Optimizing Wind Energy Forecasting: A Comparative Study of ANN and Decision Tree Algorithms

Accurate wind power forecasting is crucial for efficient power grid operation with significant wind power integration. This study compares tree-based learning algorithms and advanced machine learning methods to optimize forecasting. Seventeen-year models using tree-based algorithms showed accuracy degradation with longer sampling intervals and height extrapolation. Models were tested with varied heights and locations, demonstrating tree-based algorithms' long-term forecasting effectiveness. Combined datasets and the k-nearest neighbors algorithm filled missing data. Statistical indicators showed ANN's robust forecasting capabilities. MATLAB and Jupyter Python provided insights into optimizing models for reliable wind energy production.