Causal Sentiment-Driven Reinforcement Learning Framework for Adaptive Solar Energy Policy Design

Causal Sentiment-Driven Reinforcement Learning Framework for Adaptive Solar Energy Policy Design | 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 Causal Sentiment-Driven Reinforcement Learning Framework for Adaptive Solar Energy Policy Design Asif Jabbar, Jingbo Yuan, Salma Idris, Muhammad Inshal, Wenjing Wang, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7086935/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 The global energy transition demands not only technological innovation but also adaptive policy mechanisms that respond to shifting public sentiment and socio economic conditions. This paper presents a novel sentiment aware reinforcement learning framework for solar energy policy optimization that integrates transformer based natural language processing, structural causal modeling, and deep reinforcement learning. First, public sentiment is extracted from diverse unstructured textual sources including social media and global news using RoBERTa, a transformer model fine tuned on energy related discourse. These sentiment scores are then incorporated into a Structural Causal Model (SCM) to estimate their causal impact on solar adoption, adjusting for key confounders such as GDP, electricity pricing, and subsidy levels. The inferred causal insights are used to guide a Proximal Policy Optimization (PPO) reinforcement learning agent that recommends adaptive policy actions including subsidy changes, awareness campaigns, and regulatory shifts aimed at maximizing long term adoption and minimizing economic cost. The framework is evaluated across multiple geographic regions and policy scenarios using real world economic and social data streams. Results demonstrate that sentiment-driven policies significantly outperform static or heuristic baselines in both adoption rate and fiscal efficiency. The inclusion of explainability modules, such as SHAP values and policy entropy metrics, enhances model transparency and stakeholder trust. This end-to-end system not only captures the complex, dynamic interaction between public perception and policy response but also offers a scalable, interpretable, and real-time decision support tool for energy governance. The proposed methodology has broader implications for AI-driven policymaking across climate, mobility, and public health domains. Earth and environmental sciences/Environmental social sciences Physical sciences/Mathematics and computing Sentiment Analysis Reinforcement Learning Structural Causal Models Solar Energy Policy Public Opinion Modeling Explainable AI Full Text Additional Declarations No competing interests reported. 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-7086935","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":489329757,"identity":"71e8080b-4f48-4169-a91d-fca75eb579c9","order_by":0,"name":"Asif Jabbar","email":"","orcid":"","institution":"College of Management, Shenzhen University","correspondingAuthor":false,"prefix":"","firstName":"Asif","middleName":"","lastName":"Jabbar","suffix":""},{"id":489329758,"identity":"caa1f840-ddb0-44ab-a34a-5dd04c554f72","order_by":1,"name":"Jingbo Yuan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA40lEQVRIiWNgGAWjYBACxmYGBmYIk/3ggw8GNnakaOFJNpxRkJZMlE1QLQxmwjwfDjE2EFTeznv4dUHF4cT+2Q1pzDYGB5gZ2A8f3YDfYXxp1jPOHE6ccefgscc5Bnf4GHjS0m7g18JjZszbdjix4UZCunGOwTNmBgkeM+K0zL+RYCZtYXCYsYEILcaPQVo2gLQwEKnFjHnGmXTjjTdykg17DNKS2Qj5xbD/jPHnggpr2Xk30g8++PHHxo6f/fAx/FoaGNgkGBiaESJs+JSDgDwwaj4wMNQRUjcKRsEoGAUjGQAAMapNYdgjfT8AAAAASUVORK5CYII=","orcid":"","institution":"College of Management, Shenzhen University","correspondingAuthor":true,"prefix":"","firstName":"Jingbo","middleName":"","lastName":"Yuan","suffix":""},{"id":489329759,"identity":"f64def57-5c97-4e63-b6f7-6064064d38e6","order_by":2,"name":"Salma Idris","email":"","orcid":"","institution":"CCIS Prince Sultan University","correspondingAuthor":false,"prefix":"","firstName":"Salma","middleName":"","lastName":"Idris","suffix":""},{"id":489329760,"identity":"ce55a5ea-49ca-4ce2-aa6f-b1c752d1feda","order_by":3,"name":"Muhammad Inshal","email":"","orcid":"","institution":"CCIS Prince Sultan University","correspondingAuthor":false,"prefix":"","firstName":"Muhammad","middleName":"","lastName":"Inshal","suffix":""},{"id":489329761,"identity":"ee5f00ef-719c-4365-9689-d5ee70e523e0","order_by":4,"name":"Wenjing Wang","email":"","orcid":"","institution":"College of Management, Shenzhen University","correspondingAuthor":false,"prefix":"","firstName":"Wenjing","middleName":"","lastName":"Wang","suffix":""},{"id":489329762,"identity":"a850355e-b919-4c51-bed7-86e51f0a38e6","order_by":5,"name":"Tariq Mahmood","email":"","orcid":"","institution":"CCIS Prince Sultan University","correspondingAuthor":false,"prefix":"","firstName":"Tariq","middleName":"","lastName":"Mahmood","suffix":""}],"badges":[],"createdAt":"2025-07-09 19:38:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7086935/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7086935/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87581443,"identity":"5d6aef76-ee9a-4659-a021-60632760e993","added_by":"auto","created_at":"2025-07-25 12:54:11","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2405054,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptUpdated1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7086935/v1_covered_fdd01ddb-433f-4e99-bd72-785ca20e58a4.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Causal Sentiment-Driven Reinforcement Learning Framework for Adaptive Solar Energy Policy Design","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"Sentiment Analysis, Reinforcement Learning, Structural Causal Models, Solar Energy Policy, Public Opinion Modeling, Explainable AI","lastPublishedDoi":"10.21203/rs.3.rs-7086935/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7086935/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The global energy transition demands not only technological innovation but also adaptive policy mechanisms that respond to shifting public sentiment and socio economic conditions. This paper presents a novel sentiment aware reinforcement learning framework for solar energy policy optimization that integrates transformer based natural language processing, structural causal modeling, and deep reinforcement learning. First, public sentiment is extracted from diverse unstructured textual sources including social media and global news using RoBERTa, a transformer model fine tuned on energy related discourse. These sentiment scores are then incorporated into a Structural Causal Model (SCM) to estimate their causal impact on solar adoption, adjusting for key confounders such as GDP, electricity pricing, and subsidy levels. The inferred causal insights are used to guide a Proximal Policy Optimization (PPO) reinforcement learning agent that recommends adaptive policy actions including subsidy changes, awareness campaigns, and regulatory shifts aimed at maximizing long term adoption and minimizing economic cost. The framework is evaluated across multiple geographic regions and policy scenarios using real world economic and social data streams. Results demonstrate that sentiment-driven policies significantly outperform static or heuristic baselines in both adoption rate and fiscal efficiency. The inclusion of explainability modules, such as SHAP values and policy entropy metrics, enhances model transparency and stakeholder trust. This end-to-end system not only captures the complex, dynamic interaction between public perception and policy response but also offers a scalable, interpretable, and real-time decision support tool for energy governance. The proposed methodology has broader implications for AI-driven policymaking across climate, mobility, and public health domains.","manuscriptTitle":"Causal Sentiment-Driven Reinforcement Learning Framework for Adaptive Solar Energy Policy Design","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-24 05:31:16","doi":"10.21203/rs.3.rs-7086935/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":"e2520116-395c-4f9f-bc6f-1c4ed5f30ccb","owner":[],"postedDate":"July 24th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":51949926,"name":"Earth and environmental sciences/Environmental social sciences"},{"id":51949927,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2025-07-25T12:53:38+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-24 05:31:16","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7086935","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7086935","identity":"rs-7086935","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}