分层级故障诊断策略与规则的优先权排序方法  

Multiple Hierarchical Diagnosis Strategy and Priority Ranking Method of Rules

作　　者：赖于树[1] 熊燕[2] 

机构地区：[1]重庆三峡学院电子与信息工程学院,重庆404100 [2]重庆三峡学院机械工程学院,重庆404100

出　　处：《计算机测量与控制》2012年第12期3161-3164,共4页Computer Measurement &Control

基　　金：重庆市自然科学基金项目资助(2010BB7316);重庆市教委科技项目(KJ111113)

摘　　要：针对某些复杂设备很难有效地创建基于元器件层次分解的诊断树的特点,提出基于故障类别信息的分层级故障诊断推理策略,给出相应的规则存储模式并引入规则优先权;然后提出基于模糊多属性群决策理论的规则优先权排序方法;最后给出该诊断推理策略的可信度的传递算法和具体推理流程;应用实例和分析表明,所提出的方法在冲突消解方面具有很好的效果,可有效提升诊断效率。Abstract~ In equipment fault diagnosis, a multiple hierarchical diagnosis method based on fault categories is proposed, considering the very difficult circumstances to achieve the fault trees based on components framework. Accordingly, the storage model of rules is designed, and the concept of rule priority is introduced. Then, a priority ranking method based on fuzzy multi--attribute group decision making method is presented. Finally, the inexact reasoning flow and the algorithm of confidence level transfer based on the reasoning strategy and method are presented. The application case and analysis show that, the proposed method has good performance to conflict resolution, can improve diagnosis efficiency markedly.

关 键 词：故障诊断 不确定性推理 分层级诊断 模糊多属性群决策 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]