The Neural Network Adaptive Control for the Nonlinear Load of the Permanant Magnet Synchronous Motor

To solve the electromagnetic torque ripple caused by uncertain nonlinear factor of the Permanent Magnet Synchronous Motor (PMSM) and improve the quick response and the smooth trajectory tracking of the servo system, a robust smooth trajectory tracking method based on Neural Network compensation is designed to the servo control system in this paper. Based on the mathematical model of the PMSM and it's nonlinear load, a Neural Network backstepping control method and two-order nonlinear smooth trajectory filter is presented in this paper. Finally, the validity and effectiveness of this control method are verified through the practical DSP experiments applied into AC servo control systems.