PMSM伺服系统的RBF网络自适应滑模控制

针对位置伺服系统的非线性、外干扰及参数摄动的不确定性,通过引入RBF神经网络对未知干扰和不确定性进行自学习,提高伺服系统的鲁棒性.为了分析伺服系统的不确定性,将系统模型划分为名义模型和不确定模型两部分.其中,名义模型采用状态反馈方法进行控制,不确定部分采用RBF网络作为滑模动态补偿器进行控制,解决了伺服系统不确定性上界难以测量的问题,降低了抖振.仿真结果验证了该设计方案的有效性.

RBF Neural Network Adaptive Sliding Mode Control for PMSM Servo System

In order to improve the robustness of the servo system,the RBF neural network is used to learn the unknown disturbances and uncertainties for the nonlinear, external disturbances and parameter perturbations of the position servo system. The system model is divided into two parts, the nominal model and the uncertain model,which are used to control the uncertainty of the servo system. Among them,the nominal model is controlled by the state feedback method, and the RBF network is used as the sliding mode dynamic compensator in the uncertain part. The scheme solves the problem that the upper bound of the uncertainty of the servo system is difficult to measure and the chattering is reduced. The simulation results verify the effectiveness of the design scheme.