Day‐Ahead Forecasting of Wind Power Generation and Electricity Market Prices Using Machine Learning

The rapid expansion of wind energy is fundamentally reshaping electricity price formation in liberalized power markets,increasing both volatility and forecasting uncertainty. This study investigates how short‐term wind generation forecasts influence day‐ahead electricity prices through a two‐stage data‐driven framework applied to long‐term, high‐resolution operational and market data. In the first stage, wind power generation is forecast using gradient boosting methods optimized via Bayesian techniques. In the second stage, the forecasted wind power is explicitly integrated into electricity price prediction models. Results demonstrate that incorporating wind forecasts substantially improves the representation of renewable‐driven market dynamics, capturing the merit‐order effect associated with low marginal‐cost generation. While wind power forecasting achieves high short‐term accuracy ( R 2 ≈ 0.88), its integration into price forecasting reduces systematic bias and enhances explanatory power ( R 2 ≈ 0.84) relative to standalone price models. The findings highlight the importance of renewable‐aware forecasting frameworks for efficient market participation, improved bidding strategies, and effective integration of wind energy into day‐ahead electricity markets.