Remaining Lifetime Prediction by Integrating Degradation Data with Lifetime Data

摘要

Aiming at the field degradation data of the target product and the lifetime data of the similar products, remaining lifetime (RL) prediction method is put forward. Wiener parameters are assumed to obey to union conjugate normal-inverse gamma distribution. Bayesian estimation model for Wiener parameters is built based on the target product field measured degradation data. Entire likelihood functions for degradation data and lifetime data are built. Prior estimation model for hyper parameter is built. By expectation maximization (EM) algorithm, the posterior hyper parameter estimates is obtained. The accuracy and applicability of the present method is verified by an example.