Sliding Mode Periodic Adaptive Learning Operation Control Method For Medium-Speed Maglev Trains Based On Particle Swarm Optimization

With some advantages of small turning radius,strong climbing ability and low noise,the medium-speed maglev train is a new type of rail transportation and has a good application prospect.During the train operation,many disturbances such as air resistance,ramp resistance and magnetic resistance will degrade the train operation control performance.Therefore,it is of great theoretical significance and application value to design a running resistance compensation control algorithm to improve the train operation control performance by using modern control theory.Considering the periodic characteristics of running resistance when the train runs back and forth on a fixed line,combining the sliding mode control(SMC),model reference adaptive control(MRAC)and periodic adaptive learning control(PALC)algorithms,a sliding mode-model reference adaptive control-periodic adaptive learning control(SM-MRAC-PALC)compensation method is proposed in this thesis to estimate and compensate the resistance in real time and improve the train operation control performance.The main contributions are as follows:(1)The working principles of traction control and long-stator linear synchronous motor(LSM)of the medium-speed maglev train are analyzed,and the train kinematics model is established considering the traction force and various kinds of resistances during the train operation.(2)An MRAC-PALC running resistance compensation control algorithm is proposed,and the stability of train operation control system under the proposed MRACPALC controller is analyzed through the Lyapunov theory.Based on a semi-real medium-speed maglev train test line,the control performance with MRAC-PALC controller is verified.The results show that the proposed MRAC-PALC approach in this thesis can accurately estimate and effectively compensate the train running resistance.Compared with the conventional PID method,the proposed MRAC-PALC scheme has smaller position and speed tracking errors and better operation control performance.(3)To further improve the robustness of the train operation control system,combining the PD-type sliding mode control method,an SM-MRAC-PALC resistance compensation scheme is proposed,and the controller parameters are tuned offline using particle swarm optimization(PSO)algorithm.Under undisturbed and disturbed conditions,some performance comparisons of SM-MRAC-PALC and MRAC-PALC controllers are given.The results show that the proposed SM-MRAC-PALC controller has better control performance and stronger robustness.