Induction motor fault diagnosis using multiple class feature selection

This paper presents an effective and practical multiple class feature selection (MCFS) approach for induction motor fault diagnosis. Wavelet transform is applied to extracting energy features at some specific frequency components from both stator current signals and vibration signals. These energy features are collected to form a high-dimensional feature vector. MCFS algorithm is then introduced to select representative ones from the feature vector and used as input to a random forest classifier for induction motor fault pattern recognition. Experimental study performed on a machine fault simulator indicates that the MCFS can be used as an effective algorithm for feature dimension reduction in the field of induction motor fault diagnosis.

[1]  Xiaofei He,et al.  Locality Preserving Projections , 2003, NIPS.

[2]  Andrew R. Jamieson,et al.  Exploring nonlinear feature space dimension reduction and data representation in breast Cadx with Laplacian eigenmaps and t-SNE. , 2009, Medical physics.

[3]  F. Musharavati.  RECONFIGURABLE MANUFACTURING SYSTEMS , 2010 .

[4]  Gang Niu,et al.  Multi-agent decision fusion for motor fault diagnosis , 2007 .

[5]  Xiaoming Zhang,et al.  Feature Fusion Using Locally Linear Embedding for Classification , 2010, IEEE Transactions on Neural Networks.

[6]  Hyunjin Park,et al.  ISOMAP induced manifold embedding and its application to Alzheimer's disease and mild cognitive impairment , 2012, Neuroscience Letters.

[7]  Andrew Y. C. Nee,et al.  Integrated Condition Monitoring and Fault Diagnosis for Modern Manufacturing Systems , 2000 .

[8]  Jon Atli Benediktsson,et al.  A Novel Technique for Optimal Feature Selection in Attribute Profiles Based on Genetic Algorithms , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[9]  Robert X. Gao,et al.  PCA-based feature selection scheme for machine defect classification , 2004, IEEE Transactions on Instrumentation and Measurement.

[10]  Ashutosh Kumar Singh,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2010 .

[11]  Hans-Christian Möhring,et al.  Process monitoring with sensory machine tool components , 2010 .

[12]  Ahmad Ghasemloonia,et al.  Application and comparison of an ANN-based feature selection method and the genetic algorithm in gearbox fault diagnosis , 2011, Expert Syst. Appl..

[13]  Sukhendu Das,et al.  A Survey of Decision Fusion and Feature Fusion Strategies for Pattern Classification , 2010, IETE Technical Review.

[14]  R. Tibshirani,et al.  Least angle regression , 2004, math/0406456.

[15]  Antero Arkkio,et al.  Signatures of electrical faults in the force distribution and vibration pattern of induction motors , 2006 .

[16]  F. Filippetti,et al.  AI techniques in induction machines diagnosis including the speed ripple effect , 1996 .

[17]  Stephen Lin,et al.  Graph Embedding and Extensions: A General Framework for Dimensionality Reduction , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Deng Cai,et al.  Unsupervised feature selection for multi-cluster data , 2010, KDD.

[19]  Jafar Zarei,et al.  Induction motors bearing fault detection using pattern recognition techniques , 2012, Expert Syst. Appl..

[20]  Jong-Duk Son,et al.  A Comparison of Classifier Performance for Fault Diagnosis of Induction Motor using Multi-type Signals , 2007 .

[21]  Mohamed Benbouzid,et al.  A review of induction motors signature analysis as a medium for faults detection , 1998, IECON '98. Proceedings of the 24th Annual Conference of the IEEE Industrial Electronics Society (Cat. No.98CH36200).

[22]  Robert X. Gao,et al.  Current envelope analysis for defect identification and diagnosis in induction motors , 2012 .