Manchester Metropolitan University's Research Repository

Assessing railway vehicle derailment potential using neural networks

Iwnicki, Simon D., Parkinson, Howard and Stow, Julian M. (1999) Assessing railway vehicle derailment potential using neural networks. Professional Engineering Publications Ltd..


Download (479kB) | Preview


Current methods for ensuring the safe running of railway vehicles assess the track and vehicle condition against fixed limits. Any exceedence of these limits requires remedial action to be taken. The setting of these limits is based on past experience or on computer modelling of vehicle track interaction. This paper describes the initial results of a novel method aimed at predicting vehicle behaviour from track measurements using an artificial neural network. The speed of the neural network method would allow quick analysis of all the data measured by the track recording coach and would also allow maintenance decisions to be based on the effect of track condition on the vehicle behaviour rather than on simple limits.

Impact and Reach


Activity Overview
6 month trend
6 month trend

Additional statistics for this dataset are available via IRStats2.

Actions (login required)

View Item View Item