关键词: 永磁同步电机, RBF神经网络, PID控制器, 梯度下降法, 变速积分

Abstract:

The RBF neural network was applied to the parameter tuning of the PID for permanent magnet synchronous motor. The parameters of RBF neural network were modified by the gradient descent algorithm. The variable integral was adopted to improve the control precision for the integral part. The simulation results show that the controller shortens the overshoot of the speed regulation and fastens the system responses. And it has better adaptation, robustness and preventing disturbance ability.

Key words: permanent magnet synchronous motor (PMSM), RBF neural network, PID controller, gradient descent algorithm, variable integral