基于多层前传网络的异步电机转子磁链和转速辨识

Identification of Rotor Flux and Speed of Induction Motors by Multilayer Forward Network

文章利用多层前传网络对异步电机转子磁链和转速进行直接辨识,多层前传网络具有良好的函数逼近能力,利用误差反传算法对多层前传网络进行训练,使神经网络准确地反映电机转子磁链和转速,在此基础上建立了一个异步电机矢量控制系统.系统仿真结果表明,这种方法能较好地辨识电机转子磁链信号和转速,系统具有良好的动态性能.

This paper proposes a method for identifying directly the rotor flux and speed of induction motors using the multiplayer forward networks. The multilayer forward network has the capacity of good function approximations. We use the error backward transmission algorithm to train the multilayer forward network, so that the neural network can identify accurately the rotor flux and speed of induction motor. Based on this fact, we set up an induction motor vector control system and the results of the system simulation shows that this technique can effectively identify the rotor flux signal and speed of the motor, and that the system has good dynamical performance.