An iterative methodology for the prediction of dynamic vehicle–track interaction and long-term periodic rail wear

Artikel i vetenskaplig tidskrift, 2018

In this study, a versatile numerical method for the prediction of long-term growth of rail roughness is presented and its
functionality is demonstrated for the development of rail corrugation on small radius curves. The procedure includes
two sub-modules: (1) a time-domain model for the simulation of dynamic vehicle–track interaction in a wide range of
frequencies by using a commercial software for multibody dynamics and (2) a post-calculation of sliding wear based
on the Archard’s model in combination with a non-Hertzian and transient wheel–rail contact model. The structural
flexibility of the wheelset is accounted for by using the finite element method. The rail wear generated by a large number
of passing trains is assessed by recurrently updating the rail surface based on the wear depth calculated in each post-
processing step. The current work sets out from a previous study in which a model for the prediction of long-term
growth of rail roughness on small radius curves was developed in a general-purpose programming language. By transfer-
ring the model into a commercial software, the aim is to develop an engineering tool that is more applicable for different
operational conditions, such as various vehicle and track designs and track alignments. The proposed method is verified
by comparing the simulation results against those obtained with the pre-existing software. Conditions similar to a 120 m
radius curve on the Stockholm metro exposed to corrugation growth on the low rail are considered. The corrugation is
found to be generated by the leading wheelsets. The prevailing wavelength-fixing mechanisms are identified and
discussed.