【摘要】 在信度网基础上发展起来的因果图模型,克服了信度网的一些不足,具有重要的工业应用价值。经过文献犤5犦犤6犦犤7犦的研究,目前因果图已发展成一个能够处理离散变量和连续变量的混合因果图模型。文章讨论了将因果图用于复杂系统故障诊断的关键问题。给出了基于因果图的故障知识表达方法,给出了故障模式的静态求取方法和动态求取方法以及故障模式的概率计算方法。以核电站二回路系统为研究对象,在自行研制的因果图智能故障诊断平台上进行了故障诊断测试,诊断结果与实际情况相符,诊断迅速、效果较好。更多还原

【Abstract】 Causality Diagram based on Belief Network overcomes some shortages of Belief Network,so it is useful for industry application.After searched by referenc ,now Causality Diagram has become a mixed model,which can deal with discrete variable and continuous variable.This paper discusses the vital problem applying Causality Dia-gram to fault diagnosis of complex system.This paper presents the knowledge expression applying the Causality Diagram to fault diagnosis,the state method and the dynamic method acquiring fault mode,and the computation method of con-dition probability of fault mode.Taking the2th loop of the unclear plant as fault diagnosis object,on Causality Diagram intellect fault diagnosis platform developed by myself,the fault has been diagnoses.The diagnosis result coincides with the practice,diagnose rapidity,and the result is satisfied.更多还原