基于模糊神经滑模观测器的永磁同步电机无传感器控制

为了解决传统滑模观测器的抖振问题,提出了一种用于永磁同步电机的模糊神经滑模观测器。分析了滑模增益对抖振的影响,并采用模糊神经网络动态调整滑模增益以改善抖振,利用李雅普诺夫函数证明了模糊神经网络观测器的稳定性。利用锁相环方法提取转子位置与速度信息,减小由高频噪声引起的误差。仿真结果表明:改进后的滑模观测器能够对永磁同步电机转子位置进行精确辨识,有效地抑制了抖振,实现了高性能的永磁同步电机无传感器控制。

SENSORLESS CONTROL OF PMSM BASED ON FUZZY-NEURAL NETWORK SLIDING MODE OBSERVER

In order to solve the chattering problem of conventional sliding mode observer for a permanent magnet synchronous motor (PMSM), a fuzzy-neural network sliding mode observer was proposed. The effect of sliding mode gain to chattering was analyzed. The fuzzy-neural network dynamically adjusts sliding mode gain to reduce the chattering. The stability of the fuzzy-neural network sliding mode observer was proved through Lyapunov function. The method that the phase-locked loop extracts the rotor position and speed information was utilized in order to reduce the error caused by the high-frequency noise. The validity of the proposed algorithm has been demonstrated with experiments.