In rotating machinery, the malfunctions of rolling bearings are one of the most common faults. To prevent machine breakdown, the pattern recognition of rolling bearing faults has been a pivotal issue for fault identification and classification. This study proposes a new feature extraction method based on ensemble empirical mode decomposition (EEMD) and singular value decomposition (SVD) for fault classification. The proposed E–S method (EEMD combined with SVD using feature parameters) intends to enhance the faults identification capability in different working conditions, including various fault types (FT), fault severities (FS), and fault loads (FL). In this study, the E–S method is adopted to analyze the simulated signals. And the experiment further discusses three cases of different FT, FS, and FL separately under six different classifiers. The experimental results show that different fault classes can be effectively distinguished by the proposed E–S in comparison with other traditional feature extraction methods. Hence, the proposed method is verified to have an effective and excellent performance in bearing fault classification.
El-ThaljiIJantunenE. A summary of fault modelling and predictive health monitoring of rolling element bearings. Mech Syst Signal Process2015; 60–61: 252–272.
2.
JardineAKSLinDBanjevicD. A review on machinery diagnostics and prognostics implementing condition-based maintenance. Mech Syst Signal Process2006; 20: 1483–1510.
3.
RandallRBAntoniJ. Rolling element bearing diagnostics—A tutorial. Mech Syst Signal Process2011; 25: 485–520.
4.
IguyonIElisseeffA. An introduction to variable and feature selection. J Mach Learn Res2003; 3: 1157–1182.
5.
YuJ. Local and nonlocal preserving projection for bearing defect classification and performance assessment. IEEE Trans Ind Electron2012; 59: 2363–2376.
6.
RajASMuraliN. Early classification of bearing faults using morphological operators and fuzzy inference. IEEE Trans Ind Electron2013; 60: 567–574.
7.
ZoubekHVillwockSPacasM. Frequency response analysis for rolling-bearing damage diagnosis. IEEE Trans Ind Electron2008; 55: 4270–4276.
8.
DingXXHeQBLuoNW. A fusion feature and its improvement based on locality preserving projections for rolling element bearing fault classification. J Sound Vib2015; 335: 367–383.
9.
LeiYHeZZiY. Application of the EEMD method to rotor fault diagnosis of rotating machinery. Mech Syst Signal Process2009; 23: 1327–1338.
10.
VanMKangH-J. Wavelet Kernel local fisher discriminant analysis with particle swarm optimization algorithm for bearing defect classification. IEEE Trans Instrum Meas2015; 64: 3588–3600.
11.
KankarPKSharmaSCHarshaSP. Fault diagnosis of ball bearings using continuous wavelet transform. Appl Soft Comput J2011; 11: 2300–2312.
12.
RauberTWde Assis BoldtFVarejaoFM. Heterogeneous feature models and feature selection applied to bearing fault diagnosis. IEEE Trans Ind Electron2015; 62: 637–646.
13.
LinJQuL. Feature extraction based on Morlet wavelet and its application for mechanical fault diagnosis. J Sound Vib2000; 234: 135–148.
14.
WangHGaoJJiangZet al.Rotating machinery fault diagnosis based on EEMD time-frequency energy and SOM neural network. Arabian J Sci Eng2014; 39: 5207–5217.
15.
SaidiLAliJBFnaiechF. Bi-spectrum based-EMD applied to the non-stationary vibration signals for bearing faults diagnosis. ISA Trans2014; 53: 1650–1660.
16.
TianYMaJLuCet al.Rolling bearing fault diagnosis under variable conditions using LMD-SVD and extreme learning machine. Mech Mach Theory2015; 90: 175–186.
17.
WangZLuCWangZet al.Fault diagnosis and health assessment for bearings using the Mahalanobis–Taguchi system based on EMD-SVD. Trans Inst Meas Control2013; 35: 798–807.
18.
JiangHMChenJDongGMet al.Study on Hankel matrix-based SVD and its application in rolling element bearing fault diagnosis. Mech Syst Signal Process2015; 52–53: 338–359.
19.
MuruganathamBSanjithMAKrishnakumarBet al.Roller element bearing fault diagnosis using singular spectrum analysis. Mech Syst Signal Process2013; 35: 150–166.
20.
ZhangSBLuSLHeQBet al.Time-varying singular value decomposition for periodic transient identification in bearing fault diagnosis. J Sound Vib2016; 379: 213–231.
21.
JiangHLiCLiH. An improved EEMD with multiwavelet packet for rotating machinery multi-fault diagnosis. Mech Syst Signal Process2013; 36: 225–239.
22.
ZhaoXYeB. Selection of effective singular values using difference spectrum and its application to fault diagnosis of headstock. Mech Syst Signal Process2011; 25: 1617–1631.
23.
ZhaoXZYeBY. Singular value decomposition packet and its application to extraction of weak fault feature. Mech Syst Signal Process2016; 70–71: 73–86.
24.
Qiao ZJ and Pan ZR. SVD principle analysis and fault diagnosis for bearings based on the correlation coefficient. Meas Sci Technol 2015; 26.
25.
SharmaAAmarnathMKankarPK. Feature extraction and fault severity classification in ball bearings. J Vib Control2016; 22: 176–192.
26.
CongFChenJDongGet al.Short-time matrix series based singular value decomposition for rolling bearing fault diagnosis. Mech Syst Signal Process2013; 34: 218–230.
27.
CongFYZhongWTongSGet al.Research of singular value decomposition based on slip matrix for rolling bearing fault diagnosis. J Sound Vib2015; 344: 447–463.
28.
Hernandez-VargasMCabal-YepezEGarcia-PerezA. Real-time SVD-based detection of multiple combined faults in induction motors. Comput Electr Eng2014; 40: 2193–2203.
29.
ŽvokeljMZupanSPrebilI. Non-linear multivariate and multiscale monitoring and signal denoising strategy using Kernel Principal Component Analysis combined with Ensemble Empirical Mode Decomposition method. Mech Syst Signal Process2011; 25: 2631–2653.
30.
XiaZXiaSWanLet al.Spectral regression based fault feature extraction for bearing accelerometer sensor signals. Sensors (Switzerland)2012; 12: 13694–13719.
31.
HeQ. Vibration signal classification by wavelet packet energy flow manifold learning. J Sound Vib2013; 332: 1881–1894.
32.
TiwariRGuptaVKKankarPK. Bearing fault diagnosis based on multi-scale permutation entropy and adaptive neuro fuzzy classifier. J Vib Control2015; 21: 461–467.
