{"title":"基于多传感器特征融合的多分支卷积关注网络在旋转机械智能故障诊断中的应用","authors":"Ke Wu, Zirui Li, Chong Chen, Zhenguo Song, Jun Wu","doi":"10.1080/08982112.2023.2257762","DOIUrl":null,"url":null,"abstract":"AbstractMulti-sensor data fusion approaches based on deep learning are widely used for fault diagnosis of rotating machinery. Massive sensor data bring not only abundant information but also technical challenges for the fault diagnosis. The main challenge is how to discriminate the discrepancies between multi-sensor data and efficiently fuze these sensor data to improve diagnostic performance. To overcome the challenge, a novel multi-branch convolutional attention network (MBCAN) is proposed for fault diagnosis of the rotating machinery. In this method, a feature extractor with attention mechanism is established to extract fault-related features from the sensor data and reduce irrelevant noise interference hidden in the sensor data. Meanwhile, a multi-branch convolutional pathway is designed to enrich the features. Furthermore, a feature fusion module is constructed by defining an adaptive weighted fusion rule to fuze the extracted features. The performance of the MBCAN is verified on gearbox and centrifugal pump under different noise environments. The experimental results show that the proposed MBCAN has more excellent anti-noise ability and reliable diagnostic performance than other existing approaches.Keywords: convolutional neural networkfault diagnosisfeature fusionrotating machinery Disclosure statementNo potential conflict of interest was reported by the author(s).Additional informationFundingThis work was supported in part by the National Natural Science Foundation of China under Grant No. 51875225, in part by the Ministry of Industry and Information Technology of China under Grant No. TC210804R-1, and in part by Hubei Provincial Natural Science Foundation for Innovation Groups under Grant No. 2021CFA026.Notes on contributorsKe WuKe Wu received his B.S. and M.S. degrees in mechanical engineering from Hunan University of Science and Technology, in 2017 and 2020, respectively. He is currently a Ph.D candidate at marine engineering with the School of Naval Architecture and Ocean Engineering from Huazhong University of Science and Technology, China. His main research interests include big data analytics, health monitoring and fault diagnosis for equipment.Zirui LiZirui Li received the B.S. degree in marine engineering from the Huazhong University of Science and Technology (HUST), China, in 2020, where he is currently pursuing the master’s degree with the School of Naval Architecture and Ocean Engineering at HUST. His main research interests include health monitoring for equipment, deep learning and reinforcement learning.Chong ChenChong Chen received his B.S. degree in Nuclear science and technology from Harbin Engineering University, China, in 2011, and received his M.S. degrees in Nuclear science and technology from Harbin Engineering University, in 2013. His Ph.D degree in Nuclear science and technology with the school of Fundamental Science on Nuclear Safety and Simulation Technology Laboratory from Harbin Engineering University. His main research interests include reactor thermal-hydraulic and big data analytics.Zhenguo SongZhenguo Song received his B.S. and M.S. degree in marine engineering from Dalian Maritime University, China. He is currently a senior engineer at China Ship Development and Design Center. His main research interests include Integrated design of ship dynamics and big data analytics.Jun WuJun Wu received his B.S. degree in mechanical engineering from Hubei University of Technology, China, in 2001, and received his M.S. and Ph.D. degrees in mechanical engineering from Huazhong University of Science and Technology (HUST), in 2004 and 2008, respectively. He is currently a full Professor of School of Naval Architecture and Ocean Engineering at HUST. He worked as a visiting scholar at Stanford University, CA, USA from 2014 to 2015, and 2019, where he conducted technical research in the area of structure health monitoring. His research interests include equipment health monitoring, fault diagnosis and remaining useful life prediction. He has more than 80 publications and the award of 12 patents, and receives several awards for his teaching activities.","PeriodicalId":20846,"journal":{"name":"Quality Engineering","volume":"49 1","pages":"0"},"PeriodicalIF":1.3000,"publicationDate":"2023-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Multi-branch convolutional attention network for multi-sensor feature fusion in intelligent fault diagnosis of rotating machinery\",\"authors\":\"Ke Wu, Zirui Li, Chong Chen, Zhenguo Song, Jun Wu\",\"doi\":\"10.1080/08982112.2023.2257762\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"AbstractMulti-sensor data fusion approaches based on deep learning are widely used for fault diagnosis of rotating machinery. Massive sensor data bring not only abundant information but also technical challenges for the fault diagnosis. The main challenge is how to discriminate the discrepancies between multi-sensor data and efficiently fuze these sensor data to improve diagnostic performance. To overcome the challenge, a novel multi-branch convolutional attention network (MBCAN) is proposed for fault diagnosis of the rotating machinery. In this method, a feature extractor with attention mechanism is established to extract fault-related features from the sensor data and reduce irrelevant noise interference hidden in the sensor data. Meanwhile, a multi-branch convolutional pathway is designed to enrich the features. Furthermore, a feature fusion module is constructed by defining an adaptive weighted fusion rule to fuze the extracted features. The performance of the MBCAN is verified on gearbox and centrifugal pump under different noise environments. The experimental results show that the proposed MBCAN has more excellent anti-noise ability and reliable diagnostic performance than other existing approaches.Keywords: convolutional neural networkfault diagnosisfeature fusionrotating machinery Disclosure statementNo potential conflict of interest was reported by the author(s).Additional informationFundingThis work was supported in part by the National Natural Science Foundation of China under Grant No. 51875225, in part by the Ministry of Industry and Information Technology of China under Grant No. TC210804R-1, and in part by Hubei Provincial Natural Science Foundation for Innovation Groups under Grant No. 2021CFA026.Notes on contributorsKe WuKe Wu received his B.S. and M.S. degrees in mechanical engineering from Hunan University of Science and Technology, in 2017 and 2020, respectively. He is currently a Ph.D candidate at marine engineering with the School of Naval Architecture and Ocean Engineering from Huazhong University of Science and Technology, China. His main research interests include big data analytics, health monitoring and fault diagnosis for equipment.Zirui LiZirui Li received the B.S. degree in marine engineering from the Huazhong University of Science and Technology (HUST), China, in 2020, where he is currently pursuing the master’s degree with the School of Naval Architecture and Ocean Engineering at HUST. His main research interests include health monitoring for equipment, deep learning and reinforcement learning.Chong ChenChong Chen received his B.S. degree in Nuclear science and technology from Harbin Engineering University, China, in 2011, and received his M.S. degrees in Nuclear science and technology from Harbin Engineering University, in 2013. His Ph.D degree in Nuclear science and technology with the school of Fundamental Science on Nuclear Safety and Simulation Technology Laboratory from Harbin Engineering University. His main research interests include reactor thermal-hydraulic and big data analytics.Zhenguo SongZhenguo Song received his B.S. and M.S. degree in marine engineering from Dalian Maritime University, China. He is currently a senior engineer at China Ship Development and Design Center. His main research interests include Integrated design of ship dynamics and big data analytics.Jun WuJun Wu received his B.S. degree in mechanical engineering from Hubei University of Technology, China, in 2001, and received his M.S. and Ph.D. degrees in mechanical engineering from Huazhong University of Science and Technology (HUST), in 2004 and 2008, respectively. He is currently a full Professor of School of Naval Architecture and Ocean Engineering at HUST. He worked as a visiting scholar at Stanford University, CA, USA from 2014 to 2015, and 2019, where he conducted technical research in the area of structure health monitoring. His research interests include equipment health monitoring, fault diagnosis and remaining useful life prediction. He has more than 80 publications and the award of 12 patents, and receives several awards for his teaching activities.\",\"PeriodicalId\":20846,\"journal\":{\"name\":\"Quality Engineering\",\"volume\":\"49 1\",\"pages\":\"0\"},\"PeriodicalIF\":1.3000,\"publicationDate\":\"2023-09-19\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Quality Engineering\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1080/08982112.2023.2257762\",\"RegionNum\":4,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"ENGINEERING, INDUSTRIAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Quality Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1080/08982112.2023.2257762","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENGINEERING, INDUSTRIAL","Score":null,"Total":0}
Multi-branch convolutional attention network for multi-sensor feature fusion in intelligent fault diagnosis of rotating machinery
AbstractMulti-sensor data fusion approaches based on deep learning are widely used for fault diagnosis of rotating machinery. Massive sensor data bring not only abundant information but also technical challenges for the fault diagnosis. The main challenge is how to discriminate the discrepancies between multi-sensor data and efficiently fuze these sensor data to improve diagnostic performance. To overcome the challenge, a novel multi-branch convolutional attention network (MBCAN) is proposed for fault diagnosis of the rotating machinery. In this method, a feature extractor with attention mechanism is established to extract fault-related features from the sensor data and reduce irrelevant noise interference hidden in the sensor data. Meanwhile, a multi-branch convolutional pathway is designed to enrich the features. Furthermore, a feature fusion module is constructed by defining an adaptive weighted fusion rule to fuze the extracted features. The performance of the MBCAN is verified on gearbox and centrifugal pump under different noise environments. The experimental results show that the proposed MBCAN has more excellent anti-noise ability and reliable diagnostic performance than other existing approaches.Keywords: convolutional neural networkfault diagnosisfeature fusionrotating machinery Disclosure statementNo potential conflict of interest was reported by the author(s).Additional informationFundingThis work was supported in part by the National Natural Science Foundation of China under Grant No. 51875225, in part by the Ministry of Industry and Information Technology of China under Grant No. TC210804R-1, and in part by Hubei Provincial Natural Science Foundation for Innovation Groups under Grant No. 2021CFA026.Notes on contributorsKe WuKe Wu received his B.S. and M.S. degrees in mechanical engineering from Hunan University of Science and Technology, in 2017 and 2020, respectively. He is currently a Ph.D candidate at marine engineering with the School of Naval Architecture and Ocean Engineering from Huazhong University of Science and Technology, China. His main research interests include big data analytics, health monitoring and fault diagnosis for equipment.Zirui LiZirui Li received the B.S. degree in marine engineering from the Huazhong University of Science and Technology (HUST), China, in 2020, where he is currently pursuing the master’s degree with the School of Naval Architecture and Ocean Engineering at HUST. His main research interests include health monitoring for equipment, deep learning and reinforcement learning.Chong ChenChong Chen received his B.S. degree in Nuclear science and technology from Harbin Engineering University, China, in 2011, and received his M.S. degrees in Nuclear science and technology from Harbin Engineering University, in 2013. His Ph.D degree in Nuclear science and technology with the school of Fundamental Science on Nuclear Safety and Simulation Technology Laboratory from Harbin Engineering University. His main research interests include reactor thermal-hydraulic and big data analytics.Zhenguo SongZhenguo Song received his B.S. and M.S. degree in marine engineering from Dalian Maritime University, China. He is currently a senior engineer at China Ship Development and Design Center. His main research interests include Integrated design of ship dynamics and big data analytics.Jun WuJun Wu received his B.S. degree in mechanical engineering from Hubei University of Technology, China, in 2001, and received his M.S. and Ph.D. degrees in mechanical engineering from Huazhong University of Science and Technology (HUST), in 2004 and 2008, respectively. He is currently a full Professor of School of Naval Architecture and Ocean Engineering at HUST. He worked as a visiting scholar at Stanford University, CA, USA from 2014 to 2015, and 2019, where he conducted technical research in the area of structure health monitoring. His research interests include equipment health monitoring, fault diagnosis and remaining useful life prediction. He has more than 80 publications and the award of 12 patents, and receives several awards for his teaching activities.
期刊介绍:
Quality Engineering aims to promote a rich exchange among the quality engineering community by publishing papers that describe new engineering methods ready for immediate industrial application or examples of techniques uniquely employed.
You are invited to submit manuscripts and application experiences that explore:
Experimental engineering design and analysis
Measurement system analysis in engineering
Engineering process modelling
Product and process optimization in engineering
Quality control and process monitoring in engineering
Engineering regression
Reliability in engineering
Response surface methodology in engineering
Robust engineering parameter design
Six Sigma method enhancement in engineering
Statistical engineering
Engineering test and evaluation techniques.