Residual life prediction based on dynamic weighted Markov model and particle filtering

In order to improve the prediction accuracy of non-Gaussian data and build reasonably the prediction model, a novel residual life prediction method is proposed. A dynamic weighted Markov model is constructed by real time data and historical data, and the residual life is predicted by particle filter. The particles of the state vector are predicted and updated instantaneously using particle filter. The probability distribution of the predicted value is estimated by the updated particles. The residual life can be predicted using the set threshold of the state. This method improves the accuracy of residual life prediction. Finally, the advantage of this proposed method was shown experimentally using the bearings’ full cycle data.

[1]  James Kuria Kimotho,et al.  Machinery Prognostic Method Based on Multi-Class Support Vector Machines and Hybrid Differential Evolution – Particle Swarm Optimization , 2013 .

[2]  Ying Peng,et al.  A novel method for online health prognosis of equipment based on hidden semi-Markov model using sequential Monte Carlo methods , 2012 .

[3]  Robert Kohn,et al.  Auxiliary particle ltering within adaptive Metropolis-Hastings sampling , 2010, 1006.1914.

[4]  Jun Wu,et al.  Failure time prediction for mechanical device based on the degradation sequence , 2015, J. Intell. Manuf..

[5]  Zhengjia He,et al.  Remaining life prognostics of rolling bearing based on relative features and multivariable support vector machine , 2013 .

[6]  Xing Wang,et al.  A dynamic multi-scale Markov model based methodology for remaining life prediction , 2011 .

[7]  Rong Li,et al.  Residual-life distributions from component degradation signals: A Bayesian approach , 2005 .

[8]  Noureddine Zerhouni,et al.  CNC machine tool's wear diagnostic and prognostic by using dynamic Bayesian networks , 2012 .

[9]  Iqbal Gondal,et al.  Multi-step support vector regression and optimally parameterized wavelet packet transform for machine residual life prediction , 2013 .

[10]  Dustin G. Mixon,et al.  On a Markov‐modulated shock and wear process , 2009 .

[11]  Kwok L. Tsui,et al.  A naive Bayes model for robust remaining useful life prediction of lithium-ion battery , 2014 .

[12]  Ping Yi Chao,et al.  An improved neural network model for the prediction of cutting tool life , 1997, J. Intell. Manuf..

[13]  J. A. Hartigan,et al.  A k-means clustering algorithm , 1979 .

[14]  Steven M. Cox,et al.  Stochastic models for degradation-based reliability , 2005 .

[15]  Sami Ekici,et al.  Support vector machines models for surface roughness prediction in CNC turning of AISI 304 austenitic stainless steel , 2012, J. Intell. Manuf..

[16]  Marcos Eduardo Orchard,et al.  A Particle Filtering-based Framework for On-line Fault Diagnosis and Failure Prognosis , 2007 .

[17]  Chaochao Chen,et al.  Machine remaining useful life prediction: An integrated adaptive neuro-fuzzy and high-order particle filtering approach , 2012 .

[18]  Enrico Zio,et al.  Ensemble neural network-based particle filtering for prognostics , 2013 .

[19]  Zhigang Tian,et al.  An artificial neural network method for remaining useful life prediction of equipment subject to condition monitoring , 2012, J. Intell. Manuf..