Detection of Stator, Bearing and Rotor Faults in Induction Motors

Abstract

In this paper, faults in induction motors were diagnosed using the Common Vector Approach (CVA). In order to verify the performance of CVA, a database including stator current signals for normal and faulty cases with 1.5A loading condition was used. The current signals belong to six identical induction motors one of which is normal and remaining ones are faulty motors. The 2-step One-Dimensional Discrete Wavelet Transform (1D-DWT) is applied on the current signals in order to construct feature vectors of each class in the database. While performing CVA, the leave-30-out strategy was followed to test all feature vectors in the database. Substantially satisfactory recognition results were obtained for wavelet energy component-based features.