症状推理在诊断信息融合中的应用研究

RESEARCH ON APPLYING THEORY OF EVIDENCE REASONING IN DIAGNOSIS INFORMATION FUSION

本文对诊断信息融合中的Dempster-Shafer症状推理理论进行了研究,给出了故障诊断过程中症状推理的过程描述,确定了故障识别框架和mass函数,给出了故障诊断中的证据合成规则,提出了用症状熵来描述诊断信息融合结果的可信度,最后,通过具体的诊断实例对以上的问题进行验证,取得了满意的结果。

In this paper, Dempster-Shafer theory of evidence reasoning in diagnosis information fusion is studied. The process description in evidence reasoning for fault diagnosis is shown, the fault identification framework and mass function are presented, and evidence combination rules are given. The notion of evidence entropy is proposed to be used for quantifying the belief degree of the fault diagnosis. Finally, a diagnosis example is used to demonstrate the above solution, and a satisfying result is obtained.