ABSTRACT
This paper addresses a practical case study dealing with the analysis of membrane aging degradation in the valve of a Nuclear Power Plant. This mechanism is modeled by a discrete-state, continuous-time stochastic transport process, and state transition times are assumed to obey exponential distributions. The aim of this work is to exploit available field data to estimate the parameters of the stochastic model, giving due account to the effects of the maintenance policy on the evolution of the degradation mechanism.
Maximum Likelihood Estimation (MLE) is used for the estimation, and Fischer Information Matrix (FIM) is used to characterize the uncertainty in the estimates. The estimated parameters are used in the model to optimize the maintenance policy.
