文献类型：期刊文献

英文题名：Dynamic-scale graph neural network for fault detection

作者：Lin, Zhengqing[1];Hu, Zhengwei[1];Peng, Jingchao[1];Zhao, Haitao[1]

机构：[1]East China Univ Sci & Technol, Sch Informat Sci & Engn, Automat Dept, Shanghai 200237, Peoples R China

年份：2022

卷号：168

起止页码：953

外文期刊名：PROCESS SAFETY AND ENVIRONMENTAL PROTECTION

收录：;EI(收录号：20225013255520);WOS:【SCI-EXPANDED(收录号:WOS:000892582300002)】；

基金：This work is supported by National Natural Science Foundation of China (NSFC) under Grant 62173143 and 61973122.

语种：英文

外文关键词：Process monitoring; Fault detection; Dynamic feature extraction; Dimension reduction

摘要：Traditional graph-based dynamic fault detection methods describe the dynamic characteristic through con-structing a single neighborhood graph at the current sample with some history samples. However, they ignore the diversity of dynamic properties of the variables in complex chemical processes. To overcome this problem, a novel neural network structure combining multiscale subgraphs is proposed, named dynamic-scale graph neural network (DSGNN), which divides variables into multiple groups according to their dynamic properties. DSGNN constructs a subgraph in each group. In traditional graph-based methods, the scale of the graph is usually manually designed. In DSGNN, the scale of each subgraph is decided by the dynamic properties of the variables in this subgraph. To aggregate the dynamic information, DSGNN utilizes convolution operations. The weights assigned to the neighbors in each subgraph are determined according to the similarity between the current data and its neighbors. Low-dimensional features are extracted through the back-propagation technique from the updated high-dimensional features produced by convolution operations. Two case studies on a multivariate dynamic process and the Tennessee Eastman process are conducted to show the superiority of DSGNN.