Robust stator fault detection under load variation in induction motors using AI techniques

Detection of stator faults in their early stage is of great importance since they propagate rapidly and may cause further damage to the motor. Some variations in induction motors such as torque load anomalies must be considered in order to reliably detect stator faults. This paper presents robust artificial intelligence (AI) techniques for interturn short circuit (ITSC) fault detection of stator in three phase induction motors. In this work, the focus is first on the application of artificial neural networks and then fuzzy logic systems to reduce significantly the effect of load variations on fault detection procedure. The proposed ANN methodology has the merit to detect and locate ITSC fault, while the Fuzzy approach is capable of detecting and diagnosing the severity of ITSC fault. The simulation and experimental results are also given to verify the efficiency of both approaches under ITSC fault and load change.