基于小波神经网络的永磁同步电机无速度传感器滑模控制

Speed Sensorless Control of Permanent Magnet Synchronous Motor Based on Wavelet Neural Networks

永磁同步电机(PMSM)是一个多变量、非线性、强耦合的复杂系统,对外界扰动及内部摄动极为敏感。为提高系统鲁棒性,基于小波神经网络的时频局部特性、变焦特性、自学习、自适应、鲁棒性及很强的非线性逼近能力等特性,建立非线性映射,估计转子位置并计算转子的输出速度,引入滑模变结构控制策略对PMSM进行控制,并在同等条件下对PI控制进行了仿真研究,结果表明所提出的控制策略能有效提高系统的调速性能。

Permanent-magnet synchronous-motor is a muti-variable, non-linear, strong coupling system that is highly sensitive to the outer interfere and inter perturbation. To improve the system robustness, based on wavelet neural network time-frequency localization properties, zoom features, self-learning, adaptive, robust, and strong nonlinear approximation ability of the character and to establish non-linear mapping, it is estimated rotor position and calculate the output of the rotor speed, the introduction of sliding mode variable structure control strategy for PMSM control, and under the same conditions also PI control of a simulation study results showed that the proposed control strategy can effectively improve the system spped performance.