Research On Predictive Control Method Of High Speed Train Based On Improved GA-Elman Neural Network

Due to its advantages of safety,speediness and comfort,high-speed Railway has developed vigorously in China in recent years.At present,most emus running on high-speed railways in our country adopt manual driving mode.The operation effect of emus is too dependent on the operator’s Operation Experience and technology,the improper operation or fault of the driver may cause the train delay or even threaten the safety of the train.The research of the high-speed Train Operation Control method is the key to the realization of the high-speed train automatic driving.Because of the complex running environment of high-speed train,the running process of high-speed train is a complex one with strong nonlinearity,strong interference and multi-constraints,etc..It is of great significance to design a control strategy with strong adaptive ability and good performance of Shandong rod for high-speed train speed tracking control with high precision.Therefore,this paper presents a high-speed train predictive control based on improved Ga-Elman neural network.The specific research is as follows:1.The advantages and disadvantages of Elman neural network are analyzed,and the genetic algorithm is proposed to optimize the neural network for its slow training speed,the initial weights and thresholds have great influence on the results.The performance of GA is analyzed,and it is found that GA is easy to fall into local extremum and premature convergence.Therefore,the selection operator of GA is improved.2.A detailed and comprehensive analysis is made of the forces acting on the train in the course of its operation,and a dynamic model of the train is established by theoretical analysis,Elman neural network optimized by improved genetic algorithm is used to identify the running process of high-speed train.3.Taking CRH380 AL high-speed train as the research object,the improved Ga-Elman neural network combined with generalized prediction is used to track the train speed.The control effect is tested by simulation experiment,which provides a new strategy for high-speed train speed tracking with high precision and high-speed train running safety.