Please wait a minute...

Frontiers of Mechanical Engineering

Front. Mech. Eng.    2018, Vol. 13 Issue (2) : 301-310     https://doi.org/10.1007/s11465-017-0449-7
RESEARCH ARTICLE
An adaptive data-driven method for accurate prediction of remaining useful life of rolling bearings
Yanfeng PENG1,2,3, Junsheng CHENG1,2(), Yanfei LIU3, Xuejun LI3, Zhihua PENG4
1. State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha 410082, China
2. College of Mechanical and Vehicle Engineering, Hunan University, Changsha 410082, China
3. Hunan Provincial Key Laboratory of Health Maintenance for Mecha- nical Equipment, Hunan University of Science and Technology, Xiangtan 411201, China
4. School of Mathematics and Physics, University of South China, Hengyang 421001, China
Download: PDF(455 KB)   HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks
Abstract

A novel data-driven method based on Gaussian mixture model (GMM) and distance evaluation technique (DET) is proposed to predict the remaining useful life (RUL) of rolling bearings. The data sets are clustered by GMM to divide all data sets into several health states adaptively and reasonably. The number of clusters is determined by the minimum description length principle. Thus, either the health state of the data sets or the number of the states is obtained automatically. Meanwhile, the abnormal data sets can be recognized during the clustering process and removed from the training data sets. After obtaining the health states, appropriate features are selected by DET for increasing the classification and prediction accuracy. In the prediction process, each vibration signal is decomposed into several components by empirical mode decomposition. Some common statistical parameters of the components are calculated first and then the features are clustered using GMM to divide the data sets into several health states and remove the abnormal data sets. Thereafter, appropriate statistical parameters of the generated components are selected using DET. Finally, least squares support vector machine is utilized to predict the RUL of rolling bearings. Experimental results indicate that the proposed method reliably predicts the RUL of rolling bearings.

Keywords Gaussian mixture model      distance evaluation technique      health state      remaining useful life      rolling bearing     
Corresponding Author(s): Junsheng CHENG   
Just Accepted Date: 07 June 2017   Online First Date: 07 July 2017    Issue Date: 19 March 2018
 Cite this article:   
Yanfeng PENG,Junsheng CHENG,Yanfei LIU, et al. An adaptive data-driven method for accurate prediction of remaining useful life of rolling bearings[J]. Front. Mech. Eng., 2018, 13(2): 301-310.
 URL:  
http://journal.hep.com.cn/fme/EN/10.1007/s11465-017-0449-7
http://journal.hep.com.cn/fme/EN/Y2018/V13/I2/301
Fig.1  Illustration of GMM
Fig.2  Flowchart of the proposed method
Fig.3  Bearing test rig
Fig.4  Vibration signals of three data sets. (a) Data set 1; (b) data set 2; (c) data set 3
Data setTestBearingBreak time/minDegradation typeMaximum magnitude/(m·s–2)
11321560Inner race5
21421560Roller4
3219840Outer race5
Tab.1  Information of the experimental data sets
Fig.5  Clustering results of data set 1. (a) Clustered health state; (b) skewness
Fig.6  Clustering results of data set 2. (a) Clustered health state; (b) skewness
Fig.7  Clustering results of data set 3. (a) Clustered health state; (b) skewness
Fig.8  Salient features of data set 1
Fig.9  Salient features of data set 2
Fig.10  Salient features of data set 3
Fig.11  RUL prediction results of data set 1. (a) Method 1; (b) Method 2; (c) Method 3; (d) Method 4
Fig.12  RUL prediction results of data set 2. (a) Method 1; (b) Method 2; (c) Method 3; (d) Method 4
Fig.13  RUL prediction results of data set 3. (a) Method 1; (b) Method 2; (c) Method 3; (d) Method 4
MethodCA1CA2CA3CAEECC
11.00001.00000.65830.94290.02580.9482
20.89120.6750 0.16740.57800.05170.8970
30.38260.92670.16670.76040.03710.9127
41.00000.99080.64170.93180.03320.9325
Tab.2  Experimental results of data set 1
MethodCA1CA2CA3CA4CA5CAEECC
11.00000.99440.97600.96150.97660.98600.00480.9904
20.86010.13890.90280.54550.98610.68660.02880.9484
30.94231.00000.94400.21150.75780.88550.00940.9812
41.00000.99440.93600.92310.98440.97770.00620.9877
Tab.3  Experimental results of data set 2
MethodCA1CA2CA3CAEECC
11.00001.00000.47830.96330.01760.9663
20.80730.83490.13760.59330.03190.9368
31.00000.31500.21740.67890.02740.9484
41.00000.98430.26090.94190.02160.9588
Tab.4  Experimental results of data set 3
1 Marble S, Morton B P. Predicting the remaining life of propulsion system bearings. In: Proceedings of IEEE Aerospace Conference. IEEE, 2006, 1–8
https://doi.org/10.1109/AERO.2006.1656121
2 Liao H, Zhao W, Guo H. Predicting remaining useful life of an individual unit using proportional hazards model and logistic regression model. In: Proceedings of IEEE Annual Reliability and Maintainability Symposium Conference. Newport Beach: IEEE, 2006, 127–132
https://doi.org/10.1109/RAMS.2006.1677362
3 Tian Z, Liao H. Condition based maintenance optimization for multi-component systems using proportional hazards model. Reliability Engineering & System Safety, 2011, 96(5): 581–589
https://doi.org/10.1016/j.ress.2010.12.023
4 Sikorska J Z, Hodkiewicz M, Ma L. Prognostic modelling options for remaining useful life estimation by industry. Mechanical Systems and Signal Processing, 2011, 25(5): 1803–1836
https://doi.org/10.1016/j.ymssp.2010.11.018
5 Gebraeel N Z, Lawley M A, Liu R, et al.. Residual life predictions from vibration-based degradation signals: A neural network approach. IEEE Transactions on Industrial Electronics, 2004, 51(3): 694–700
https://doi.org/10.1109/TIE.2004.824875
6 Di Maio F, Tsui K L, Zio E. Combining relevance vector machines and exponential regression for bearing residual life estimation. Mechanical Systems and Signal Processing, 2012, 31(1): 405–427
https://doi.org/10.1016/j.ymssp.2012.03.011
7 Ben Ali J, Chebel-Morello B, Saidi L, et al.. Accurate bearing remaining useful life prediction based on Weibull distribution and artificial neural network. Mechanical Systems and Signal Processing, 2015, 56–57: 150–172
https://doi.org/10.1016/j.ymssp.2014.10.014
8 Pan D, Liu J, Cao J. Remaining useful life estimation using an inverse Gaussian degradation model. Neurocomputing, 2016, 185: 64–72
https://doi.org/10.1016/j.neucom.2015.12.041
9 Zhao M, Tang B, Tan Q. Bearing remaining useful life estimation based on time-frequency representation and supervised dimensionality reduction. Measurement, 2016, 86: 41–55
https://doi.org/10.1016/j.measurement.2015.11.047
10 Chen C, Vachtsevanos G, Orchard M E. Machine remaining useful life prediction: An integrated adaptive neuro-fuzzy and high-order particle filtering approach. Mechanical Systems and Signal Processing, 2012, 28: 597–607
https://doi.org/10.1016/j.ymssp.2011.10.009
11 Lu C, Chen J, Hong R, et al.. Degradation trend estimation of slewing bearing based on LSSVM model. Mechanical Systems and Signal Processing, 2016, 76–77: 353–366
https://doi.org/10.1016/j.ymssp.2016.02.031
12 Loutas T H, Roulias D, Georgoulas G. Remaining useful life estimation in rolling bearings utilizing data-driven probabilistic e-support vectors regression. IEEE Transactions on Reliability, 2013, 62(4): 821–832
https://doi.org/10.1109/TR.2013.2285318
13 Khanmohammadi S, Chou C A. A Gaussian mixture model based discretization algorithm for associative classification of medical data. Expert Systems with Applications, 2016, 58: 119–129
https://doi.org/10.1016/j.eswa.2016.03.046
14 Elguebaly T, Bouguila N. Simultaneous high-dimensional clustering and feature selection using asymmetric Gaussian mixture models. Image and Vision Computing, 2015, 34: 27–41
https://doi.org/10.1016/j.imavis.2014.10.011
15 Yu J. Bearing performance degradation assessment using locality preserving projections and Gaussian mixture models. Mechanical Systems and Signal Processing, 2011, 25(7): 2573–2588
https://doi.org/10.1016/j.ymssp.2011.02.006
16 Heyns T, Heyns P S, de Villiers J P. Combining synchronous averaging with a Gaussian mixture model novelty detection scheme for vibration-based condition monitoring of a gearbox. Mechanical Systems and Signal Processing, 2012, 32: 200–215
https://doi.org/10.1016/j.ymssp.2012.05.008
17 Yang B S, Han T, Huang W W. Fault diagnosis of rotating machinery based on multi-class support vector machines. Journal of Mechanical Science and Technology, 2005, 19(3): 846–859
https://doi.org/10.1007/BF02916133
18 Zeng M, Yang Y, Zheng J, et al.. Maximum margin classification based on flexible convex hulls for fault diagnosis of roller bearings. Mechanical Systems and Signal Processing, 2016, 66–67: 533–545
https://doi.org/10.1016/j.ymssp.2015.06.006
19 Lei Y, He Z, Zi Y, et al.. New clustering algorithm-based fault diagnosis using compensation distance evaluation technique. Mechanical Systems and Signal Processing, 2008, 22(2): 419–435
https://doi.org/10.1016/j.ymssp.2007.07.013
20 Choi S W, Park J H, Lee I B. Process monitoring using a Gaussian mixture model via principal component analysis and discriminant analysis. Computers & Chemical Engineering, 2004, 28(8): 1377–1387
https://doi.org/10.1016/j.compchemeng.2003.09.031
21 Lei Y, Lin J, He Z, et al.. A review on empirical mode decomposition in fault diagnosis of rotating machinery. Mechanical Systems and Signal Processing, 2013, 35(1–2): 108–126
https://doi.org/10.1016/j.ymssp.2012.09.015
22 Gai G. The processing of rotor startup signals based on empirical mode decomposition. Mechanical Systems and Signal Processing, 2006, 20(1): 222–235
https://doi.org/10.1016/j.ymssp.2004.07.001
23 Huang N E, Zheng S, Long S R, et al.. The empirical mode decomposition and the Hilbert spectrum for non linear and non-stationary time series analysis. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 1998, 454(1971): 903–995
https://doi.org/10.1098/rspa.1998.0193
24 Yeh M H. The complex bidimensional empirical mode decomposition. Signal Processing, 2012, 92(2): 523–541
https://doi.org/10.1016/j.sigpro.2011.08.019
25 NASA. IMS bearings data set. 2014. Retrieved from
26 Qiu H, Lee J, Lin J, et al.. Robust performance degradation assessment methods for enhanced rolling element bearing prognostics. Advanced Engineering Informatics, 2003, 17(3–4): 127–140
https://doi.org/10.1016/j.aei.2004.08.001
27 Qiu H, Lee J, Lin J, et al.. Wavelet filter-based weak signature detection method and its application on rolling element bearing prognostics. Journal of Sound and Vibration, 2006, 289(4–5): 1066–1090
https://doi.org/10.1016/j.jsv.2005.03.007
Related articles from Frontiers Journals
[1] Lie SUN,Ang LI. Rolling-element bearings in China: From ancient times to the 20th century[J]. Front. Mech. Eng., 2016, 11(1): 33-43.
[2] Yulun CHI, Haolin LI. Simulation and analysis of grinding wheel based on Gaussian mixture model[J]. Front Mech Eng, 2012, 7(4): 427-432.
Viewed
Full text


Abstract

Cited

  Shared   0
  Discussed