Effective degradation indicator and robust prediction model are very important for residual life prediction. Thus a new residual life prediction based on Markov indicator and support vector is proposed. Since the Markov model is good at dealing with stochastic characteristics in time domain, Markov model is joined with multiple fault features for the construction of an effective degradation indicator of rolling element bearings. The support vector regression is used to construct an adaptive prediction model composed of two prediction models that are, respectively, based on historical data and online data. Thus the ultimate prediction result is obtained by taking a weighted average of the two prediction results captured by the two prediction models, and the weights are adjusted by the LMS to enhance the prediction accuracy. The experimental results show that the Markov indicator is more sensitive than the common features, and the proposed prediction method is more effective in comparison to other methods.
Rolling element bearings Degradation indicator Markov model Support vector regression Adaptive prediction LMS
This is a preview of subscription content, log in to check access.
The work described in this paper was supported by a grant from the National Defence Researching Fund (No. 9140A27020413JB11076).
H. Qiu, J. Lee, J. Lin et al., Robust performance degradation assessment methods for enhanced rolling element bearing prognostics. Adv. Eng. Inform. 17(3), 127–140 (2003)CrossRefGoogle Scholar
J. Yan, C. Guo, X. Wang, A dynamic multi-scale Markov model based methodology for remaining life prediction. Mech. Syst. Signal Process. 25(4), 1364–1376 (2011)CrossRefGoogle Scholar
J. Yu, Bearing performance degradation assessment using locality preserving projections and Gaussian mixture models. Mech. Syst. Signal Process. 25(7), 2573–2588 (2011)CrossRefGoogle Scholar
L. Liao, Discovering prognostic features using genetic programming in remaining useful life prediction. IEEE Trans. Ind. Electron. 61(5), 2464–2472 (2014)CrossRefGoogle Scholar
C. Chen, G. Vachtsevanos, M.E. Orchard, Machine residual useful life prediction: An integrated adaptive neuro-fuzzy and high-order particle filtering approach. Mech. Syst. Signal Process. 28, 597–607 (2012)CrossRefGoogle Scholar
Z. Tian, An artificial neural network method for remaining useful life prediction of equipment subject to condition monitoring. J. Intell. Manuf. 23(2), 227–237 (2012)CrossRefGoogle Scholar
P.Y. Chao, Y.D. Hwang, An improved neural network model for the prediction of cutting tool life. J. Intell. Manuf. 8(2), 107–115 (1997)CrossRefGoogle Scholar
N.Z. Gebraeel, M.A. Lawley, R. Li et al., Residual-life distributions from component degradation signals: a Bayesian approach. IIE Trans. 37(6), 543–557 (2005)CrossRefGoogle Scholar
J.P. Kharoufeh, D.G. Mixon, On a Markov-modulated shock and wear process. Naval Res. Logist. (NRL) 56(6), 563–576 (2009)CrossRefGoogle Scholar
Orchard M E. AParticle filtering-based framework for on-line fault diagnosis and failure prognosis. Georgia Institute of Technology, 2007Google Scholar
P. Baraldi, M. Compare, S. Sauco et al., Ensemble neural network-based particle filtering for prognostics. Mech. Syst. Signal Process. 41(1), 288–300 (2013)CrossRefGoogle Scholar
M.S. Arulampalam, S. Maskell, N. Gordon et al., A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking. IEEE Trans. Sig. Process. 50(2), 174–188 (2002)CrossRefGoogle Scholar
J.M. Lee, S.J. Kim, Y. Hwang et al., Diagnosis of mechanical fault signals using continuous hidden Markov model. J. Sound Vib. 276(3), 1065–1080 (2004)CrossRefGoogle Scholar
S. Zhou, J. Zhang, S. Wang, Fault diagnosis in industrial processes using principal component analysis and hidden Markov model[C]//American Control Conference. Proc. 2004 IEEE 2004(6), 5680–5685 (2004)Google Scholar
B.S. Yang, A. Widodo, Support vector machine for machine fault diagnosis and prognosis. J. Syst. Des. Dyn. 2, 12–23 (2008)Google Scholar
T. Benkedjouh, K. Medjaher, N. Zerhouni et al., Remaining useful life estimation based on nonlinear feature reduction and support vector regression. Eng. Appl. Artif. Intell. 26(7), 1751–1760 (2013)CrossRefGoogle Scholar
R.I.A. Davis, B.C. Lovell, Comparing and evaluating HMM ensemble training algorithms using train and test and condition number criteria. Formal Pattern Anal. Appl. 6(4), 327–335 (2004)Google Scholar
D. Basak, S. Pal, D.C. Patranabis, Support vector regression. Neural Inf. Process. Lett. Rev. 11(10), 203–224 (2007)Google Scholar