Digital Twin and Machine Learning Approaches for Renewable Energy System Optimization

Digital Twin and Machine Learning Approaches for Renewable Energy System Optimization | 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 Digital Twin and Machine Learning Approaches for Renewable Energy System Optimization IDOWU OLUGBENGA ADEWUMI, Victoria Bola Oyekunle, Waheed Azeez Ajani This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7390493/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This research introduced a combined digital twin and machine learning framework aimed at enhancing biofuel conversion efficiency and energy return on investment (EROI), while concurrently simulating battery degradation dynamics for sustainable energy systems. Six forecasting models were evaluated using leave-one-out cross-validation (LOOCV). In terms of conversion efficiency, linear regression delivered the highest performance (MAE = 1.22, RMSE = 1.49, R² = 0.84), whereas gradient boosting slightly enhanced predictive consistency in predicted–actual visualizations. In the case of EROI, linear regression consistently surpassed other methods (MAE = 0.21, RMSE = 0.26, R² = 0.89), demonstrating a 20–25% decrease in error relative to tree-based models. Analysis of feature importance showed that fermentation time (+ 0.47) and ethanol yield (+ 0.44) were the key predictors for conversion efficiency and EROI, whereas energy input (–0.56) had the most significant negative impact. Comparisons between actual and predicted outcomes from five experimental batches revealed average deviations of ± 1.1% for efficiency and ± 0.21 for EROI, demonstrating robust model generalization. The optimization phase of the digital twin utilized reinforcement learning alongside AutoML frameworks. Bayesian optimization provided the best results with yield increases of 28.1 L, an 18.2% boost in EROI, and a 12.2% reduction in energy consumption, reaching convergence within just 200 epochs. In contrast, deep Q-learning reached a reduced yield (26.3 L) and demonstrated a slower convergence rate (500 epochs). Complementary battery degradation modeling showed that capacity retention dropped to 82% after 300 cycles, accompanied by thermal variations reaching 45°C during discharge, highlighting the necessity for combined electrochemical-thermodynamic monitoring. In summary, the suggested framework resulted in significant improvements in yield (+ 15–18%), EROI (+ 12–18%), and energy savings (8–12%), while offering a scalable solution for sustainable bioenergy and battery management systems. Artificial Intelligence and Machine Learning Biofuel optimization cassava-to-ethanol conversion conversion efficiency energy return on investment (EROI) battery degradation modeling digital twin reinforcement learning Bayesian optimization AutoML feature importance machine learning LSTM thermal profile capacity retention discharge curves sustainable energy systems. Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted 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-7390493","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":501404883,"identity":"8170640c-22b8-41c8-a0b5-a4c81b7b074a","order_by":0,"name":"IDOWU OLUGBENGA ADEWUMI","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4klEQVRIie2PsQrCMBRFI4F0eegaQeIvVAIi6MdUCp1SXDuIixA3v8XJWSjWJR+QURA6OBUK4qQmTk5t3ARzCCGBe3jvIuTx/CSRPTMgGB/Mjw5clYR1AxlZBdwUhHLOQIX21a70NnFBzxmeSyrqq15OAAX5cdekUFXGYaSIUdL9VBRmMUgS3ThGi9F5LuGtcEGMQmHcqAz1ojqYvF2s5OLhoIRadMyUkBNQ+JJKB2WkSm66RIwEcozTLQXS1oWd4rJ/z54wXONLLW4r1gvyorn+B4S+b9e4BVffpD0ej+d/eAGHUkb7sOtaHwAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-7005-3306","institution":"Department of Computer and Information Engineering, Faculty of Natural and Applied Science, Lead City University, Ibadan, Nigeria","correspondingAuthor":true,"prefix":"","firstName":"IDOWU","middleName":"OLUGBENGA","lastName":"ADEWUMI","suffix":""},{"id":501404884,"identity":"c4382364-f7a8-4c24-834f-675a94560b1c","order_by":1,"name":"Victoria Bola Oyekunle","email":"","orcid":"","institution":"Department of Computer and Information Engineering, Faculty of Natural and Applied Science, Lead City University, Ibadan, Nigeria","correspondingAuthor":false,"prefix":"","firstName":"Victoria","middleName":"Bola","lastName":"Oyekunle","suffix":""},{"id":501404885,"identity":"5081aa7d-eae5-44ef-a44e-2da2038df4b4","order_by":2,"name":"Waheed Azeez Ajani","email":"","orcid":"","institution":"Department of Computer and Information Engineering, Faculty of Natural and Applied Science, Lead City University, Ibadan, Nigeria","correspondingAuthor":false,"prefix":"","firstName":"Waheed","middleName":"Azeez","lastName":"Ajani","suffix":""}],"badges":[],"createdAt":"2025-08-17 05:17:36","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7390493/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7390493/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89419051,"identity":"c05527c4-28bd-45a4-aaa8-39b414f35587","added_by":"auto","created_at":"2025-08-19 18:05:17","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":696994,"visible":true,"origin":"","legend":"","description":"","filename":"DigitalTwinandMachineLearningApproachesforRenewableEnergySystemOptimization1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7390493/v1_covered_379b85f5-bce2-4742-8140-3cf513f2df5b.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eDigital Twin and Machine Learning Approaches for Renewable Energy System Optimization\u003c/strong\u003e\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"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":"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":"Biofuel optimization, cassava-to-ethanol conversion, conversion efficiency, energy return on investment (EROI), battery degradation modeling, digital twin, reinforcement learning, Bayesian optimization, AutoML, feature importance, machine learning, LSTM, thermal profile, capacity retention, discharge curves, sustainable energy systems.","lastPublishedDoi":"10.21203/rs.3.rs-7390493/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7390493/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis research introduced a combined digital twin and machine learning framework aimed at enhancing biofuel conversion efficiency and energy return on investment (EROI), while concurrently simulating battery degradation dynamics for sustainable energy systems. Six forecasting models were evaluated using leave-one-out cross-validation (LOOCV). In terms of conversion efficiency, linear regression delivered the highest performance (MAE\u0026thinsp;=\u0026thinsp;1.22, RMSE\u0026thinsp;=\u0026thinsp;1.49, R\u0026sup2; = 0.84), whereas gradient boosting slightly enhanced predictive consistency in predicted\u0026ndash;actual visualizations. In the case of EROI, linear regression consistently surpassed other methods (MAE\u0026thinsp;=\u0026thinsp;0.21, RMSE\u0026thinsp;=\u0026thinsp;0.26, R\u0026sup2; = 0.89), demonstrating a 20\u0026ndash;25% decrease in error relative to tree-based models. Analysis of feature importance showed that fermentation time (+\u0026thinsp;0.47) and ethanol yield (+\u0026thinsp;0.44) were the key predictors for conversion efficiency and EROI, whereas energy input (\u0026ndash;0.56) had the most significant negative impact. Comparisons between actual and predicted outcomes from five experimental batches revealed average deviations of \u0026plusmn;\u0026thinsp;1.1% for efficiency and \u0026plusmn;\u0026thinsp;0.21 for EROI, demonstrating robust model generalization. The optimization phase of the digital twin utilized reinforcement learning alongside AutoML frameworks. Bayesian optimization provided the best results with yield increases of 28.1 L, an 18.2% boost in EROI, and a 12.2% reduction in energy consumption, reaching convergence within just 200 epochs. In contrast, deep Q-learning reached a reduced yield (26.3 L) and demonstrated a slower convergence rate (500 epochs). Complementary battery degradation modeling showed that capacity retention dropped to 82% after 300 cycles, accompanied by thermal variations reaching 45\u0026deg;C during discharge, highlighting the necessity for combined electrochemical-thermodynamic monitoring. In summary, the suggested framework resulted in significant improvements in yield (+\u0026thinsp;15\u0026ndash;18%), EROI (+\u0026thinsp;12\u0026ndash;18%), and energy savings (8\u0026ndash;12%), while offering a scalable solution for sustainable bioenergy and battery management systems.\u003c/p\u003e","manuscriptTitle":"Digital Twin and Machine Learning Approaches for Renewable Energy System Optimization","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-19 17:33:12","doi":"10.21203/rs.3.rs-7390493/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":"6254ff2e-f4df-4f5c-85fe-7fae86909cd0","owner":[],"postedDate":"August 19th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":53256706,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2026-01-26T06:09:12+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-19 17:33:12","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7390493","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7390493","identity":"rs-7390493","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}