基于RBF神经网络的永磁同步电机速度PI-IP控制

摘要

Aiming at the problem that the control method of speed loop PI and IP of PMSM servo system is not good enough, this paper presents a speed PI-IP control parameters autotuning algorithm of servo system based on RBF neural network. A speed PI-IP controller based on RBF neural network was designed through combining the advantages of RBF neural network and PI-IP composite control. The control effect of tradi-tional PI and IP controller was improved by setting speed loop control parameters. It is verified through simulation finally. The simulation results show that, compared with traditional PI and IP control algorithm, the proposed algorithm can effectively reduce the speed overshoot, control interference. The servo system designed by this algorithm has good tracking performance and strong anti interference capability of load torque.%针对PMSM伺服系统速度环PI和IP两种控制方式控制效果不佳的问题,提出一种基于RBF神经网络的伺服系统速度PI-IP控制参数自整定方法.该方法结合RBF神经网和PI-IP复合控制二者的优点,构造了基于RBF神经网络的速度PI-IP控制器,对速度环控制参数进行整定,改善常规PI和IP两种控制器的控制效果,最后通过仿真实验进行了验证.仿真结果表明:与常规PI和IP两种控制方法相比,该方法能有效减小速度超调,抑制扰动,用该方法设计的伺服系统具有良好的跟踪给定性能和较强的抗负载转矩扰动能力.