Feature Extraction using Wavelet Transform for Multi-class Fault Detection of Induction Motor

In this paper the theoretical aspects and feature extraction capabilities of continuous wavelet transform (CWT) and discrete wavelet transform (DWT) are experimentally verified from the point of view of fault diagnosis of induction motors. Vertical frame vibration signal is analyzed to develop a wavelet based multi-class fault detection scheme. The redundant and high dimensionality information of CWT makes it computationally in-efficient. Using greedy-search feature selection technique (Greedy-CWT) the redundancy is eliminated to a great extent and found much superior to the widely used DWT technique, even in presence of high level of noise. The results are verified using MLP, SVM, RBF classifiers. The feature selection technique has enabled determination of the most relevant CWT scales and corresponding coefficients. Thus, the inherent limitations of CWT like proper selection of scales and redundant information are eliminated. In the present investigation ‘db8’ is found as the best mother wavelet, due to its long period and higher number of vanishing moments, for detection of motor faults.

[1]  A. Prochazka,et al.  Signal Analysis and Prediction , 1998 .

[2]  Dae-Seong Kang,et al.  A study on fault diagnosis of induction motor using neural-wavelet , 2008 .

[3]  Peter Tavner,et al.  Condition monitoring of electrical machines , 1987 .

[4]  Wang Bin,et al.  Motor Fault Diagnosis Based on the Vibration Signal Testing and Analysis , 2009, 2009 Third International Symposium on Intelligent Information Technology Application.

[5]  Gaurav Pandey,et al.  SVM BASED FAULT DIAGNOSIS OF MONOBLOCK CENTRIFUGAL PUMP USING STATIONARY WAVELET FEATURES , 2011 .

[6]  Atif Iqbal,et al.  Comparative performance analysis of Thyristor and IGBT based induction motor soft starters , 2010 .

[7]  Rich Caruana,et al.  Greedy Attribute Selection , 1994, ICML.

[8]  M. El Mansori,et al.  On the optimal choice of wavelet function for multiscale honed surface characterization , 2011 .

[9]  Satish C. Sharma,et al.  Fault diagnosis of ball bearings using continuous wavelet transform , 2011, Appl. Soft Comput..

[10]  B. O. Al-Bedoor,et al.  WAVELET ANALYSIS OF VIBRATION SIGNALS OF AN OVERHANG ROTOR WITH A PROPAGATING TRANSVERSE CRACK , 2001 .

[11]  Xin-Ping Xie,et al.  Gene expression pattern extraction based on wavelet analysis , 2009, 2009 International Conference on Information and Automation.

[12]  J. Rafiee,et al.  A novel technique for selecting mother wavelet function using an intelligent fault diagnosis system , 2009, Expert Syst. Appl..

[13]  Iqbal Gondal,et al.  Envelope-Wavelet Packet Transform for Machine Condition Monitoring , 2011 .

[14]  Ratna Dahiya,et al.  Rotor Faults Detection in Induction Motor by Wavelet Analysis , 2009 .

[15]  P. Konar,et al.  Bearing fault detection of induction motor using wavelet and Support Vector Machines (SVMs) , 2011, Appl. Soft Comput..

[16]  R. Merry.  Wavelet theory and applications : a literature study , 2005 .

[17]  K. S. Verma,et al.  Wavelet transform based power quality events classification using artificial neural network and SVM , 2012 .

[18]  W. R. Finley,et al.  An analytical approach to solving motor vibration problems , 1999, Industry Applications Society 46th Annual Petroleum and Chemical Technical Conference (Cat.No. 99CH37000).

[19]  Bo-Suk Yang,et al.  Application of nonlinear feature extraction and support vector machines for fault diagnosis of induction motors , 2007, Expert Syst. Appl..

[20]  Satish C. Sharma,et al.  Fault diagnosis of ball bearings using machine learning methods , 2011, Expert Syst. Appl..

[21]  Subrata Karmakar,et al.  Novel diagnosis technique of mass unbalance in rotor of induction motor by the analysis of motor starting current at no load through wavelet transform , 2010, International Conference on Electrical & Computer Engineering (ICECE 2010).

[22]  S. Seker,et al.  Transfer Function approach based upon wavelet transform for bearing damage detection in electric motors , 2008, 2008 IEEE International Symposium on Industrial Electronics.

[23]  Bo-Suk Yang,et al.  Fault Diagnosis System of Induction Motors Using Feature Extraction, Feature Selection and Classification Algorithm , 2006 .

[24]  Neil Kelley,et al.  USING TIME-FREQUENCY AND WAVELET ANALYSIS*' TO ASSESS TURBULENCE/ROTOR INTERACTIONS , 2000 .

[25]  Mostefa Mesbah,et al.  Wavelet Decomposition for the Detection and Diagnosis of Faults in Rolling Element Bearings , 2009 .

[26]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[27]  Gang Yu,et al.  Machine fault diagnosis using a cluster-based wavelet feature extraction and probabilistic neural networks , 2009 .

[28]  Mo-Yuen Chow,et al.  Neural-network-based motor rolling bearing fault diagnosis , 2000, IEEE Trans. Ind. Electron..

[29]  Qiang Shen,et al.  Fuzzy-Rough Sets Assisted Attribute Selection , 2007, IEEE Transactions on Fuzzy Systems.

[30]  A. Ray,et al.  Broken Rotor Bar detection using Symbolic Wavelet Analysis , 2006, Proceedings of the 45th IEEE Conference on Decision and Control.

[31]  Pedro Vicente,et al.  Current-, force-, and vibration-based techniques for induction motor condition monitoring , 2007 .

[32]  Bo-Suk Yang,et al.  Combination of independent component analysis and support vector machines for intelligent faults diagnosis of induction motors , 2007, Expert Syst. Appl..

[33]  Li Xiaolei,et al.  The application of wavelet transform and artificial neural networks in machinery fault diagnosis , 1996, Proceedings of Third International Conference on Signal Processing (ICSP'96).