ALMformer: a modified Transformer based on Adaptive frequency enhanced attention, large kernel convolution, and multi-scale implementation for bearing fault diagnosis

ALMformer: a modified Transformer based on Adaptive frequency enhanced attention, large kernel convolution, and multi-scale implementation for bearing fault diagnosis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article ALMformer: a modified Transformer based on Adaptive frequency enhanced attention, large kernel convolution, and multi-scale implementation for bearing fault diagnosis Xiao Chang, Shaobin Cai, Wanchen Cai, Yuchang Mo, Liansuo Wei This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7080172/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 25 Sep, 2025 Read the published version in Scientific Reports → Version 1 posted 11 You are reading this latest preprint version Abstract Bearing fault diagnosis has attracted increasing attention due to its critical role in monitoring the health of rotating machinery. Data-driven models based on deep learning (DL) have demonstrated powerful capabilities in feature extraction. However, their performance often degrades under strong noise interference, limiting their applicability in real-world industrial scenarios. To address this issue, this paper proposes a novel attention-enhanced Transformer model that integrates large-kernel convolution and multi-scale CNN structures for robust fault diagnosis. The proposed framework effectively combines spatial-temporal feature modeling with adaptive frequency-domain enhancement, enabling it to suppress noise and emphasize informative diagnostic features. Experimental results on the Paderborn University and Case Western Reserve University datasets show that the proposed method achieves superior recognition accuracy under various signal-to-noise ratios, outperforming several state-of-the-art models. Furthermore, ablation studies and visualization analyses validate the effectiveness and rationality of the proposed architecture. Physical sciences/Engineering Physical sciences/Mathematics and computing Rolling bearing fault diagnosis Multi-scale convolution Adaptive Frequency-Domain Attention Noise robustness Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 25 Sep, 2025 Read the published version in Scientific Reports → Version 1 posted Reviews received at journal 20 Jul, 2025 Reviews received at journal 19 Jul, 2025 Reviewers agreed at journal 17 Jul, 2025 Reviewers agreed at journal 17 Jul, 2025 Reviewers agreed at journal 17 Jul, 2025 Reviewers agreed at journal 16 Jul, 2025 Reviewers invited by journal 16 Jul, 2025 Editor assigned by journal 16 Jul, 2025 Editor invited by journal 16 Jul, 2025 Submission checks completed at journal 14 Jul, 2025 First submitted to journal 14 Jul, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7080172","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":488079889,"identity":"d5e29bf4-fc08-4d23-8e11-f3838d87be1c","order_by":0,"name":"Xiao Chang","email":"","orcid":"","institution":"Huzhou University","correspondingAuthor":false,"prefix":"","firstName":"Xiao","middleName":"","lastName":"Chang","suffix":""},{"id":488079890,"identity":"35c19ad2-aad7-4168-b08e-ff4b3297b0a3","order_by":1,"name":"Shaobin Cai","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYFACxgZmIJnAxsB8gEGyASSSQLQWtgRitTAwMEOU8RgAtROhhb/9cPPnghqbPD7pns8fLHccZuBnzzFg+LkDtxaJM4lt0jOOpRWzyZzdYCB55jCDZM8bA8beM7i1GDAktjHzsB1ObJPI3ZAg2XaYweBGjgEzYxseLfwPmz/z/ANpyXlwAKTFnqAWicQGad42sBbGBrAtEgS0SNx42CbN2wf0i0SaMYNkWzqPxJlnBQd78Wjh709//Jnnm02e/Izkx58l26zl+NuTNz74iUcLCmCWAMYOiHGASA3AhPCBaKWjYBSMglEwkgAA+SROnikyp/IAAAAASUVORK5CYII=","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":true,"prefix":"","firstName":"Shaobin","middleName":"","lastName":"Cai","suffix":""},{"id":488079891,"identity":"0dc7734e-db67-44b2-918c-6def891046d1","order_by":2,"name":"Wanchen Cai","email":"","orcid":"","institution":"National Taiwan University","correspondingAuthor":false,"prefix":"","firstName":"Wanchen","middleName":"","lastName":"Cai","suffix":""},{"id":488079892,"identity":"ecc6aaff-b007-4061-b832-9e4583968b4f","order_by":3,"name":"Yuchang Mo","email":"","orcid":"","institution":"Huaqiao University","correspondingAuthor":false,"prefix":"","firstName":"Yuchang","middleName":"","lastName":"Mo","suffix":""},{"id":488079893,"identity":"3e97a654-a99d-416b-8365-b25c0b69bc79","order_by":4,"name":"Liansuo Wei","email":"","orcid":"","institution":"Suqian University","correspondingAuthor":false,"prefix":"","firstName":"Liansuo","middleName":"","lastName":"Wei","suffix":""}],"badges":[],"createdAt":"2025-07-09 05:53:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7080172/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7080172/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-18187-4","type":"published","date":"2025-09-25T15:58:07+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":92430573,"identity":"8d5ff22a-f087-4892-bb83-b348cd87bf32","added_by":"auto","created_at":"2025-09-29 16:06:01","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1454936,"visible":true,"origin":"","legend":"","description":"","filename":"file.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7080172/v1_covered_e82031e2-804c-4a4c-b042-18d541cdd5df.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"ALMformer: a modified Transformer based on Adaptive frequency enhanced attention, large kernel convolution, and multi-scale implementation for bearing fault diagnosis","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Rolling bearing fault diagnosis, Multi-scale convolution, Adaptive Frequency-Domain Attention, Noise robustness","lastPublishedDoi":"10.21203/rs.3.rs-7080172/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7080172/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBearing fault diagnosis has attracted increasing attention due to its critical role in monitoring the health of rotating machinery. Data-driven models based on deep learning (DL) have demonstrated powerful capabilities in feature extraction. However, their performance often degrades under strong noise interference, limiting their applicability in real-world industrial scenarios. To address this issue, this paper proposes a novel attention-enhanced Transformer model that integrates large-kernel convolution and multi-scale CNN structures for robust fault diagnosis. The proposed framework effectively combines spatial-temporal feature modeling with adaptive frequency-domain enhancement, enabling it to suppress noise and emphasize informative diagnostic features. Experimental results on the Paderborn University and Case Western Reserve University datasets show that the proposed method achieves superior recognition accuracy under various signal-to-noise ratios, outperforming several state-of-the-art models. Furthermore, ablation studies and visualization analyses validate the effectiveness and rationality of the proposed architecture.\u003c/p\u003e","manuscriptTitle":"ALMformer: a modified Transformer based on Adaptive frequency enhanced attention, large kernel convolution, and multi-scale implementation for bearing fault diagnosis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-21 03:23:58","doi":"10.21203/rs.3.rs-7080172/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2025-07-20T14:18:58+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-20T03:13:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"266947320849889974352782916512579924662","date":"2025-07-18T01:46:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"11225977417049774508106300790467200794","date":"2025-07-18T01:46:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"41771197624258084499699276580049231546","date":"2025-07-17T04:11:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"140596146000970232321958605386731089334","date":"2025-07-17T02:00:07+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-07-17T01:58:48+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-17T01:56:44+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-07-16T11:25:34+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-14T14:48:35+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-07-14T14:44:05+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"97968785-6b40-4c02-8002-a6d4889c815c","owner":[],"postedDate":"July 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":51810420,"name":"Physical sciences/Engineering"},{"id":51810421,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2025-09-29T16:02:00+00:00","versionOfRecord":{"articleIdentity":"rs-7080172","link":"https://doi.org/10.1038/s41598-025-18187-4","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-09-25 15:58:07","publishedOnDateReadable":"September 25th, 2025"},"versionCreatedAt":"2025-07-21 03:23:58","video":"","vorDoi":"10.1038/s41598-025-18187-4","vorDoiUrl":"https://doi.org/10.1038/s41598-025-18187-4","workflowStages":[]},"version":"v1","identity":"rs-7080172","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7080172","identity":"rs-7080172","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}