{"paper":{"title":"PF$\\Delta$: A Benchmark Dataset for Power Flow under Load, Generation, and Topology Variations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"The PFΔ benchmark provides 859,800 power flow instances to test solvers and ML methods under load, generation, topology, and contingency variations.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alvaro Carbonero, Ana K. Rivera, Anvita Bhagavathula, Priya Donti","submitted_at":"2025-10-24T22:09:09Z","abstract_excerpt":"Power flow (PF) calculations are the backbone of real-time grid operations, across workflows such as contingency analysis (where repeated PF evaluations assess grid security under outages) and topology optimization (which involves PF-based searches over combinatorially large action spaces). Running these calculations at operational timescales or across large evaluation spaces remains a major computational bottleneck. Additionally, growing uncertainty in power system operations from the integration of renewables and climate-induced extreme weather also calls for tools that can accurately and ef"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"PFΔ contains 859,800 solved power flow instances spanning six bus system sizes, three contingency types (N, N-1, N-2), and close-to-infeasible cases near steady-state voltage stability limits; evaluations of traditional solvers and GNN-based methods highlight key areas where existing approaches struggle.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The synthetic variations in load, generation, and topology together with the chosen contingency scenarios and near-infeasible points are sufficiently representative of real-world power system uncertainties to serve as a useful benchmark for ML methods.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"PFΔ is a benchmark dataset of 859,800 power flow solutions across six bus system sizes with N/N-1/N-2 contingencies and close-to-infeasible cases to evaluate traditional solvers and GNN methods.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"The PFΔ benchmark provides 859,800 power flow instances to test solvers and ML methods under load, generation, topology, and contingency variations.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"c14e304ebc341be58593d268d3ba2f7498c0d013a9b5dc718b71f25077d23b7e"},"source":{"id":"2510.22048","kind":"arxiv","version":4},"verdict":{"id":"11336eb3-5275-4596-96ef-f921f00a3c8d","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-18T04:05:22.957619Z","strongest_claim":"PFΔ contains 859,800 solved power flow instances spanning six bus system sizes, three contingency types (N, N-1, N-2), and close-to-infeasible cases near steady-state voltage stability limits; evaluations of traditional solvers and GNN-based methods highlight key areas where existing approaches struggle.","one_line_summary":"PFΔ is a benchmark dataset of 859,800 power flow solutions across six bus system sizes with N/N-1/N-2 contingencies and close-to-infeasible cases to evaluate traditional solvers and GNN methods.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The synthetic variations in load, generation, and topology together with the chosen contingency scenarios and near-infeasible points are sufficiently representative of real-world power system uncertainties to serve as a useful benchmark for ML methods.","pith_extraction_headline":"The PFΔ benchmark provides 859,800 power flow instances to test solvers and ML methods under load, generation, topology, and contingency variations."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2510.22048/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"34909c6150894f601e8f9b9a93f4a6ca43b5b59b6b0201ca23314bebbf23ded0"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}