Research On Speed Control Algorithm Of Freight Train Based On Parameter Identification

Enhancing the intelligent level of freight railways can effectively improve quality of services,which is the trend of freight railways development in the future,and the new technologies represented by autonomous driving technology are leading a new round of technological innovation.Speed tracking control for freight train is one of the core technologies of freight train automatic driving,which aims to replace manual train speed regulation.However,the long-distance transportation of freight trains and the changeable environment cause the basic resistance of the train to appear nonlinear and time-varying,which greatly affects the speed tracking control accuracy of freight trains.In addition,the discrete level control of freight trains needs to meet the level switching rules in the speed tracking control,which puts higher requirements on the speed tracking controller.Therefore,this paper focuses on freight trains and research on the speed tracking control algorithm of freight trains based on parameter identification to realize the safe and stable operation of trains.Firstly,for the nonlinear and time-varying characteristics of the basic running resistance coefficients of freight trains,an online parameter estimation algorithm based on genetic algorithm is designed.By sampling the train operation data and aiming at speed fitting,the genetic algorithm updates the search range of the basic resistance coefficients after the train runs for a long distance,and the precise value of the basic running resistance coefficients is calculated within this range after the single-step operation of the train is controlled.The simulation shows that the method can accurately estimate the basic running resistance online,and overcome the data overfitting phenomenon caused by the least square method.Secondly,the fuzzy controller of freight train with double-layer structure is designed by using the interval type-2 fuzzy logic control principle and the type-1 fuzzy logic control principle.The interval type-2 fuzzy logic control principle takes into account the train speed tracking,acceleration change and handle level switching,which is used to calculate the amount of handle level change.The type-1 fuzzy logic control principle converts the train working condition rules into logical values,and uses matrix operation to judge whether the handle level and working conditions calculated by the type-2 fuzzy logic control principle meet the actual requirements.The simulation shows that the control effect of the fuzzy controller is consistent with the actual situation,and the interval type-2 fuzzy logic control principle can effectively reduce the switching frequency of the handle level.Finally,the model predictive control principle is introduced to optimize the membership function parameters of the interval type-2 fuzzy controller.The online feedback correction of the train prediction model is performed using the online estimation algorithm of the train resistance coefficients based on genetic algorithm.The pattern search algorithm is used to optimize the scaling factor of the membership function,so that the fuzzy controller for freight trains in the next control domain has a better control effect and achieves accurate tracking of the target speed while minimizing the frequent switching of the handle level.Simulation shows that the fuzzy predictive controller can effectively improve the accuracy of train speed tracking,and has strong anti-interference ability under the time-varying basic running resistance.