On-Line Neural Network Stator Resistance Estimation in Direct Torque Controlled Induction Motor Drive

This paper presents an on-line estimation for the stator resistances of the induction motor in the direct torque controlled drive, using artificial neural networks. The variation of stator resistance due to changes in temperature or frequency degrades the performance of such control strategy. In order to solve this issue, a backpropagation algorithm is used for training of the neural network. The error between the desired state variable of an induction motor and the actual state variable of a neural model is back propagated to adjust the weights of the neural model, so that the actual state variable tracks the desired value. Simulation results show the good performance of these resistance estimator and torque response of the drive.