基于维纳过程金属化膜电容器的剩余寿命预测

摘要

金属化膜电容器是惯性约束聚变激光装置能源系统最重要的元器件,对个体电容器的剩余寿命进行有效的预测对整个装置的可靠性水平有着重要的影响.为有效地预测个体电容器的剩余寿命,提出了融合单个电容器性能退化数据与先验性能退化数据信息的预测方法.采用Wiener过程对其性能退化过程进行建模,并根据先验退化数据信息构造参数的先验分布,由Bayes方法融合个体退化数据实现剩余寿命的实时预测,通过一个实例表明了本文的研究意义和实用价值.%The metallized film capacitor is one of the most important components of the inertial confinement fusion (ICF) facility. Ihe 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.