Time domain 3D finite element modelling of train-induced vibration at high speed

A. El Kacimi, Peter Keith Woodward, O. Laghrouche, G. Medero

    Research output: Contribution to journalArticlepeer-review

    166 Citations (Scopus)

    Abstract

    The purpose of this paper is to investigate a 3D finite element (FE) coupled train-track model for the numerical modelling of the ground induced vibration due to the passage of a single high speed train locomotive. The track components such as the sleepers, the ballast and the subgrade are represented by 20 noded brick elements. The rail is modelled by using 3D beam-column elements. A quarter train model is coupled to the 3D railway track model, through the interaction points between the wheels and the rail, based on the nonlinear Hertzian contact theory. A damping model, based on Rayleigh damping approach, is used. The 3D FE model is capable of simulating multi-layered ground and radiation damping by using viscous boundary conditions. Material nonlinearity, especially of the ballast and subgrade layers can be taken into account as appropriate. Numerical experiments are carried out, using the proposed 3D FE coupled train-track model, to study the train fundamental passing frequency effect on the dynamic railway track response for train speeds belonging to the subcritical, critical and supercritical ranges. The influence of the soil material damping is also investigated. The results clearly show an increase in track deflection with train speed. The material damping model allows a realistic prediction of the track vibration and train body dynamics at high speeds.

    Original languageEnglish
    Pages (from-to)66 - 73
    Number of pages8
    JournalComputers and Structures
    Volume118
    Issue numberMarch 2013
    Early online date19 Aug 2012
    DOIs
    Publication statusPublished - 19 Aug 2012

    Fingerprint

    Dive into the research topics of 'Time domain 3D finite element modelling of train-induced vibration at high speed'. Together they form a unique fingerprint.

    Cite this