Causal-Trivial Attention Graph Neural Network for Fault Diagnosis of Complex Industrial Processes

计算机科学 人工神经网络 分类器(UML) 因果模型 一般化 图形 人工智能 数据挖掘 理论计算机科学 机器学习 数学 统计 数学分析
作者
Hao Wang,Ruonan Liu,Steven X. Ding,Qinghua Hu,Zengxiang Li,Hongkuan Zhou
出处
期刊:IEEE Transactions on Industrial Informatics [Institute of Electrical and Electronics Engineers]
卷期号:20 (2): 1987-1996 被引量:1
标识
DOI:10.1109/tii.2023.3282979
摘要

In modern industrial systems, components have complex interactions with each other, which makes it become a challenging task to identify the operational conditions of industrial systems. Considering that an industrial system, the embedded components and their interactions can be expressed as nodes and edges in a graph, respectively. Therefore, graph representation algorithms are powerful tools for fault diagnosis of industrial systems. As one of the most commonly used graph representation algorithms, Graph Neural Networks (GNN) mainly follow the law of “learning to attend”. GNN extract training data features, learn the statistical correlations between features and labels, resulting in the attended graph favoring for accessing non-causal features as a shortcut for prediction. This shortcut feature is unstable and depends on the data distribution characteristics in the training dataset, which reduces the generalization ability of the classifier. By performing the causal analysis of GNN modeling for graph representation, the results show that shortcut features act as confounding factors between causal features and predictions, causing classifiers to learn wrong correlations. Therefore, to discover patterns of causality and weaken the confounding effects of shortcut features, a Causal-Trivial Attention Graph Neural Network (CTA-GNN) strategy is proposed. Firstly, node and edge representations are given by estimating soft masks. Secondly, through disentanglement, both causal features and shortcut features are obtained from the graph. Thirdly, the backdoor adjustment of the causal theory is parameterized to combine each causal feature with a variety of shortcut features. Finally, comparative experiments on the Three-Phase Flow Facility (TFF) dataset illustrate the effectiveness of the proposed method.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
大幅提高文件上传限制,最高150M (2024-4-1)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
发natrue发布了新的文献求助10
刚刚
wwwwww发布了新的文献求助10
刚刚
slim完成签到,获得积分10
1秒前
辞清完成签到 ,获得积分10
2秒前
筑梦完成签到 ,获得积分10
2秒前
安静的绿海完成签到,获得积分10
2秒前
viough完成签到,获得积分10
2秒前
2秒前
小酒完成签到,获得积分20
3秒前
rudjs完成签到,获得积分10
3秒前
cm2303发布了新的文献求助10
4秒前
4秒前
微笑的水桃完成签到 ,获得积分10
4秒前
Ying完成签到,获得积分10
5秒前
萧水白应助evan_L采纳,获得10
5秒前
优秀的半双完成签到,获得积分10
6秒前
6秒前
6秒前
7秒前
悠然完成签到,获得积分10
7秒前
7秒前
优雅的帽子关注了科研通微信公众号
7秒前
7秒前
YKB完成签到,获得积分20
8秒前
rafa完成签到 ,获得积分10
8秒前
8秒前
谷粱紫槐完成签到,获得积分10
8秒前
海丽完成签到 ,获得积分10
8秒前
8秒前
ljr完成签到 ,获得积分10
9秒前
company发布了新的文献求助10
10秒前
xxxgoldxsx完成签到,获得积分10
10秒前
Niki完成签到,获得积分10
11秒前
穆紫月懒阳阳完成签到,获得积分10
11秒前
书涵给书涵的求助进行了留言
11秒前
diu举报老刘不吃香菜求助涉嫌违规
11秒前
庞伟泽完成签到,获得积分10
11秒前
tivyg'lk完成签到,获得积分10
11秒前
南宫炽滔完成签到 ,获得积分10
12秒前
威武冷雪发布了新的文献求助10
12秒前
高分求助中
The late Devonian Standard Conodont Zonation 2000
The Lali Section: An Excellent Reference Section for Upper - Devonian in South China 1500
Nickel superalloy market size, share, growth, trends, and forecast 2023-2030 1000
Smart but Scattered: The Revolutionary Executive Skills Approach to Helping Kids Reach Their Potential (第二版) 1000
Mantiden: Faszinierende Lauerjäger Faszinierende Lauerjäger 800
PraxisRatgeber: Mantiden: Faszinierende Lauerjäger 800
A new species of Coccus (Homoptera: Coccoidea) from Malawi 500
热门求助领域 (近24小时)
化学 医学 生物 材料科学 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 基因 遗传学 催化作用 物理化学 免疫学 量子力学 细胞生物学
热门帖子
关注 科研通微信公众号,转发送积分 3245290
求助须知:如何正确求助?哪些是违规求助? 2888921
关于积分的说明 8256346
捐赠科研通 2557298
什么是DOI,文献DOI怎么找? 1385998
科研通“疑难数据库(出版商)”最低求助积分说明 650265
邀请新用户注册赠送积分活动 626527