永磁同步电机的神经网络滑模与预测控制

Neural Network Sliding Mode and Predictive Control of Permanent Magnet Synchronous Motor

针对永磁同步电机(permanent magnet synchronous motor,PMSM)伺服系统转子位置跟踪问题,提出了一种基于Elman神经网络的互补滑模控制方法。首先,考虑系统的不确定性建立PMSM数学模型,利用Elman神经网络来估计系统的不确定性,以实现对PMSM高度的位置跟踪。然后,提出了一种PMSM无差拍预测电流控制方法,该方法相当于高增益的比例控制方法,从原理上增加了电流环的带宽,从而改善电流环的性能。计算机仿真和半实物仿真的结果表明,与传统的矢量控制相比,无差拍预测电流控制有效提高了电流环的动态性能和稳态精度,基于Elman神经网络的互补滑模控制有更好的跟踪性能。

For the problem of rotor position tracking of permanent magnet synchronous motor(PMSM)servo system,a complementary sliding mode control method based on Elman neural network is proposed.Firstly,the PMSM mathematical model is established considering the uncertainty of the system,and Elman neural network is used to estimate the uncertainty of the system,so as to realize the position tracking of PMSM height.Then,a deadbeat predictive current control method for PMSM is proposed,which is equivalent to high gain proportional control method.This method increases the bandwidth of the current loop in principle,so as to improve the performance of the current loop.The results of computer simulation and semi-physical simulation show that compared with the traditional vector control,the deadbeat predictive current control effectively improves the dynamic performance and steady-state accuracy of the current loop,and the complementary sliding mode control based on Elman neural network has better tracking performance.