Fault diagnosis in induction motor using soft computing techniques

Induction motor is the most commonly used electrical machine in industry. Condition monitoring is necessary in order to avoid motor failure. Different fault monitoring techniques for induction motor can be broadly categorized as model based, signal processing based and soft computing techniques. Soft computing techniques give good analysis of a faulty system even if accurate models are unavailable. They are easy to extend and modify and also give improved performance. Here the different soft computing techniques for fault diagnosis are discussed. A methodology based on park's vector approach employing Fuzzy logic and Artificial Neural Network (ANN) is used to diagnose electrical faults. Simulation study was carried out corresponding to faults like stator voltage unbalance; stator open phase and stator short-circuit. An experimental setup to model stator fault is developed and the results obtained are utilized to create a fuzzy logic based fault diagnosis scheme to detect the fault and to quantify its severity. In addition, a hardware setup was developed for on-line monitoring of induction motor conditions and the proposed fault diagnosis scheme employing fuzzy logic was successfully implemented.