{"paper":{"title":"Load constrained wind farm flow control through multi-objective multi-agent reinforcement learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Multi-agent reinforcement learning lets wind farm turbines steer wakes for higher total power while keeping load increases below set thresholds.","cross_cats":["cs.LG","cs.SY"],"primary_cat":"eess.SY","authors_text":"Iasonas Tsaklis, Marcus Binder Nilsen, Nikolay Dimitrov, Pierre-Elouan R\\'ethor\\'e, Teodor {\\AA}strand, Tuhfe G\\\"o\\c{c}men","submitted_at":"2026-04-13T12:39:26Z","abstract_excerpt":"This study presents a multi-agent reinforcement learning (MARL) framework for load-constrained wind farm flow control (WFFC). While wake steering can enhance total wind farm power, it often introduces increased structural loads on downstream turbines. To address this, we integrate an Independent Soft Actor-Critic (I-SAC) architecture with a data-driven, local inflow sector-averaged surrogate model to provide real-time estimates of Damage Equivalent Loads (DELs). By incorporating these estimates into a shaped reward function, turbine-specific agents are trained to maximize power production whil"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The MARL agents successfully learn collaborative policies that prioritise power gain while actively retreating from high-DEL control strategies.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The data-driven, local inflow sector-averaged surrogate model supplies sufficiently accurate real-time estimates of Damage Equivalent Loads that can be inserted directly into the shaped reward without introducing large policy errors.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A multi-agent RL system using Independent Soft Actor-Critic and a local-inflow surrogate for damage-equivalent loads learns policies that raise wind-farm power while respecting explicit load-increase limits.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Multi-agent reinforcement learning lets wind farm turbines steer wakes for higher total power while keeping load increases below set thresholds.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"ebf50d4adb154c0d05c3cf26e7ffd74b8827876ee27de8b674309b48f29693a1"},"source":{"id":"2604.22795","kind":"arxiv","version":2},"verdict":{"id":"dd0f3b85-7e2d-4d99-b420-8d93fe878dde","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T15:46:23.314141Z","strongest_claim":"The MARL agents successfully learn collaborative policies that prioritise power gain while actively retreating from high-DEL control strategies.","one_line_summary":"A multi-agent RL system using Independent Soft Actor-Critic and a local-inflow surrogate for damage-equivalent loads learns policies that raise wind-farm power while respecting explicit load-increase limits.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The data-driven, local inflow sector-averaged surrogate model supplies sufficiently accurate real-time estimates of Damage Equivalent Loads that can be inserted directly into the shaped reward without introducing large policy errors.","pith_extraction_headline":"Multi-agent reinforcement learning lets wind farm turbines steer wakes for higher total power while keeping load increases below set thresholds."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.22795/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":15,"sample":[{"doi":"","year":1999,"title":"Thomsen K and Sørensen P 1999 Fatigue loads for wind turbines operating in wakesJournal of Wind Engineering and Industrial Aerodynamics80121–136 ISSN 0167-6105","work_id":"1fe1d1d6-29b4-4477-9e66-b448cd79ee6b","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Debusscher C M J, G¨ o¸ cmen T and Andersen S J 2022 Probabilistic surrogates for flow control using combined control strategiesJournal of Physics: Conference Series2265032110","work_id":"130f4bcd-6f09-43cb-9378-da01af7a2dd4","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Padullaparthi V R, Nagarathinam S, Vasan A, Menon V and Sudarsanam D 2022 Falcon- farm level control for wind turbines using multi-agent deep reinforcement learningRenewable Energy181445–456 ISSN 0960","work_id":"b6d99ab4-a896-43c3-a77b-4d9f6391e535","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"Damiani R, Dana S, Annoni J, Fleming P, Roadman J, van Dam J and Dykes K 2018 Assessment of wind turbine component loads under yaw-offset conditionsWind Energy Science3173–189 ISSN 2366-7443 publisher","work_id":"24250f56-c859-4e3f-9c05-5bcd8d3847db","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"G¨ o¸ cmen T, Liew J, Kadoche E, Dimitrov N, Riva R, Andersen S J, Lio A W, Quick J, R´ ethor´ e P E and Dykes K 2025 Data-driven wind farm flow control and challenges towards field implementation: A ","work_id":"e1058bd2-fa5f-46bb-9cb6-c8408c0479a8","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":15,"snapshot_sha256":"7929e01c7d3f9f1f372526ef21ae5e6e9e01b0b6cba4afd57f4eb48fe7083073","internal_anchors":1},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}