Multibody simulation benchmark for dynamic vehicle–track interaction in switches and crossings: results and method statements

Y. Bezin, B.A. Pålsson, W. Kik, P. Schreiber, J. Clarke, V. Beuter, M. Sebes, I. Persson, Hugo Pacheco Magalhaes, Ping Wang, P. Klauser

Research output: Contribution to journalArticlepeer-review

20 Citations (Scopus)

Abstract

A benchmark of railway multibody dynamics software application to switches and crossings (S&C) is presented, comparing all major commercially available software and a few independent codes. Two different representative S&C have been implemented, using the Manchester Benchmark passenger vehicle. The final results show that all software offer a reliable and efficient way to understand the kinematic and dynamics forces between the wheels and the track elements. The highest challenges are found when modelling a combination of multiple rails in simultaneous contact with a wheel (check-stock or switch-stock), large longitudinal variations in rail shape (crossings) and high lateral steering forces (diverging cases in tight radius). In those cases, the codes able to account for the exact relative motion of each wheels with respect to each rails independently are the most apt. The most significant variations between software are found in the contact prediction with an influence on the detailed contact tangential and normal forces. The user variability is found to be very small, with the most time-consuming and error prone being the task of handling the input data for the variable rails definition. All software could benefit from improvements to assist the user and ensure higher reliability and efficiency generally.
Original languageEnglish
Pages (from-to)660-697
Number of pages38
JournalVehicle System Dynamics
Volume61
Issue number3
Early online date23 Aug 2021
DOIs
Publication statusPublished - 1 Mar 2023

Fingerprint

Dive into the research topics of 'Multibody simulation benchmark for dynamic vehicle–track interaction in switches and crossings: results and method statements'. Together they form a unique fingerprint.

Cite this