永磁同步电机位置伺服系统的RBF神经网络滑模控制策略

RBF neural network sliding mode control strategy for permanent magnet synchronous motor position servo system

针对永磁同步电机(PMSM)位置伺服控制系统的负载扰动、外部不确定性干扰和模型参数摄动的特点,常规PID控制策略难以达到满意的控制效果.利用滑模控制对系统扰动和参数变化不灵敏的优点,提出一种基于神经网络和滑模控制相结合的位置伺服优化控制策略.在常规滑模控制器设计的基础上,引入RBF神经网络调节滑模控制器的切换增益,削弱系统的抖振,并通过在系统中设计干扰观测器实现对扰动的补偿.仿真结果表明:与常规滑模控制和常规PID控制相比,不同参数下本文所提出的优化控制策略超调量最多降低22%,调节时间最多减少9.2 s,有效提高位置伺服系统的鲁棒性、抗干扰能力和跟踪精度,且系统抖振得到有效遏制.

In view of the characteristics of load disturbance,external uncertainty interference and model parameter perturbation of the permanent magnet synchronous motor(PMSM)position servo control system,conventional PID control strategies are difficult to achieve satisfactory control results.Taking advantage of sliding mode control vhich is not sensitive to system disturbances and parameter changes,a position servo optimization control strategy based on the combination of neural network and sliding mode control is proposed.On the basis of the conventional sliding mode controller design,the RBF neural network is introduced to adjust the switching gain of the sliding mode controller,weaken the chattering of the system,and realize the compensation of disturbance by designing a disturbance observer in the system.The simulation results show that compared with conventional sliding mode control and conventional PID control,the optimized control strategy proposed in this paper reduces the overshoot by up to 22%and the adjustment time by up to 9.2 s under different parameters,which effectively improves the robustness of the position servo system.Performance,anti-interference ability and tracking accuracy,and the system chattering is effectively curbed.