Neural Network Sliding Mode Control For Permanent Magnet Linear Synchronous Motor

With the continuous development of automation control technology and industrial Internet of Things,permanent magnet linear synchronous motor(PMLSM)is characterized by its simple structure,high reliability and high efficiency.The simple mechanical structure of the permanent magnet synchronous linear motor also causes its performance to be easily affected.When the internal parameters change,external sudden load and motor thrust pulsation occur,the stability performance of linear motor will decline sharply.At present,the traditional control algorithm is not ideal for improving the dynamic performance of the linear motor,so that the performance and efficiency of the linear motor drive system cannot be fully utilized.Therefore,the design of a suitable controller is the key to ensure the dynamic performance of the linear motor control system.In order to improve the position tracking performance of PMLSM,this paper builds a mathematical model of a permanent magnet synchronous linear motor.Then,due to the nonlinear and strong coupling characteristics of the linear motor model,in order to facilitate analysis and simplify the calculation,the mathematical model of the motor was established in the coordinate system.Aiming at the problem that permanent magnet linear synchronous motor is vulnerable to parameter changes and external interference,the radial basis function neural network adaptive backstepping controller is designed for permanent magnet synchronous linear motor system.By observing the unknown disturbances and parameter changes in the system,the position tracking performance and anti-interference ability of the system are improved.Then,a model was established in Matlab/Simulink for simulation experiments to verify the effectiveness of the controller.In order to improve the position tracking performance and anti-interference ability of the system,a terminal sliding mode control method is introduced on the basis of neural network backstepping control.The tracking performance and anti-disturbance performance of permanent magnet synchronous linear motor control system are guaranteed by using the strong robustness of sliding mode control,high stability under external disturbance and insensitive to the changes of system parameters.At the same time,the neural network observer and sliding mode control are combined to keep the advantages of sliding mode control,and the disturbance is observed by neural network,which can suppress the short chattering phenomenon of sliding mode controller.Matlab / Simulink simulation results show that the neural network adaptive terminal sliding mode control not only has fast response speed,but also has strong antiinterference ability.Finally,in order to verify the actual application effect of the algorithm,a permanent magnet synchronous linear motor experimental platform is built based on the dSPACE simulation system,and the position tracking performance of the proposed algorithm is experimentally studied.By comparing with traditional sliding mode control and adaptive control,the feasibility and effectiveness of the algorithm in this paper are verified in the permanent magnet synchronous linear motor control system.