Research on online identification of the stator resistance using wavelet neural network

The change of the stator resistance in induction motor greatly affects the performances of direct torque control (DTC) system run at low speeds. It is hard to form an accurate math model, for the change of the resistance value is nonlinear and time varying. According to the terminal temperature of winding and the temperature variation, this paper presents a wavelet neural network used as resistance on-line identification. After trained with recursion arithmetic, the network was used to measure the resistance. The results show that this identifier can precisely measure the value of resistance and efficiently improve the low-speed performances of DTC system.