COMPARISION OF DWT & WPT TO DETECT BEARING FAULTS IN 3 PHASE INDUCTION MOTOR USING CURRENT SIGANTURE ANALYSIS

Abstract

Detection of bearing faults has become a central problem in industry over the past decade. A solution to embark upon this problem is by using a sensor less monitoring. Sensor less monitoring of the induction motors can radically reduce the cost of maintenance by allowing the early detection of faults and this is more predominant than the past conventional methods used. Therefore, in this work bearing fault components are estimated by Motor Current Signature Analysis (MCSA). The convectional FFT analysis has a fervent pace in the area of analyzing the current signature, on the other hand wavelets unambiguously represents the faithful detection and analysis of bearing fault using current signature analysis. In this paper, comparison of Discrete Wavelet Transform (DWT) and Wavelet Packet Decomposition (WPT) is done to detect the nascent stage bearing faults. The experimental results have affirmed the effectiveness of the proposed technique.