Abstract:

An improved sliding mode model reference adaptive system (MRAS) was designed to improve the control performance of traditional permanent magnet synchronous motor (PMSM) sliding mode MRAS. Based on the integral sliding mode surface of traditional sliding mode MRAS, the non-singular fast terminal sliding mode surface was introduced, which constitutes an integral non-singular fast terminal sliding mode surface to improve the convergence performance. A nonlinear exponential function was used to replace the switching function in the traditional sliding mode MRAS, and an improved gray wolf optimization (GWO) algorithm was used to optimize the sliding mode surface parameters globally to improve the control precision. The simulation results of MATLAB/Simulink software show that compare with the traditional sliding mode MRAS, the improved sliding mode MRAS can reduce the torque ripple amplitude by 75%, reduce the actual speed jitter amplitude by 80%, control the actual speed overshot within 2.3%, control the adjustment time within 10 ms, and reduce the observation error of rotor position by 60%.

Key words: permanent magnet synchronous motor (PMSM), model reference adaptive system (MRAS), sliding model MRAS, grey wolf optimization (GWO), non-singular fast terminal sliding surface, nonlinear exponential function