Neural Network Based Temperature Field Mapping Model for CRTS II Type Ballastless Track

In order to investigate the mapping relationship between the meteorological parameters and the internal temperature field of the track plate, this paper takes the CRTS II type ballastless track of a passenger dedicated line in East China as the research object, and builds an on-line monitoring system for the temperature field of the intelligent track plate, and carries out real-time data on the ambient temperature, solar radiation, wind speed and the internal temperature of the track plate. Based on the BP neural network method, the variation rule of the temperature field inside the track plate is studied, and the mapping relationship between the environmental meteorological parameters and the temperature of the track plate is established. The prediction results show that the prediction accuracy of the nonlinear mapping model is 92.4% and the prediction accuracy of the temperature gradient is 81.5%, and the prediction accuracy is 96% and 93% respectively for the 13:00-16:00 time section of the overall temperature and the maximum positive temperature gradient of the track plate.