Neural network approach for inter-turn short-circuit detection in induction motor stator winding

Abstract

This work deals with neural network approach for automatic detection of stator winding fault of the induction motor. The problem is faced through modeling of induction motor with short circuit in stator winding. Instantaneous phase voltages, peak values of phase currents, and parameters derived from these data are used to train artificial neural network. The output of the neural network classifies the condition of the stator winding. The proposed architecture performs with the selection of a significant feature set, and accurate fault detection results have been obtained.