{"paper":{"title":"Privacy-Preserving Generation Fraud Detection for Distributed Photovoltaic Systems: A Solar Irradiance-Fused Federated Learning Framework","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A federated learning framework fuses solar irradiance data via co-attention to detect photovoltaic generation fraud while keeping household data private across communities.","cross_cats":["cs.CE"],"primary_cat":"cs.LG","authors_text":"Chenghao Huang, Hao Wang, Xiaolu Chen, Yanru Zhang","submitted_at":"2026-05-16T15:19:14Z","abstract_excerpt":"The wide adoption of residential photovoltaic (PV) systems introduces new challenges for generation fraud detection (FD). Unlike traditional electricity theft detection, which focuses on electricity consumption-side behavior, PV generation fraud detection (PVG-FD) is complicated by the inherent intermittency and uncertainty of PV generation. The distributed nature of PV systems poses further challenges for centralized PVG-FD approaches due to scalability and privacy concerns. This paper develops a privacy-preserving distributed PVG-FD framework based on federated learning (FL). In this framewo"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Extensive experiments on a real-world residential PV dataset validate the effectiveness of the developed method and demonstrate that it outperforms state-of-the-art FL methods across various scenarios, with scalability across community sizes and strong robustness to class imbalance.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that discrepancies between reported PV generation and fused solar irradiance/weather data reliably indicate fraud, and that prototype alignment sufficiently mitigates class imbalance without distorting normal generation patterns (implicit in the co-attention and FL aggregation description).","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A federated learning framework fuses solar irradiance and PV generation data via co-attention, uses prototype alignment for imbalance, and aggregates models across communities to detect generation fraud while preserving privacy.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A federated learning framework fuses solar irradiance data via co-attention to detect photovoltaic generation fraud while keeping household data private across communities.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"3f6e156252b24b4c26143665f74855f1e9e6d2d72f9dec96f2871dacafa831ff"},"source":{"id":"2605.17039","kind":"arxiv","version":1},"verdict":{"id":"2aab4dc2-268e-48b5-9c4e-f8d2f4753841","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T20:13:04.550723Z","strongest_claim":"Extensive experiments on a real-world residential PV dataset validate the effectiveness of the developed method and demonstrate that it outperforms state-of-the-art FL methods across various scenarios, with scalability across community sizes and strong robustness to class imbalance.","one_line_summary":"A federated learning framework fuses solar irradiance and PV generation data via co-attention, uses prototype alignment for imbalance, and aggregates models across communities to detect generation fraud while preserving privacy.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that discrepancies between reported PV generation and fused solar irradiance/weather data reliably indicate fraud, and that prototype alignment sufficiently mitigates class imbalance without distorting normal generation patterns (implicit in the co-attention and FL aggregation description).","pith_extraction_headline":"A federated learning framework fuses solar irradiance data via co-attention to detect photovoltaic generation fraud while keeping household data private across communities."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.17039/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T20:31:18.921308Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T20:20:42.557293Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"citation_quote_validity","ran_at":"2026-05-19T19:49:41.692432Z","status":"skipped","version":"0.1.0","findings_count":0},{"name":"cited_work_retraction","ran_at":"2026-05-19T18:51:56.308326Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T18:41:56.160143Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:22.996445Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"a4ad3a8b6b656cc061000691205101e382f0cdde273496636d4400bf4a3c6577"},"references":{"count":44,"sample":[{"doi":"","year":2024,"title":"International Energy Agency, “Renewables 2024,” 2023. [Online]. Available: https://www.iea.org/reports/renewables-2024","work_id":"9d20969d-56a8-48d5-992d-6c18d9adb9bb","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Renewable energy market update,","work_id":"d2b1b7c0-e739-4dc4-8d9d-f257cb3054d3","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"A review of cyber security risks of power systems: from static to dynamic false data attacks,","work_id":"fcf1fba8-831a-405b-b6ea-9adf52f8fbf6","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Detection methods in smart meters for electricity thefts: A survey,","work_id":"e263eaeb-f3b5-4b9d-ab94-88beed82073e","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Deep learning detection of electricity theft cyber-attacks in renewable distributed generation,","work_id":"39468152-171b-4eb0-af24-9ec47e012ebc","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":44,"snapshot_sha256":"c376420fcbdd6c7bd7a431d499b9c705271f605f82d0a917c580b999f3c7c778","internal_anchors":0},"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"}