Research on Life Prediction of Navigation Satellite Based on Bayesian Network

Navigation satellite is composed of three types of products with different failure mechanisms: random, loss and consumption. In order to solve the problems that the failure data of random product is too little and the actual internal and external structure is not considered, based on the comprehensive consideration some factors, such as product redundant backups, common-cause failure, functional backup, this paper uses Bayesian network to build the navigation satellite life prediction model based on random failure product, and present a method of predicting the remaining life of navigation satellite due to sudden failure. Combined with the life prediction results of loss and consumption failure product, the research on the navigation satellite life prediction is completed. The case shows that the accuracy and credibility of the navigation satellite life prediction can be improved by considering the redundancy backup of product, common-cause failure and functional backup.