of the railway vehicle and the optimization of its dynamic performance, and later on for examining the issues emerging during exploitation. In the recent years, the specifications regarding the homologation of the railway vehicles from the perspective of the dynamic behaviour in terms of safety, track fatigue and ride quality [4,5] have been added to a string of requirements so as to carry out the tests based on numerical simulations, thus regulating the so-called procedure of ‚virtual homologation' [6,7].…”