Multi-Stage Linear Prediction Model for Railway Track Irregularity

Abstract

Railway track irregularity has significant impact on traffic safety. It can ensure high smooth of the track from predicting the deterioration of track irregularities accurately, and then fundamentally eliminate hidden dangers. In this paper, a multi-stage linear prediction method for track local irregularities based on time series is proposed. On the basis of the track irregularity management standard in our country, this paper has researched the changing regularity of irregularities in unit section and built the local irregularities deterioration model for each unit section to forecast the development of irregularities during a maintenance period. After comparing predicting data with actual data and analyzing the error, the accuracy of the model was verified.