Yan Hou, Jinggao Sun, Xing Liu, Ziqing Wei, Haitao Yang
{"title":"An Industrial Fault Diagnosis Method Based on Graph Attention Network","authors":"Yan Hou, Jinggao Sun, Xing Liu, Ziqing Wei, Haitao Yang","doi":"10.1021/acs.iecr.4c02220","DOIUrl":null,"url":null,"abstract":"In the field of industrial production, the precise and timely implementation of fault diagnosis methods is crucial for improving product quality, enhancing operational safety, reducing downtime, and minimizing losses. Recent studies have shown that most CNN-based fault diagnosis models are more suitable for handling Euclidean data such as images or videos but are not suitable for dealing with non-Euclidean sensor data. In practical industrial scenarios, chemical process data with imbalanced fault patterns may lead data-driven models to assign different attentions to fault patterns. The SMOTE algorithm is commonly used to generate new data, but it often tends to overfit when there are very few nearest neighbor samples. To address these issues, we designed an efficient fault diagnosis model named KRGAT. To fully utilize the spatial structural information on sensor data, we employed graph attention networks (GATs), which are well-suited for handling non-Euclidean data. Additionally, we introduced the top-k loss method to select hard samples, thereby increasing the weight of these samples. Furthermore, we improved DropMessage to enhance the model’s accuracy and robustness. Experimental results demonstrate that our model outperforms the baseline model under both balanced and imbalanced conditions.","PeriodicalId":39,"journal":{"name":"Industrial & Engineering Chemistry Research","volume":null,"pages":null},"PeriodicalIF":3.8000,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Industrial & Engineering Chemistry Research","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1021/acs.iecr.4c02220","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, CHEMICAL","Score":null,"Total":0}
引用次数: 0
摘要
在工业生产领域,准确及时地实施故障诊断方法对于提高产品质量、加强操作安全、减少停机时间和降低损失至关重要。最新研究表明,大多数基于 CNN 的故障诊断模型更适合处理图像或视频等欧几里得数据,但不适合处理非欧几里得传感器数据。在实际工业场景中,具有不平衡故障模式的化学过程数据可能会导致数据驱动模型对故障模式分配不同的关注点。SMOTE 算法常用于生成新数据,但当近邻样本极少时,该算法往往会出现过拟合。为解决这些问题,我们设计了一种名为 KRGAT 的高效故障诊断模型。为了充分利用传感器数据的空间结构信息,我们采用了图注意网络(GAT),它非常适合处理非欧几里得数据。此外,我们还引入了 top-k loss 方法来选择硬样本,从而提高这些样本的权重。此外,我们还改进了 DropMessage,以提高模型的准确性和鲁棒性。实验结果表明,在平衡和不平衡条件下,我们的模型都优于基线模型。
An Industrial Fault Diagnosis Method Based on Graph Attention Network
In the field of industrial production, the precise and timely implementation of fault diagnosis methods is crucial for improving product quality, enhancing operational safety, reducing downtime, and minimizing losses. Recent studies have shown that most CNN-based fault diagnosis models are more suitable for handling Euclidean data such as images or videos but are not suitable for dealing with non-Euclidean sensor data. In practical industrial scenarios, chemical process data with imbalanced fault patterns may lead data-driven models to assign different attentions to fault patterns. The SMOTE algorithm is commonly used to generate new data, but it often tends to overfit when there are very few nearest neighbor samples. To address these issues, we designed an efficient fault diagnosis model named KRGAT. To fully utilize the spatial structural information on sensor data, we employed graph attention networks (GATs), which are well-suited for handling non-Euclidean data. Additionally, we introduced the top-k loss method to select hard samples, thereby increasing the weight of these samples. Furthermore, we improved DropMessage to enhance the model’s accuracy and robustness. Experimental results demonstrate that our model outperforms the baseline model under both balanced and imbalanced conditions.
期刊介绍:
ndustrial & Engineering Chemistry, with variations in title and format, has been published since 1909 by the American Chemical Society. Industrial & Engineering Chemistry Research is a weekly publication that reports industrial and academic research in the broad fields of applied chemistry and chemical engineering with special focus on fundamentals, processes, and products.