文献类型：期刊文献

中文题名：基于不同故障传播路径差异化的故障诊断方法

英文题名：Fault Propagation Path-aware Network:A Fault Diagnosis Method

作者：谭帅[1];王一帆[1];姜庆超[1];侍洪波[1];宋冰[1]

机构：[1]华东理工大学信息科学与工程学院能源化工过程智能制造教育部重点实验室,上海200237

年份：2025

卷号：51

期号：1

起止页码：161

中文期刊名：自动化学报

外文期刊名：Acta Automatica Sinica

收录：;EI(收录号：20250617810998);北大核心:【北大核心2023】；

基金：国家自然科学基金(62273147);上海市自然科学基金(22ZR1417000)资助。

语种：中文

中文关键词：故障诊断;图神经网络;故障源图;故障根源;故障传播路径

外文关键词：Fault diagnosis;graph neural network;fault source graph;fault root cause;fault propagation path

摘要：针对工业过程中故障发生源与故障信息在传播过程中的差异性问题,提出了一种基于不同故障传播路径差异化(Fault propagation path-aware network,FPPAN)的故障诊断方法.该方法分别从故障源邻域信息关系和故障信息传播两个角度出发,设计了基于k近邻筛选(k-nearest-neighbor,k-NN)和基于剪枝的k跳可达路径选择(Pruning-based k-hop reachable path selection,k-PHop)的两种故障源图的构建方式,构建“故障源图”.从故障在变量间的差异化表现着手,将基于特征的分类问题转换为基于结构关系的图匹配问题,利用该结构化信息优化过程特征,提升模型故障诊断性能.最后,通过田纳西?伊斯曼(Tennessee-Eastman,TE)过程和某海底盾构掘进施工过程进行仿真验证,实验结果证明了所提方法的有效性.
In order to address the issue of variability of fault sources and fault information in the propagation process in industrial processes,this paper proposes a fault diagnosis method based on fault propagation path-aware network(FPPAN).The method is based on two perspectives of fault source neighbourhood information relationship and fault information propagation,and designs two ways of constructing fault source graphs based on k-nearestneighbour(k-NN)filtering and pruning-based k-hop reachable path selection(k-PHop)to construct a“fault source graph”.Based on the differentiation of faults among variables,the feature-based classification problem is viewed as a graph matching problem based on structural relationships,and the structural information is used to optimise the process features and improve the fault diagnosis performance of the model.Finally,the simulation is verified by the Tennessee-Eastman(TE)process and a submarine shield boring construction process,and the experimental results prove the effectiveness of the proposed method.