摘要
基于符号有向图(SDG)的故障诊断方法具有良好的完备性和易于解释性,但其存在分辨率差的缺陷,为此提出基于模糊概率SDG模型和贝叶斯推理的半定量故障诊断方法.用模糊变量表示节点变量,用条件概率表(CPT)表达节点间的定性因果关系,利用贝叶斯推理和回溯搜索找出故障源候选解的集合,并对候选解进行排序.最后建立了某卫星一次电源系统的诊断模型.仿真结果表明,该方法有效地提高了诊断的分辨率,适用于航天器在轨故障诊断.
The fault diagnosis approach based on signed directed graph(SDG) has better completeness and explanation facility, and has the disadvatage of the lower diagnostic resolution. Therefore, the semi-quantitative fault diagnosis approach is proposed based on the model of fuzzy probabilistic SDG and Bayesian inference. The node variable is expressed as fuzzy variable. The cause-effect relationship between the nodes is described by conditional probabilities table (CPT). The set of failure source candidates is found out by using Bayesian inference and backtracking algorithm. Furthermore, the candidates in the set are ranked according to the rate of fault possibility. The primary electrical power supply system in certain a satellite is modeled with the proposed approach. The diagnosis simulation results show that the diagnostic resolution can be improved significantly, and the approach is feasible to be applied to on-board diagnosis for spacecraft.
出处
《控制与决策》
EI
CSCD
北大核心
2009年第5期692-696,共5页
Control and Decision
基金
国家863计划项目(2005AA735080)