Residual-life estimation for components with non-symmetric priors

Abstract

Condition monitoring uses sensory signals to assess the health of engineering systems. A degradation model is a mathematical characterization of the evolution of a condition signal. Our recent research focuses on using degradation models to compute residuallife distributions for degrading components. Residual-life distributions are important for providing probabilistic estimates of failure time for use in maintenance planning and spare parts inventory management. To obtain residual-life distributions, our earlier work assumed the degradation model's stochastic parameters to be normally distributed. This paper investigates the performance of these residual-life distributions when the underlying normality assumptions are not satisfied. The paper also develops methods for estimating residual-life when the stochastic parameters of the degradation model follow more general distributions.

Residual life prediction for highly reliable products with prior accelerated degradation data

“…The acceleration factor constant principle was used to deduce 2) the relationships that the parameters of the Wiener process with a time function should satisfy. It offered a feasible approach to constructing the acceleration models for the parameters of the Wiener process.…”

“…They assumed the random parameters to obey conjugate prior distributions for mathematical tractability, and obtained the prior distributions of random parameters using the historical degradation data of the population of devices, then predicted the residual life for an individual with the real-time degradation data. Chakraborty et al [2] also studied the exponential degradation model with random parameters, but they assumed the random parameters to obey non-conjugate prior distributions, and adopted the Metropolis-Hasting algorithm to estimate the posterior means of random parameters. Gebraeel et al [6] proposed a Bayesian method, which takes the failure time data as historical information while takes the real-time degradation data of an individual as filed information, to predict the residual life distribution for a rotating machinery.…”

Failure time prediction for mechanical device based on the degradation sequence

“…Gebraeel has been doing many researches on the failure time prediction of bearings using monitoring sensory signals. The posterior estimation over different prior assumptions has been thoroughly discussed (Chakraborty et al 2009;Gebraeel and Lawley 2008;Gebraeel and Pan 2008;Gebraeel et al 2005Gebraeel et al , 2009. As an unsupervised NN, self-organized mapping (SOM) NN shows good efficacy in the state recognition of bearing applications (Huang et al 2007;Qiu et al 2003;Yu 2011).…”

Bearing Remaining Useful Life Prediction Based on a Nonlinear Wiener Process Model

“…Gebraeel et al [22] put forward an exponential degradation model for bearing RUL prediction. This model was extensively used and improved by many researchers [23][24][25][26]. However, it is restricted to the exponential degradation process and cannot describe other nonlinear degradation process.…”