Title: Wavelet-based signal processing technique for multiple fault detection in induction motor: ANN approach

Authors: Anjali U. Jawadekar; G.M. Dhole; S.R. Paraskar; M.A. Beg

Addresses: Department of Electrical Engineering, S.S.G.M.C.E., Shegaon, Buldhana, Maharashtra, 444203, India. ' Department of Electrical Engineering, S.S.G.M.C.E., Shegaon, Buldhana, Maharashtra, 444203, India. ' Department of Electrical Engineering, S.S.G.M.C.E., Shegaon, Buldhana, Maharashtra, 444203, India. ' Department of Electrical Engineering, S.S.G.M.C.E., Shegaon, Buldhana, Maharashtra, 444203, India

Abstract: Faults and failures of induction motor can lead to excessive downtimes and generate large losses in terms of maintenance and revenue. This motivates motor monitoring, incipient fault detection and diagnosis. Most recurrent faults in induction motor are turn to turn short circuit, bearing deterioration, and cracked rotor bar. This paper presents a novel online methodology for multiple fault detection in induction motor. Motor line currents are recorded and analysed using modern signal processing tool and processed by applying discrete wavelet transformation. Feed forward ANN is used for fault characterisation based on DWT features extracted from motor current, which identifies bearing defects: interturn fault, broken rotor bar and voltage unbalance conditions in three phase induction motor. In order to demonstrate the efficiency of DWT detail analysis of fault classification is done based on time domain approach. Results are presented below to show the effectiveness of proposed methodology.

Keywords: signal processing; induction motors; discrete wavelet transform; DWT; multiple fault detection; ANNs; artificial neural networks; bearing defects; interturn faults; broken rotor bars; voltage unbalance; fault classification.

DOI: 10.1504/IJCSYSE.2012.050233

International Journal of Computational Systems Engineering, 2012 Vol.1 No.2, pp.100 - 107

Published online: 28 Aug 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article