Abstract:The metallized film capacitor is one of the most important components of the inertial confinement fusion (ICF) facility. The effective residual life forecasting of capacitors affects the reliability of the whole facility. In order to forecast the residual life of a single capacitor effectively, a method which fuses prior degradation information and its own degradation data was presented. Firstly, the capacitor's degradation process was modeled with Wiener process, whose parameter's prior distribution was established on the basis of prior degradation information. Then the real-time residual life forecasting of the capacitor was proposed by Bayes method. Finally, the significance and practical value of this research was verified by an example.