Loading…

Toward compound fault diagnosis via EMAGAN and large kernel augmented few-shot learning

Bearings are essential in machinery. Damage to them can cause financial losses and safety risks at industrial sites. Therefore, it is necessary to design an accurate diagnostic model. Although many bearing fault diagnosis methods have been proposed recently, they still cannot meet the requirements o...

Full description

Saved in:
Bibliographic Details
Published in:Frontiers in mechanical engineering 2024-07, Vol.10
Main Authors: Xu, Wenchang, Zhang, Zhexian, Wang, Zhijun, Wang, Tianao, He, Zijian, Dong, Shijie
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites cdi_FETCH-LOGICAL-c238t-3e4fb330d1e35069a8b177b8f286f72c32c4764dbee8294fb40268bb816a7d890
container_end_page
container_issue
container_start_page
container_title Frontiers in mechanical engineering
container_volume 10
creator Xu, Wenchang
Zhang, Zhexian
Wang, Zhijun
Wang, Tianao
He, Zijian
Dong, Shijie
description Bearings are essential in machinery. Damage to them can cause financial losses and safety risks at industrial sites. Therefore, it is necessary to design an accurate diagnostic model. Although many bearing fault diagnosis methods have been proposed recently, they still cannot meet the requirements of high-accurate prediction of bearing faults. There are several challenges in this: 1) In practical settings, gathering sufficient and balanced sample data for training diagnostic network models proves challenging. 2) The damage to bearings in real industrial production sites is not singular, and compound faults are also a huge challenge for diagnostic networks. To address these issues, this study introduces a novel fault diagnosis model called EMALKNet that integrates DCGAN with Efficient Multi-Scale Attention (EMAGAN) and RepLKNet-XL, enhancing the detection and analysis of bearing faults in industrial machinery. This model employs EMAGAN to explore the underlying distribution of raw data, thereby enlarging the fault sample pool and enhancing the model’s diagnostic capabilities; The large kernel structure of RepLKNet-XL is different from the current mainstream small kernel and has stronger representation extraction ability. The proposed method has been validated on the Paderborn University dataset and the Huazhong University of Science and Technology dataset.
doi_str_mv 10.3389/fmech.2024.1430542
format article
fullrecord <record><control><sourceid>doaj_cross</sourceid><recordid>TN_cdi_doaj_primary_oai_doaj_org_article_d8fec15a9d7c4bc28d77d898db13fa86</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><doaj_id>oai_doaj_org_article_d8fec15a9d7c4bc28d77d898db13fa86</doaj_id><sourcerecordid>oai_doaj_org_article_d8fec15a9d7c4bc28d77d898db13fa86</sourcerecordid><originalsourceid>FETCH-LOGICAL-c238t-3e4fb330d1e35069a8b177b8f286f72c32c4764dbee8294fb40268bb816a7d890</originalsourceid><addsrcrecordid>eNpNkMtOwzAQRS0EElXpD7DyD6T4kcbOsqpKqVRgU8TSGr_SlDSu7JSKvyd9CLGaq5m5Z3EQeqRkzLksn_zOmc2YEZaPac7JJGc3aMBYKTJORHn7L9-jUUpbQgiVQpaSDtDnOhwhWmzCbh8OrcUeDk2HbQ1VG1Kd8HcNeP46XUzfMPTnBmLl8JeLrWswHKqdazvXt9wxS5vQ4cZBbOu2ekB3HprkRtc5RB_P8_XsJVu9L5az6SozjMsu4y73mnNiqeMTUpQgNRVCS89k4QUznJlcFLnVzklW9r85YYXUWtIChJUlGaLlhWsDbNU-1juIPypArc6LECsFsatN45SV3hk6gdIKk2vDpBUnhLSacg-y6FnswjIxpBSd_-NRok6m1dm0OplWV9PkFxCrco0</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Toward compound fault diagnosis via EMAGAN and large kernel augmented few-shot learning</title><source>DOAJ Directory of Open Access Journals</source><creator>Xu, Wenchang ; Zhang, Zhexian ; Wang, Zhijun ; Wang, Tianao ; He, Zijian ; Dong, Shijie</creator><creatorcontrib>Xu, Wenchang ; Zhang, Zhexian ; Wang, Zhijun ; Wang, Tianao ; He, Zijian ; Dong, Shijie</creatorcontrib><description>Bearings are essential in machinery. Damage to them can cause financial losses and safety risks at industrial sites. Therefore, it is necessary to design an accurate diagnostic model. Although many bearing fault diagnosis methods have been proposed recently, they still cannot meet the requirements of high-accurate prediction of bearing faults. There are several challenges in this: 1) In practical settings, gathering sufficient and balanced sample data for training diagnostic network models proves challenging. 2) The damage to bearings in real industrial production sites is not singular, and compound faults are also a huge challenge for diagnostic networks. To address these issues, this study introduces a novel fault diagnosis model called EMALKNet that integrates DCGAN with Efficient Multi-Scale Attention (EMAGAN) and RepLKNet-XL, enhancing the detection and analysis of bearing faults in industrial machinery. This model employs EMAGAN to explore the underlying distribution of raw data, thereby enlarging the fault sample pool and enhancing the model’s diagnostic capabilities; The large kernel structure of RepLKNet-XL is different from the current mainstream small kernel and has stronger representation extraction ability. The proposed method has been validated on the Paderborn University dataset and the Huazhong University of Science and Technology dataset.</description><identifier>ISSN: 2297-3079</identifier><identifier>EISSN: 2297-3079</identifier><identifier>DOI: 10.3389/fmech.2024.1430542</identifier><language>eng</language><publisher>Frontiers Media S.A</publisher><subject>artificial intelligence ; compound fault ; fault diagnosis ; feature engineering ; few-shot</subject><ispartof>Frontiers in mechanical engineering, 2024-07, Vol.10</ispartof><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c238t-3e4fb330d1e35069a8b177b8f286f72c32c4764dbee8294fb40268bb816a7d890</cites></display><links><openurl>$$Topenurl_article</openurl><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>780</link.rule.ids></links><search><creatorcontrib>Xu, Wenchang</creatorcontrib><creatorcontrib>Zhang, Zhexian</creatorcontrib><creatorcontrib>Wang, Zhijun</creatorcontrib><creatorcontrib>Wang, Tianao</creatorcontrib><creatorcontrib>He, Zijian</creatorcontrib><creatorcontrib>Dong, Shijie</creatorcontrib><title>Toward compound fault diagnosis via EMAGAN and large kernel augmented few-shot learning</title><title>Frontiers in mechanical engineering</title><description>Bearings are essential in machinery. Damage to them can cause financial losses and safety risks at industrial sites. Therefore, it is necessary to design an accurate diagnostic model. Although many bearing fault diagnosis methods have been proposed recently, they still cannot meet the requirements of high-accurate prediction of bearing faults. There are several challenges in this: 1) In practical settings, gathering sufficient and balanced sample data for training diagnostic network models proves challenging. 2) The damage to bearings in real industrial production sites is not singular, and compound faults are also a huge challenge for diagnostic networks. To address these issues, this study introduces a novel fault diagnosis model called EMALKNet that integrates DCGAN with Efficient Multi-Scale Attention (EMAGAN) and RepLKNet-XL, enhancing the detection and analysis of bearing faults in industrial machinery. This model employs EMAGAN to explore the underlying distribution of raw data, thereby enlarging the fault sample pool and enhancing the model’s diagnostic capabilities; The large kernel structure of RepLKNet-XL is different from the current mainstream small kernel and has stronger representation extraction ability. The proposed method has been validated on the Paderborn University dataset and the Huazhong University of Science and Technology dataset.</description><subject>artificial intelligence</subject><subject>compound fault</subject><subject>fault diagnosis</subject><subject>feature engineering</subject><subject>few-shot</subject><issn>2297-3079</issn><issn>2297-3079</issn><fulltext>false</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>DOA</sourceid><recordid>eNpNkMtOwzAQRS0EElXpD7DyD6T4kcbOsqpKqVRgU8TSGr_SlDSu7JSKvyd9CLGaq5m5Z3EQeqRkzLksn_zOmc2YEZaPac7JJGc3aMBYKTJORHn7L9-jUUpbQgiVQpaSDtDnOhwhWmzCbh8OrcUeDk2HbQ1VG1Kd8HcNeP46XUzfMPTnBmLl8JeLrWswHKqdazvXt9wxS5vQ4cZBbOu2ekB3HprkRtc5RB_P8_XsJVu9L5az6SozjMsu4y73mnNiqeMTUpQgNRVCS89k4QUznJlcFLnVzklW9r85YYXUWtIChJUlGaLlhWsDbNU-1juIPypArc6LECsFsatN45SV3hk6gdIKk2vDpBUnhLSacg-y6FnswjIxpBSd_-NRok6m1dm0OplWV9PkFxCrco0</recordid><startdate>20240730</startdate><enddate>20240730</enddate><creator>Xu, Wenchang</creator><creator>Zhang, Zhexian</creator><creator>Wang, Zhijun</creator><creator>Wang, Tianao</creator><creator>He, Zijian</creator><creator>Dong, Shijie</creator><general>Frontiers Media S.A</general><scope>AAYXX</scope><scope>CITATION</scope><scope>DOA</scope></search><sort><creationdate>20240730</creationdate><title>Toward compound fault diagnosis via EMAGAN and large kernel augmented few-shot learning</title><author>Xu, Wenchang ; Zhang, Zhexian ; Wang, Zhijun ; Wang, Tianao ; He, Zijian ; Dong, Shijie</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c238t-3e4fb330d1e35069a8b177b8f286f72c32c4764dbee8294fb40268bb816a7d890</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>artificial intelligence</topic><topic>compound fault</topic><topic>fault diagnosis</topic><topic>feature engineering</topic><topic>few-shot</topic><toplevel>peer_reviewed</toplevel><creatorcontrib>Xu, Wenchang</creatorcontrib><creatorcontrib>Zhang, Zhexian</creatorcontrib><creatorcontrib>Wang, Zhijun</creatorcontrib><creatorcontrib>Wang, Tianao</creatorcontrib><creatorcontrib>He, Zijian</creatorcontrib><creatorcontrib>Dong, Shijie</creatorcontrib><collection>CrossRef</collection><collection>DOAJ Directory of Open Access Journals</collection><jtitle>Frontiers in mechanical engineering</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>no_fulltext</fulltext></delivery><addata><au>Xu, Wenchang</au><au>Zhang, Zhexian</au><au>Wang, Zhijun</au><au>Wang, Tianao</au><au>He, Zijian</au><au>Dong, Shijie</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Toward compound fault diagnosis via EMAGAN and large kernel augmented few-shot learning</atitle><jtitle>Frontiers in mechanical engineering</jtitle><date>2024-07-30</date><risdate>2024</risdate><volume>10</volume><issn>2297-3079</issn><eissn>2297-3079</eissn><abstract>Bearings are essential in machinery. Damage to them can cause financial losses and safety risks at industrial sites. Therefore, it is necessary to design an accurate diagnostic model. Although many bearing fault diagnosis methods have been proposed recently, they still cannot meet the requirements of high-accurate prediction of bearing faults. There are several challenges in this: 1) In practical settings, gathering sufficient and balanced sample data for training diagnostic network models proves challenging. 2) The damage to bearings in real industrial production sites is not singular, and compound faults are also a huge challenge for diagnostic networks. To address these issues, this study introduces a novel fault diagnosis model called EMALKNet that integrates DCGAN with Efficient Multi-Scale Attention (EMAGAN) and RepLKNet-XL, enhancing the detection and analysis of bearing faults in industrial machinery. This model employs EMAGAN to explore the underlying distribution of raw data, thereby enlarging the fault sample pool and enhancing the model’s diagnostic capabilities; The large kernel structure of RepLKNet-XL is different from the current mainstream small kernel and has stronger representation extraction ability. The proposed method has been validated on the Paderborn University dataset and the Huazhong University of Science and Technology dataset.</abstract><pub>Frontiers Media S.A</pub><doi>10.3389/fmech.2024.1430542</doi><oa>free_for_read</oa></addata></record>
fulltext no_fulltext
identifier ISSN: 2297-3079
ispartof Frontiers in mechanical engineering, 2024-07, Vol.10
issn 2297-3079
2297-3079
language eng
recordid cdi_doaj_primary_oai_doaj_org_article_d8fec15a9d7c4bc28d77d898db13fa86
source DOAJ Directory of Open Access Journals
subjects artificial intelligence
compound fault
fault diagnosis
feature engineering
few-shot
title Toward compound fault diagnosis via EMAGAN and large kernel augmented few-shot learning
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-03T09%3A20%3A52IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-doaj_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Toward%20compound%20fault%20diagnosis%20via%20EMAGAN%20and%20large%20kernel%20augmented%20few-shot%20learning&rft.jtitle=Frontiers%20in%20mechanical%20engineering&rft.au=Xu,%20Wenchang&rft.date=2024-07-30&rft.volume=10&rft.issn=2297-3079&rft.eissn=2297-3079&rft_id=info:doi/10.3389/fmech.2024.1430542&rft_dat=%3Cdoaj_cross%3Eoai_doaj_org_article_d8fec15a9d7c4bc28d77d898db13fa86%3C/doaj_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c238t-3e4fb330d1e35069a8b177b8f286f72c32c4764dbee8294fb40268bb816a7d890%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true