{"ID":2873261,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.06311","arxiv_id":"2509.06311","title":"WindFM: An Open-Source Foundation Model for Zero-Shot Wind Power Forecasting","abstract":"High-quality wind power forecasting is crucial for the operation of modern power grids. However, prevailing data-driven paradigms either train a site-specific model which cannot generalize to other locations or rely on fine-tuning of general-purpose time series foundation models which are difficult to incorporate domain-specific data in the energy sector. This paper introduces WindFM, a lightweight and generative Foundation Model designed specifically for probabilistic wind power forecasting. WindFM employs a discretize-and-generate framework. A specialized time-series tokenizer first converts continuous multivariate observations into discrete, hierarchical tokens. Subsequently, a decoder-only Transformer learns a universal representation of wind generation dynamics by autoregressively pre-training on these token sequences. Using the comprehensive WIND Toolkit dataset comprising approximately 150 billion time steps from more than 126,000 sites, WindFM develops a foundational understanding of the complex interplay between atmospheric conditions and power output. Extensive experiments demonstrate that our compact 8.1M parameter model achieves state-of-the-art zero-shot performance on both deterministic and probabilistic tasks, outperforming specialized models and larger foundation models without any fine-tuning. In particular, WindFM exhibits strong adaptiveness under out-of-distribution data from a different continent, demonstrating the robustness and transferability of its learned representations. Our pre-trained model is publicly available at https://github.com/shiyu-coder/WindFM.","short_abstract":"High-quality wind power forecasting is crucial for the operation of modern power grids. However, prevailing data-driven paradigms either train a site-specific model which cannot generalize to other locations or rely on fine-tuning of general-purpose time series foundation models which are difficult to incorporate domai...","url_abs":"https://arxiv.org/abs/2509.06311","url_pdf":"https://arxiv.org/pdf/2509.06311v1","authors":"[\"Hang Fan\",\"Yu Shi\",\"Zongliang Fu\",\"Shuo Chen\",\"Wei Wei\",\"Wei Xu\",\"Jian Li\"]","published":"2025-09-08T03:26:18Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[\"Transformer\"]","has_code":false,"code_links":[{"ID":610038,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2873261,"paper_url":"https://arxiv.org/abs/2509.06311","paper_title":"WindFM: An Open-Source Foundation Model for Zero-Shot Wind Power Forecasting","repo_url":"https://github.com/shiyu-coder/WindFM","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}