| In recent years,the research of PMSM has attracted great attention from scholars and engineers all over the world.To achieve fast response,robustness and high-precision control for permanent magnet synchronous motors in complex situation,this dissertation focuses on the performance of sliding mode control algorithms on the permanent magnet synchronous motor platform,and analyzes the shortcomings of the sliding mode control algorithm,meanwhile,the extreme learning machine(ELM)is introduced to improve robustness.The modelling of PMSM is a prerequisite for the design of control algorithm.The modelling of PMSM,discussion of disturbance is under d-q rotate coordinate system,which is convenient for designing sliding mode controller.In this dissertation,the traditional sliding mode control algorithm and the non-singular terminal sliding mode control algorithm are designed first,and then the stability analysis of the system is completed via rigorous mathematical proof,and the expressions of each controller are derived.After analyzing the shortcomings of the sliding mode control algorithm,a sliding mode control algorithm based on the ELM is proposed,which uses the powerful fitting ability of the ELM to estimate the disturbance of the PMSM system,and then compensates the whole system.MATLAB/SIMULINK 2018 was selected as the simulation environment to complete the verification of each controller,and DSP was used as the microcontroller to carry out experiments,compare the performance of each controller,and verify the superiority of the proposed sliding mode control algorithm based on the ELM.The experiment results show that the sliding mode control algorithm need to be improved,and the proposed control is significantly superior than the traditional sliding mode control algorithm,the response speed relative to the nonsingular terminal sliding mode algorithm and the traditional sliding mode algorithm is increased by 18% and 33% respectively,and its anti-disturbance performance is superior to the traditional sliding mode control algorithm. |