NI CompactRIO-based methodology for online detection of stator winding inter-turn insulation faults in 3-phase induction motors

Abstract

An NI-CompactRIO-based online stator winding fault diagnosis technique has been proposed for discrimination and severity detection of partial degradation and complete breakdown of the insulation in stator winding turns of 3-phase induction motors. Principal Component Analysis (PCA) method was employed on 3-phase current signals to extract significant features. The variances of the first two principal components (PC1 & PC2) are found to be key features for fault detection. Inter-turn complete insulation breakdown has been discriminated from inter-turn partial insulation failure by analyzing the range of variance of PC2. Additionally, severity of both the faults is detected by noting the trend in variance of PC1. Faults involving a minimum of 0.27% of total phase winding turns can be successfully detected online within a few milliseconds with acceptable prediction accuracy by employing k-Nearest Neighbors classifier.