Incipient Inter-turn Fault Diagnosis in Induction motors using CNN and LSTM based Methods

Induction machines are an integral part of any major industry or production process. Incipient fault diagnosis is an important topic which aims at detecting the fault at an early stage and isolating them from other ambiguous conditions. In this work, an analytical model for inter-turn fault diagnosis in induction machines has been developed. A methodology for early diagnosis of fault has been envisaged, in presence of ambiguous conditions such as voltage imbalances and load variations. The novel method is based on motor current signature analysis (MCSA), using deep learning based one dimensional convolutional neural network(1D-CNN) model and long short term model(LSTM). The results using these two methods have been compared, and this initial investigation shows that CNN is found to be more suitable than LSTM, for incipient fault diagnosis.

[1]  Nikhil R. Pal,et al.  On minimum cross-entropy thresholding , 1996, Pattern Recognit..

[2]  Sung-Bae Cho,et al.  Human activity recognition with smartphone sensors using deep learning neural networks , 2016, Expert Syst. Appl..

[3]  Bijaya K. Panigrahi,et al.  Incipient Turn Fault Detection and Condition Monitoring of Induction Machine Using Analytical Wavelet Transform , 2014, IEEE Transactions on Industry Applications.

[4]  Steven X. Ding,et al.  A Survey of Fault Diagnosis and Fault-Tolerant Techniques—Part I: Fault Diagnosis With Model-Based and Signal-Based Approaches , 2015, IEEE Transactions on Industrial Electronics.

[5]  Bijaya K. Panigrahi,et al.  Vibration Analysis Based Interturn Fault Diagnosis in Induction Machines , 2014, IEEE Transactions on Industrial Informatics.

[6]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[7]  W. T. Thomson,et al.  Current signature analysis to detect induction motor faults , 2001 .

[8]  Tara N. Sainath,et al.  Improving deep neural networks for LVCSR using rectified linear units and dropout , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[9]  Honglak Lee,et al.  An Analysis of Single-Layer Networks in Unsupervised Feature Learning , 2011, AISTATS.

[10]  Hang Li,et al.  Convolutional Neural Network Architectures for Matching Natural Language Sentences , 2014, NIPS.

[11]  Keiron O'Shea,et al.  An Introduction to Convolutional Neural Networks , 2015, ArXiv.

[12]  H.A. Toliyat,et al.  Condition Monitoring and Fault Diagnosis of Electrical Motors—A Review , 2005, IEEE Transactions on Energy Conversion.

[13]  Andrew W. Senior,et al.  Long short-term memory recurrent neural network architectures for large scale acoustic modeling , 2014, INTERSPEECH.

[14]  Xiaoming Xue,et al.  A hybrid fault diagnosis approach based on mixed-domain state features for rotating machinery. , 2017, ISA transactions.

[15]  Toshio Fukuda,et al.  Theory and applications of neural networks for industrial control systems , 1992, IEEE Trans. Ind. Electron..

[16]  Thomas G. Habetler,et al.  Transient model for induction machines with stator winding turn faults , 2000, Conference Record of the 2000 IEEE Industry Applications Conference. Thirty-Fifth IAS Annual Meeting and World Conference on Industrial Applications of Electrical Energy (Cat. No.00CH37129).

[17]  M. E. H. Benbouzid,et al.  What Stator Current Processing Based Technique to Use for Induction Motor Rotor Faults Diagnosis , 2002, IEEE Power Engineering Review.

[18]  Yurong Liu,et al.  A survey of deep neural network architectures and their applications , 2017, Neurocomputing.

[19]  Lukás Burget,et al.  Extensions of recurrent neural network language model , 2011, 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[20]  Giovanni Franceschini,et al.  On-field experience with online diagnosis of large induction motors cage failures using MCSA , 2002 .