A Bayesian-based complex system reliability analysis method

The invention discloses a Bayesian-based complex system reliability analysis method. The method sequentially comprises the following steps of: firstly, describing a complex system as an event tree diagram consisting of subsystems and components; Secondly, establishing a likelihood function of system reliability; Then establishing priori distribution, and integrating priori distribution of three types of components with accurate engineering judgment, component grouping information and no accurate engineering judgment given by a plurality of experts to obtain priori distribution of the system component; And finally, fusing priori distribution, system test data and subsystem test data of the complex system by using a Markov chain Monte Carlo method to obtain the reliability of the complex system and components. The prediction and estimation precision of the reliability of the complex system can be greatly improved. The method is suitable for the technical field of reliability engineering. 本发明公开了一种基于贝叶斯的复杂系统可靠性分析