Artificial Neural Networks Technique to Detect and Locate an Interturn Short-Circuit Fault in Induction Motor

Induction machines (IMs) faults diagnosis is of significance to enhance the reliability and security of the industrial productions. In recent years, the faults detection and diagnosis of IMs have moved from traditional techniques that using the signals analysis to artificial intelligence (AIs) techniques such as the neural networks (NNs). Unfortunately, stator faults related failures account for a large percentage of faults in IMs. In this context, this paper proposes a diagnostic technique based on NNs for detecting and locating the interturns short-circuit in one of three stator winding phases of IM. Often, the most important step in which NNs is the process of selecting the best inputs that enables higher performance and better diagnosis, in this study, the three-phase shift between the stator voltages and its currents are considered as inputs of the NNs in order to develop an automatic fault detection and classification system. The simulation results in MATLAB environment prove the efficiency of the suggested NN technique to detect and locate short-circuit between turns of the same coil when the SCIM is operating under various load levels and various fault severities.