In industry loading on bearing and its fault severity in an induction motor is unpredictable. Hence, in the present article wide range of vibration data from induction motor bearing surface has been taken for extraction of fault features and classified for the detection of mechanical faults presents in the bearing so that condition based monitoring possible. The vibration data which is selected in this paper includes four different kinds of loading and three different types of fault with three different fault sizes. Firstly, Wavelet Packet Transform (WPT) is applied to decompose the vibration signal and develop Bearing Damage Index (BDI) from the decomposed signal to select the useful signal from the original recorded signal. This BDI based useful signal is further applied for extraction of statistical features and fed to the classifier. Total eleven time domain features has been calculated and Principal Component Analysis (PCA) is applied for the selection of significant features. The selected features further used as input to the Dendogram Support Vector Machine (DSVM) classifier to identify the faults. This proposed method shows significant improvement in classification rate as compared to conventional method, which is quite promising and encouraging.
GuptaK., Vibration— A tool for machine diagnostics and condition monitoring, Sadhana22 (1997), 393–410. 10.1109/10.1007/BF02744480
2.
SchoenR.R., HabetlerT.G., KamranF. and BartfieldR.G., Motor bearing damage detection using stator current monitoring, in , IEEE Transactions on Industry Applications31 (6) (1995), 1274–1279. 10.1109/28.475697
3.
TrajinB., RegnierJ. and FaucherJ., Bearing fault indicator in induction machine using stator current spectral analysis, 2008 4th IET Conference on Power Electronics, Machines and Drives, York, 2008, pp. 592–596. 10.1049/cp:20080590
4.
SinghS. and KumarN., Detection of bearing faults in mechanical systems using stator current monitoring, in , IEEE Transactions on Industrial Informatics13 (3) (2017), 1341–1349. 10.1109/TII.2016.2641470
5.
McinernyS.A. and DaiY., Basic Vibration signal processing for bearing fault detection, IEEE Trans Educ46 (1) (2003), 149–156. 10.1109/TII.2016.2641470
6.
SeshadrinathJ., SinghB. and PanigrahiB.K., Vibration analysis based interturn fault diagnosis in induction machines, IEEE Trans Ind Informatics10 (1) (2014), 340–350. 10.1109/TII.2013.2271979
7.
SreejithB., VermaA.K. and SrividyaA., Fault diagnosis of rolling element bearing using time-domain features and neural networks, IEEE Int Conf ICIIS (2008) (1) 6–1110.1109/ICIINFS.2008.4798444
8.
MingA.B., QinZ.Y., ZhangW. and ChuF.L., Spectrum auto-correlation analysis and its application to fault diagnosis of rolling element bearings, Mech Syst Signal Process41 (2013), 141–154.
9.
GokhaleM.Y., Time domain signal analysis using wavelet packet decomposition approach, Int’l J Commun Netw Syst Sci3 (2010), 321–329.
10.
HeQ., Time–frequency manifold for nonlinear feature extraction in machinery fault diagnosis, Mech Syst Signal Process35 (1–2) (2013), 200–218. 10.1016/j.ymssp.2012.08.018
11.
MehalaN. and DahiyaR., Condition monitoring methods, failure identification and analysis for Induction machines, Int J circuits Syst Signal Process3 (1) (2009), 10–17.
12.
ChandelA.K. and PatelR.K., Bearing fault classification based on wavelet transform and artificial neural network, IETE J Res59 (3) (2013), 219–225.
13.
BessousN., ZouzouS.E., BentrahW., SbaaS. and SahraouiM., Diagnosis of bearing defects in induction motors using discrete wavelet transform, Int J Syst Assur Eng Manag (2016). 10.1007/s13198-016-0459-6
14.
SahraouiM., GhoggalA., GuedidiS. and ZouzouS.E., Detection of inter-turn short-circuit in induction motors using Park–Hilbert method, Int J Syst Assur Eng Manag (2013).
15.
PengZ.K., TseP.W. and ChuF.L., An improved Hilbert–Huang transform and its application in vibration signal analysis, J Sound Vib286 (1–2) (2005), 187–205. 10.1016/j.jsv.2004.10.005
16.
WangD., TsuiK.-L., TseP.W. and ZuoM.J., Principal components of super-high dimensional statistical features and support vector machine for improving identification accuracies of different gear crack levels under different working conditions, Shock Vib2015 (2015), 1–14. 10.1155/2015/420168
17.
RauberT.W., de Assis BoldtF., and VarejaoF.M., Heterogeneous feature models and feature selection applied to bearing fault diagnosis, IEEE Trans Ind Electron62(1) (2015), 637–646. 10.1109/TIE.2014.2327589
18.
LiB., ZhangP.L., TianH., MiS.S., LiuD.S. and RenG.Q., A new feature extraction and selection scheme for hybrid fault diagnosis of gearbox, Expert Syst Appl38 (8) (2011), 10000–10009. 10.1016/j.eswa.2011.02.008
19.
KonarP., SilJ. and ChattopadhyayP., Knowledge extraction using data mining for multi-class fault diagnosis of induction motor, Neurocomputing166 (2015), 14–25. 10.1016/j.neucom.2015.04.040
20.
WidodoA. and YangB.-S., Application of nonlinear feature extraction and support vector machines for fault diagnosis of induction motors, Expert Syst Appl33 (1) (2007), 241–250.
21.
KonarP. and ChattopadhyayP., Bearing fault detection of induction motor using wavelet and Support Vector Machines (SVMs), Appl Soft Comput11 (6) (2011), 4203–4211. 10.1016/j.asoc.2011.03.014
22.
BinG.F., GaoJ.J., LiX.J. and DhillonB.S., Early fault diagnosis of rotating machinery based on wavelet packets— Empirical mode decomposition feature extraction and neural network, Mech Syst Signal Process27 (2012) (2012),696–711.10.1016/j.ymssp.2011.08.002
23.
VermaN.K., SevakulaR.K., DixitS. and SalourA., Acoustic signals for air compressors, IEEE Trans ReliabPP (99) (2015), 1–19. 10.1109/TR.2015.2459684
24.
AsaithambiM., ManoharanS.C. and SubramanianS., Clustering based analysis of spirometric data using principal component analysis and self organizing map, in 4th International Conference, SEMCCO 2013, Chennai, India, December 19-21, 2013, Proceedings, Part II, 2013, pp. 523–533. 10.1007/978-3-319-03756-1_47
25.
PandyaD.H., UpadhyayS.H. and HarshaS.P., Fault diagnosis of rolling element bearing by using multinomial logistic regression and wavelet packet transform, Soft Comput18 (2) (2013), 255–266. 10.1007/s00500-013-1055-1
LiN., ZhouR., HuQ. and LiuX., Mechanical fault diagnosis based on redundant second generation wavelet packet transform, neighborhood rough set and support vector machine, Mech Syst Signal Process28 (2012), 608–621. 10.1016/j.ymssp.2011.10.016
28.
Ngoc-TuN.-T., LeeH.-H. and KwonJ.-M., Optimal feature selection using genetic algorithm for mechanical fault detection of induction motor, J Mech Sci Technol22 (3) (2008), 490–496. 10.1007/s12206-007-1036-3
29.
WidodoA., YangB.-S. and HanT., Combination of independent component analysis and support vector machines for intelligent faults diagnosis of induction motors, Expert Syst Appl32 (2) (2007), 299–312. 10.1016/j.eswa.2005.11.031
30.
VijayG.S., KumarH.S., Srinivasa PaiP., SriramN.S. and RajB.K.N.R., Evaluation of effectiveness of wavelet based denoising schemes using ANN and SVM for bearing condition classification, Computational Intelligence and Neuroscience2012 (2012), 1–12. 10.1155/2012/582453
