Reproduce “UniWind: Day-Ahead Wind Power Forecasting via Physics-Informed State Routing” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Ronghui Xu, Tongxin Wu, Guozhen Zhang, Yihan Li, Chenjuan Guo, Bin Yang, Yong Li
UniWind forecasts day-ahead delivered wind power by separating a site-calibrated, weather-driven physical reference from a correction for likely operating states. Future NWP weather conditions its bounded expert corrections. In cross-region zero-shot evaluation on SD_A, UniWind reports MAE 20.15±0.20 MW and RMSE 25.84±0.28 MW, versus Moirai’s 23.83±0.26 MW and 35.69±0.44 MW; its state categories are heuristic, not verified operational labels.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.