Research on the Speed Control of Automatic Train Operation System Based on the Novel PSO-B-BP-PID Controller

Abstract

By studying the control algorithms of automatic train operation system,the PSO-B-BP-PID controller based on reparametric B-Spline Neural Networks and PSO is proposed.The controller can find more suitable factor by PSO,thereby, it can receive the best reparametric B-Spline basic functions fit for searching weights of the network,at the same time, by replacing the conventional backward-propagation(BP)algorithm with PSO in the paper,it can overcome the shortcomings of easily going into local optimum.By comparing the control performance of PSO-B-BP-PID,B-BP-PID and PSO-BP-PID in experiment,the results indicate the PSO-B-BP-PID can improves punctuality,comfort and stop precision in ATO system.