摘要
为了克服基于模型的诊断方法不能确定诊断最优解的缺陷,提出以基于模型的诊断所生成的最小诊断集构建贝叶斯网络的诊断方法,它能通过贝叶斯网络确定候选诊断解的发生概率,进而确定系统最有可能的诊断解。并针对当最小诊断解较多时,构建的贝叶斯网络将非常复杂,不易诊断的问题,提出了超件的概念,并将超件应用于贝叶斯网络构建,以简化构建的贝叶斯网络,提高诊断效率。最后以某电路为例进行诊断分析。
In order to overcome the defect that optimal diagnostic results can＇t be confirmed by model-based diagnosis（MBD）,the diagnostic method that minimal diagnosis set produced by model-based diagnosis is utilized to construct Bayesian Network is presented.It can confirm the probability of each candidate diagnostic results,then confirm optimal diagnostic results of the system.When the number of minimal diagnosis set is big,Bayesian Network constructed will be complex,which will be difficult to get ultimate diagnostic results,so super-component is presented and applied to simplify Bayesian Network,thus the efficiency of diagnosis is improved.Finally,a certain circuit is taken for example to analyze.
出处
《微计算机信息》
2010年第28期114-115,61,共3页
Control & Automation
关键词
贝叶斯网络
最小诊断集
故障诊断
诊断模型
Bayesian Network
minimal diagnosis sets
fault diagnosis
diagnostic method
