用人工神经网络实现对感应电机转速的估计

为了准确估计电机的转速，提出了基于人工神经网络的转速估算方法。从感应电机的动态方程出发，推导出了转速的表达式，然后根据转速与定子电压、定子电流、转子磁链之间的非线性关系，建立了两种人工神经网络转速估计模型。仿真结果表明，这两种基于人工神经网络的转速估计模型可以准确跟踪电机转速的变化，即使转子电阻参数发生变化，这两种转速估计模型仍具有良好的性能

Speed estimation of induction motor using artificial neural network

The speed estimation of an induction motor is the key technology for high performance control in developing speed sensorless vector controlled system. In order to estimate the speed of an induction motor accurately, the methods of speed estimation based on artificial neural networks were presented. Two expressions of the rotor speed were derived from the dynamic equations of the induction motor. According to the rotor speed as the nonlinear functions of the stator voltages, the stator currents and the rotor flux, two speed estimation models based on artificial neural networks were built. Simulation results show that the neural network speed estimators can accurately track the variation of the motor speed, and have the better dynamic response characteristics when the rotor parameters are changed.