摘  要：提出基于小波神经网络的永磁同步电机无速度传感器控制方法。依据直接检测得到的电流、磁链等参数，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的特性，利用小波神经网络建立非线性映射，估计转子位置并计算转子的输出速度，实现电机无速度传感器控制。在Matlab/Simunlink仿真环境下进行仿真研究，仿真结果表明，具有较好的动态响应和鲁棒性。
关键词：永磁同步电机;小波神经网络;无速度传感器;非线性映射

Abstract: The method of speed sensorless control of permanent magnet synchronous motor based on wavelet neural networks was put forward.As the wavelet neural network has the properties of local time-frequency characteristics，zooming index，adaptive study，strong robustness and non-linear approaching,according to the parameters of current and flux directly obtained,the non-linear reflection was built to estimate the output speed of rotor and realize the speed sensorless control.The sensorless control was studied in the simulation environment Matlab/Simulink and the result shows the control strategy has good dynamic response and robustness.
Key Words: Permanent magnet synchronous motor;Wavelet neural networks;Speed sensorless;Non-linear reflection