基于多源信息融合故障树与模糊Petri网的复杂系统故障诊断方法

发布时间：2018-06-11 13:36

本文选题：复杂系统 + 故障诊断　；参考：《计算机集成制造系统》2017年08期

【摘要】：针对复杂系统故障树模型构建困难且模型冗余节点多、计算复杂的问题,提出一种基于多源信息融合故障树与模糊Petri网的故障诊断方法。该方法先将多源信息进行标准化处理,从处理后的信息中提取维修元数据,同时利用数据挖掘方法得到故障关联项集。通过维修元数据、故障关联项集和系统结构关系的映射、融合,更加全面、准确地构建复杂系统故障树模型。采用模糊Petri网对多源信息融合故障树模型进行简化和改进,并利用基于模糊Petri网的动态故障推理方法和基于关联矩阵的最小割集求解方法建立复杂系统故障诊断方法,提高了故障的诊断速度与推理效率。以汽车发动机故障诊断过程为例,证明了所提方法的合理性和有效性。
[Abstract]:A fault diagnosis method based on multi-source information fusion fault tree and fuzzy Petri net is proposed to solve the complex problem of complex system fault tree model with more redundant nodes and complicated computation. In this method, the multi-source information is standardized and the maintenance metadata is extracted from the processed information, and the fault association item set is obtained by using the data mining method. Through the mapping and fusion of maintenance metadata, fault association item set and system structure, a more comprehensive and accurate fault tree model of complex system is constructed. The fault tree model of multi-source information fusion is simplified and improved by using fuzzy Petri net, and the fault diagnosis method of complex system is established by using the dynamic fault reasoning method based on fuzzy Petri net and the minimum cut set solution method based on correlation matrix. The fault diagnosis speed and reasoning efficiency are improved. Taking the process of automobile engine fault diagnosis as an example, the rationality and effectiveness of the proposed method are proved.
【作者单位】：西南交通大学制造业产业链协同与信息化支撑技术四川省重点实验室;西南交通大学四川省现代服务科技工程技术研究中心;
【基金】：国家科技支撑计划资助项目(2015BAF32B05) 四川省科技支撑计划资助项目(2015GZ0076)~~
【分类号】：TP301.1;U472.9

【相似文献】