A Novel Park’s Vector Approach for Investigation of Incipient Stator Fault Using MCSA in Three-Phase Induction Motors

Abstract

Fault detection of induction motors at an early stage is a critical issue related to industrial maintenance. It is of utmost significance to a situation where in a sudden failure of induction motor may lead to shutdown of production practices and might result in the disruption of essential services such as medical, transportation or military. This further leads to huge economic losses and wastage of raw material. Several methods of fault detection have been proposed in literature, but the Motor Current Signature Analysis (MCSA) is the most reliable and widely used technique having advantage of noninvasiveness. The stator fault can be diagnosed at incipient stage during the time of Condition Monitoring (CM). This article presents a novel fault detection method using Park’s Vector Approach (PVA) for Inter-Turn Short (ITS) fault in a squirrel cage induction motor by monitoring three phase stator current. The method transforms the three phase stator current into two orthogonal phases and then finds the fault based on the spatial variations of Parks current vector locus. Experimental results obtained from a healthy motor (without any turn fault), motor with 2.2%, 4% and 5.1% shorted turns are presented to validate the technique. The proposed technique not only detects the fault existence but also estimates the level of fault severity.