arXiv cs.MA· Yuhang Dong, Min Xia, Jinghong Li, Yue Dai, Xinyue Wang, Shi Liang·· 12 小时前AI 评分25
Guarded Agent-in-the-Loop 残差自适应框架提升风电功率预测精度
Guarded Agent-in-the-Loop Residual Adaptation for Evolving Wind Power Regimes
AI 导读
研究提出 Guarded Agent-in-the-Loop 残差自适应框架，保持时序主干不变，仅通过 Agent 层识别气象场景并调度小型残差专家做修正。在 9 种时序主干、5 个预测步长（4/8/12/24/48）共 45 组组合上，MAE 与 RMSE 全部下降，平均 MAE 降幅 11.21% 至 20.26%。
正文
Abstract:Wind power forecasting models are often deployed as fixed temporal backbones, but the errors they make can vary across meteorological regimes such as cold waves, heat-and-calm periods, productive typhoon, and destructive storms. We present a guarded Agent-in-the-loop residual adaptation framework that leaves the temporal backbone unchanged and adapts only a small set of scene-conditioned residual experts. An Agent layer interprets external scene information, identifies the current regime, selects residual experts, and applies a correction only when recent observed errors support it. New experts enter the pool after they outperform their initializations on a validation set, keeping model growth under explicit control. Backbone-only and Agent-enhanced forecasts are compared at lead times of 4, 8, 12, 24, and 48 across nine time-series backbones. Across these runs, the guarded framework lowers both MAE and RMSE for all 45 backbone-horizon combinations, with average MAE reductions ranging from 11.21% to 20.26% across backbones. Online-selection baselines, component ablations, and routing analyses identify the roles of residual experts, scene-aware routing, and the online acceptance rule. Candidate-scene runs evaluate the promotion path under controlled conditions; long-term unattended deployment was not evaluated.
| Subjects: | Multiagent Systems (cs.MA) |
| ACM classes: | I.2.7; I.2.6 |
| Cite as: | arXiv:2610.34968 [cs.MA] |
| (or arXiv:2610.34968v1 [cs.MA] for this version) | |
| https://doi.org/10.48550/arXiv.2610.34968 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yuhang Dong [view email]
[v1]
Mon, 26 Oct 2026 07:38:33 UTC (2,446 KB)
来源：arXiv cs.MA · arxiv.org