Neural-Hawkes Graph Dynamics: A Continuous-Time Learning Framework for Fault Contagion and Reliability Assessment in Power Grids

Neural-Hawkes Graph Dynamics: A Continuous-Time Learning Framework for Fault Contagion and Reliability Assessment in Power Grids | 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 Neural-Hawkes Graph Dynamics: A Continuous-Time Learning Framework for Fault Contagion and Reliability Assessment in Power Grids koffi tinin Worou, Agbassou Guenoukpati, Adekunlé Akim Salami This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9464988/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract Modern power systems require diagnostic frameworks capable of capturing both the instantaneous network state and the stochastic dynamics of fault propagation. While graph neural networks (GNNs) provide powerful representations of spatial dependencies, their reliance on discrete-time formulations limits their ability to model the continuous and self-exciting nature of cascading failures. This study introduces a hybrid framework that integrates GATv2-based graph representations with spatiotemporal Hawkes processes. The GNN component serves as an encoder, learning latent features that reflect both grid topology and physical vulnerabilities through electrical impedance, while the Hawkes process acts as a decoder, modeling the temporal evolution of failure intensity. To incorporate physical consistency, a physics-informed excitation kernel is proposed to govern risk propagation in accordance with Kirchhoff’s laws. Evaluated on a digital twin of the IEEE 118-bus system, the approach significantly improves the likelihood fit of cascading event sequences compared to conventional Poisson and statistical models.An electrical branching ratio is further introduced as a reliability indicator, enabling early detection of instability phases preceding large-scale failures, and supporting more proactive and resilient maintenance strategies. Physical sciences/Engineering Physical sciences/Mathematics and computing Physical sciences/Physics Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 12 May, 2026 Reviewers agreed at journal 08 May, 2026 Reviews received at journal 08 May, 2026 Reviewers agreed at journal 07 May, 2026 Reviews received at journal 07 May, 2026 Reviewers agreed at journal 28 Apr, 2026 Reviewers agreed at journal 26 Apr, 2026 Reviewers invited by journal 24 Apr, 2026 Editor invited by journal 24 Apr, 2026 Editor assigned by journal 21 Apr, 2026 Submission checks completed at journal 21 Apr, 2026 First submitted to journal 19 Apr, 2026 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-9464988","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":629846841,"identity":"b95c84ee-924c-42f8-8ddb-a47e1ba36ac2","order_by":0,"name":"koffi tinin Worou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA90lEQVRIiWNgGAWjYDADPgY2IFnBwGDAAKFxAh4Ygw2s5QxUyxmitTC2EaHFnr352IOfO2rlgFoSHxfOOyxvzt58gOHgHjy28BxLN+w9c9wYqOWw8cxthw139hxLYDjwDI8WiRwzCd62Y4ltDOxt0rzbDjNuuJFjwPzhAB4t8u+/Sf5tO1YP1NL+m3fOYXuQFoYD+LRI8LBJ87bVJAAddoyZt+FwImEtZ9LMpGXbDhi2MbAlSwM9lrzhzLGEA/i0sLcffib5tq1Onp+BzfAzT4217YbjzQcf4NMCBYcZGOQfgBjNYC5hDQwMdRiMUTAKRsEoGAVwAADmj1KGYiHbNAAAAABJRU5ErkJggg==","orcid":"","institution":"University of Lomé","correspondingAuthor":true,"prefix":"","firstName":"koffi","middleName":"tinin","lastName":"Worou","suffix":""},{"id":629846845,"identity":"5ca14da0-e8fe-4e74-bdf8-27ea0da474aa","order_by":1,"name":"Agbassou Guenoukpati","email":"","orcid":"","institution":"University of Lomé","correspondingAuthor":false,"prefix":"","firstName":"Agbassou","middleName":"","lastName":"Guenoukpati","suffix":""},{"id":629846852,"identity":"c34131c9-4995-4eff-b3af-6fef2a8b6920","order_by":2,"name":"Adekunlé Akim Salami","email":"","orcid":"","institution":"University of Lomé","correspondingAuthor":false,"prefix":"","firstName":"Adekunlé","middleName":"Akim","lastName":"Salami","suffix":""}],"badges":[],"createdAt":"2026-04-19 22:23:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9464988/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9464988/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107919618,"identity":"19a893d6-6551-47ce-a606-120e31556ad0","added_by":"auto","created_at":"2026-04-27 14:39:52","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":350736,"visible":true,"origin":"","legend":"","description":"","filename":"submissionstoScientificReports12.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9464988/v1_covered_d5fba5bd-42aa-4778-9038-b668c748b014.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Neural-Hawkes Graph Dynamics: A Continuous-Time Learning Framework for Fault Contagion and Reliability Assessment in Power Grids","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"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":"","lastPublishedDoi":"10.21203/rs.3.rs-9464988/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9464988/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Modern power systems require diagnostic frameworks capable of capturing both the instantaneous network state and the stochastic dynamics of fault propagation. While graph neural networks (GNNs) provide powerful representations of spatial dependencies, their reliance on discrete-time formulations limits their ability to model the continuous and self-exciting nature of cascading failures. This study introduces a hybrid framework that integrates GATv2-based graph representations with spatiotemporal Hawkes processes. The GNN component serves as an encoder, learning latent features that reflect both grid topology and physical vulnerabilities through electrical impedance, while the Hawkes process acts as a decoder, modeling the temporal evolution of failure intensity. To incorporate physical consistency, a physics-informed excitation kernel is proposed to govern risk propagation in accordance with Kirchhoff’s laws. Evaluated on a digital twin of the IEEE 118-bus system, the approach significantly improves the likelihood fit of cascading event sequences compared to conventional Poisson and statistical models.An electrical branching ratio is further introduced as a reliability indicator, enabling early detection of instability phases preceding large-scale failures, and supporting more proactive and resilient maintenance strategies.","manuscriptTitle":"Neural-Hawkes Graph Dynamics: A Continuous-Time Learning Framework for Fault Contagion and Reliability Assessment in Power Grids","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-27 14:38:16","doi":"10.21203/rs.3.rs-9464988/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"56780498159669000040255833038784830806","date":"2026-05-12T05:00:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"219744696028487486407741288103564783185","date":"2026-05-08T04:42:14+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-08T04:12:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"247693865498361608966341209767003716993","date":"2026-05-08T03:54:50+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-08T03:47:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"151127803450256212057442473054010396152","date":"2026-04-28T14:28:00+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"60384348854707523299386035318031158182","date":"2026-04-26T20:26:56+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-24T09:02:07+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-04-24T08:35:29+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-21T09:52:28+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-21T09:52:23+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-04-19T22:11:50+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":"69e842bd-b7a0-4193-84a6-c75d79a60e43","owner":[],"postedDate":"April 27th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"56780498159669000040255833038784830806","date":"2026-05-12T05:00:25+00:00","index":38,"fulltext":""},{"type":"reviewerAgreed","content":"219744696028487486407741288103564783185","date":"2026-05-08T04:42:14+00:00","index":37,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-08T04:12:46+00:00","index":36,"fulltext":""},{"type":"reviewerAgreed","content":"247693865498361608966341209767003716993","date":"2026-05-08T03:54:50+00:00","index":35,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-08T03:47:34+00:00","index":30,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67035323,"name":"Physical sciences/Engineering"},{"id":67035324,"name":"Physical sciences/Mathematics and computing"},{"id":67035325,"name":"Physical sciences/Physics"}],"tags":[],"updatedAt":"2026-04-27T14:38:17+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-27 14:38:16","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9464988","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9464988","identity":"rs-9464988","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}