Abstract
Accurate short-term wind power forecasting is essential for grid stability and operational planning, yet remains challenging due to the complex interactions between atmospheric conditions and turbine dynamics. Existing methods fail to effectively incorporate weather forecasting with wind turbine data (i.e., SCADA), leading to suboptimal solutions. To address this, we introduce a multimodal framework that integrates historical point-based SCADA data with grid-based Numerical Weather Prediction (NWP) forecasts, which is challenging due to heterogeneous input and the complex physical wind-turbine interactions.
Our approach first explicitly decomposes inputs into scalar and vector features to better capture both site-specific and geometric dependencies. It then incorporates a geometric encoder to extract rotation-invariant features from wind vectors. Furthermore, we leverage a Fourier Neural Operator (FNO) architecture, which performs global convolutions in the frequency domain to efficiently model long-range spatiotemporal relationships. Extensive experiments on three real-world wind farms, with weather forecasting data, demonstrate that our model consistently outperforms state-of-the-art baselines, highlighting the effectiveness of its physically-informed design.
Core implementation is publicly available at: GitHub - GWPF.
Blogger's Review: This paper presents a multimodal framework that significantly enhances short-term wind power forecasting accuracy by integrating SCADA data and NWP forecasts. The importance of physical knowledge in model design is evident, showcasing the potential for future applications in power systems. The use of Fourier Neural Operator highlights the advantages of frequency domain analysis in spatiotemporal modeling, warranting further exploration and promotion.