AdditiveGDL: Generative deep learning for predicting local thermal distributions in metal 3D-printed layers

AdditiveGDL: Generative deep learning for predicting local thermal distributions in metal 3D-printed layers | 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 Research Article AdditiveGDL: Generative deep learning for predicting local thermal distributions in metal 3D-printed layers David Guirguis, Conrad Tucker, Jack Beuth This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5257658/v3 This work is licensed under a CC BY 4.0 License Status: Posted Version 3 posted You are reading this latest preprint version Show more versions Abstract In metal additive manufacturing, comprehending the intricate thermal dynamics during the printing process is paramount. These dynamics have far-reaching effects on various critical aspects including microstructure and mechanical properties, fatigue life, residual stresses, dimensional accuracy, shape integrity, and surface quality. Additionally, the scan strategy significantly impacts heat accumulation, especially at geometric features with sharp corners, overhangs, and thin walls. Thus, predicting heat distribution given the scan strategy becomes crucial. While numerical simulations using finite element methods are common, they can be computationally prohibitive for complex parts. In this paper, we propose a method that leverages constrained generative neural networks to predict the local thermal distribution given the laser toolpath. By identifying critical regions of heat accumulation, we can optimize geometry, and scan paths, ultimately enhancing the quality and reliability of 3D-printed metal components. Results show that generative deep learning offers an alternative approach to predicting the thermal field of printed layers efficiently. Materials Engineering Additive Manufacturing Deep Learning Generative AI 3D Printing Anomaly Detection Local Heat Accumulation Generative Adversarial NetworkThermal Simulation IR Thermal Imaging Generative Adversarial Network Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 3 posted You are reading this latest preprint version Show more versions 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-5257658","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":374907268,"identity":"8dd60a56-2fd1-46b5-9562-5f1f18afa45b","order_by":0,"name":"David Guirguis","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAx0lEQVRIiWNgGAWjYDCCA2DEwMPA3gDmMzYQr4XnAAlaIEAigUgtfOfXGB6u+HVYRn7m46ebeRhsZDccIKBF8sYbg4Nn+w7zGNxOM7vNw5BmTFCLwY0zBgcbe4BapHPYgFoOJxKvRX7mGZCW/0RoOd9jcLDhx2Eehhs8IC0HCGuRvMFWcLCxIZ3H4Eya2c05BsnGMwlp4Tt/ePPHhj/W9vLth5/deFNhJ9tHSAs4Ohjb4O4kpBwE+EGG/iFG5SgYBaNgFIxYAADwb02cA/NGhgAAAABJRU5ErkJggg==","orcid":"","institution":"Carnegie Mellon University","correspondingAuthor":true,"prefix":"","firstName":"David","middleName":"","lastName":"Guirguis","suffix":""},{"id":374907269,"identity":"f7061cbc-e14d-4b86-83a1-1222024fbe49","order_by":1,"name":"Conrad Tucker","email":"","orcid":"","institution":"Carnegie Mellon University","correspondingAuthor":false,"prefix":"","firstName":"Conrad","middleName":"","lastName":"Tucker","suffix":""},{"id":374907270,"identity":"9adc0015-d928-4241-a227-84f7e16bf813","order_by":2,"name":"Jack Beuth","email":"","orcid":"","institution":"Carnegie Mellon University","correspondingAuthor":false,"prefix":"","firstName":"Jack","middleName":"","lastName":"Beuth","suffix":""}],"badges":[],"createdAt":"2024-10-14 03:19:49","currentVersionCode":3,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-5257658/v3","doiUrl":"https://doi.org/10.21203/rs.3.rs-5257658/v3","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101297872,"identity":"d2554ac8-14d3-41c4-9bcd-d80e048f1377","added_by":"auto","created_at":"2026-01-28 09:29:09","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1373365,"visible":true,"origin":"","legend":"","description":"","filename":"AdditiveGDLR3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5257658/v3_covered_2d86aabe-be0d-4b4e-8015-3d15730ccac0.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"AdditiveGDL: Generative deep learning for predicting local thermal distributions in metal 3D-printed layers","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Carnegie Mellon University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Additive Manufacturing, Deep Learning, Generative AI, 3D Printing, Anomaly Detection, Local Heat Accumulation, Generative Adversarial NetworkThermal Simulation, IR Thermal Imaging, Generative Adversarial Network","lastPublishedDoi":"10.21203/rs.3.rs-5257658/v3","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5257658/v3","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn metal additive manufacturing, comprehending the intricate thermal dynamics during the printing process is paramount. These dynamics have far-reaching effects on various critical aspects including microstructure and mechanical properties, fatigue life, residual stresses, dimensional accuracy, shape integrity, and surface quality. \u0026nbsp;Additionally, the scan strategy \u0026nbsp;significantly impacts heat accumulation, especially at geometric features with sharp corners, overhangs, and thin walls. \u0026nbsp;Thus, predicting heat distribution given the scan strategy becomes crucial. While numerical simulations using finite element methods are common, they can be computationally prohibitive for complex parts. \u0026nbsp;In this paper, we propose a \u0026nbsp;method that leverages constrained generative neural networks to predict the \u003cem\u003elocal\u003c/em\u003e thermal distribution given the laser toolpath. By identifying critical regions of heat accumulation, we can optimize geometry, and scan paths, ultimately enhancing the quality and reliability of 3D-printed metal components. \u0026nbsp;Results \u0026nbsp;show that \u0026nbsp;generative deep learning offers \u0026nbsp;an alternative approach to predicting the thermal field of printed layers efficiently.\u003c/p\u003e","manuscriptTitle":"AdditiveGDL: Generative deep learning for predicting local thermal distributions in metal 3D-printed layers","msid":"","msnumber":"","nonDraftVersions":[{"code":3,"date":"2026-01-27 20:19:13","doi":"10.21203/rs.3.rs-5257658/v3","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}},{"code":2,"date":"2024-11-06 14:52:55","doi":"10.21203/rs.3.rs-5257658/v2","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}},{"code":1,"date":"2024-10-16 12:44:53","doi":"10.21203/rs.3.rs-5257658/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7c51e99c-4e40-4e30-ae07-5cb01be14d92","owner":[],"postedDate":"January 27th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":39907483,"name":"Materials Engineering"}],"tags":[],"updatedAt":"2024-10-16T12:44:53+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-27 20:19:13","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v3","identity":"rs-5257658","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5257658","identity":"rs-5257658","version":["v3"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}