arXiv：cs.AI· Delaram Pirhayatifard, Terrence Alsup·· 1 天前
MetaGrid：面向电力价格预测的跨市场元学习
Cross-Market Meta-Learning for Electricity Price Forecasting
AI 导读
MetaGrid 是一种拓扑感知 Transformer，通过跨市场元训练，用单一与区域数量无关的参数集适配未见过的电网。在五个美国 ISO 的无泄漏基准上，MetaGrid 取得最低平均 MAE，比最佳基线提升 3.1%，比同架构从零训练提升 7.7%。该工作已被 NeurIPS 2026 Workshop on AI Foundations for Power Grids 接收。
正文
Abstract:Day-ahead (DA) electricity price forecasting is usually posed per market, with a separate model fitted to each grid's own history. This caps accuracy: statistical and tree-based baselines saturate, while higher-capacity sequence models need more training data than a single market's few years of hourly prices can supply. Such methods also treat each node as an independent series, discarding the spatial dependencies induced by network topology and congestion. Because price formation across independent system operators (ISOs) is driven by shared physics and largely common market design, we instead treat cross-market history as a pretraining corpus. We present MetaGrid, a topology-aware transformer, meta-trained across markets so that a single zone-count-independent parameter set adapts to an unseen grid. On a leak-free benchmark of five U.S. ISOs, where every model receives the same prices, DA load forecasts and calendar features, MetaGrid attains the lowest average MAE, improving on the best baseline by 3.1\% and on the same architecture trained from scratch by 7.7\%.
| Comments: | Accepted at the NeurIPS 2026 Workshop on AI Foundations for Power Grids |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.17387 [cs.LG] |
| (or arXiv:2610.17387v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.17387 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Delaram Pirhayatifard [view email]
[v1]
Mon, 12 Oct 2026 20:48:53 UTC (64 KB)
来源：arXiv：cs.AI · arxiv.org