Residual Life Estimation by Fusing Few Failure Lifetime and Degradation Data from Real-Time Updating

Abstract

Residual life time estimation is significant for complex systems. In this paper, a method is proposed to estimate the residual life of products on satellite platform by fusing real time updating few failure lifetime and degradation data. The linear Wiener process is adopted to model the degradation data where the drift and diffusion parameters are both assumed to be random variables. Besides, the lifetime data is described by the inverse Gauss distribution. Through maximum likelihood function, the lifetime and degradation data are integrated. Considering the continually collected data from the satellite platform, a practical approach combined Bayesian idea is proposed to update the probability density function of the residual life time. Due to the complexity of the model, Monte Carlo Markov Chain (MCMC) is applied to estimate the parameters. Finally, an example of the infrared-sensitive single machine is provided to illustrate the effectiveness and validity of the proposed method.