Tyan-WP: A Wind Power Foundation Model for Ultra-Short-Term Probabilistic Forecasting
Researchers have developed Tyan-WP, a novel foundation model designed for ultra-short-term probabilistic forecasting of wind power generation. This model is the first of its kind, specifically engineered to address the limitations of existing methods in data-scarce or generalized scenarios. Tyan-WP leverages static site attributes and a power-aware meteorological fusion module to achieve accurate zero-shot forecasting, outperforming numerous established models in both in-domain and cross-geography evaluations. AI
IMPACT Introduces a specialized foundation model for wind power forecasting, potentially accelerating new wind farm deployments and improving risk management.