Research On Adaptive Backstepping Control Of Primary Permanent Magnet Linear Motor

In recent years,because of its advantages of low development and repair costs and large thrust,the primary permanent magnet linear motor has become a research hotspot in the field of rail transit.This paper takes the primary permanent magnet linear motor as the research object and builds the mathematical model of the primary permanent magnet linear motor considering the side-end effect.Direct thrust control is used as the basic control strategy,in which the motor’s displacement and velocity are studied.This paper’s key thesis work covers the following three parts:(1)For primary permanent magnet linear motor mathematical model of high order nonlinear and parameter time-varying characteristics,the adaptive backstepping controller is designed,in the case of time-varying parameters of motor speed and displacement under the condition of effective control,and is verified by simulation experiments,adaptive backstepping controller can make the motor actual displacement effective tend to expect.(2)The virtual control coefficients in traditional adaptive backstepping control usually adopt trial and error method,and the selection of the coefficients will directly affect the performance of the controller.The hybrid particle swarm optimization algorithm is then implemented,and genetic particle swarm optimization is paired with self-organizing particle swarm optimization and the enhanced PSO is used to set the backstepping control virtual control coefficient to increase the controller’s control accuracy.The findings of the simulation reveal that the optimized controller’s control output is clearly higher than that of the unoptimized controller.(3)In order to prevent the differential expansion phenomenon in the backstepping control from affecting the control system,a command filter is introduced in the backstepping control,and a filter compensation signal is introduced to compensate the filter error.The neural network backstepping control is used to compensate the disturbance of unmodeled load.The adaptive RBF neural network command filter backstepping control is proposed,and the controller can approach to the desired displacement faster through simulation experiments.Moreover,it has a good inhibition effect on the unmodeled load disturbance in the system,so that the controller has better stability and robustness.