Online stator resistance estimation using artificial neural network for direct torque controlled induction motor drive

For the stable and effective operation of the induction motor proper estimation of the stator resistance is very essential.This is because stator resistance keeps on increasing with the temperature when the motor is in operation which results in high torque and flux ripple.A method based on artificial neural network to estimate the stator resistance of induction motor for direct torque control drive is proposed in this paper. For the training purpose of neural network a back propagation algorithm has been used. The adjustments of the weights of the neural network has been done by back propagating the error signal between measured and estimated current of stator. From simulation it has been proved that the estimator can track stator resistance value within around 40ms when a step change of stator resistance has been applied. Effectiveness of the estimator is investigated in simulation by varying the stator resistance from the nominal value which has been done in MATLAB SIMULINK.