Construction of Wind Farm Load Combination Forecasting Model Based on GBDT, LightGBM and RF

To enhance the accuracy and effectiveness of wind farm load forecasting results, this study investigates the development of a machine learning-based wind farm load combination forecasting model. Firstly, three types of machine learning regression prediction models, namely Gradient Boosting Decision Tree (GBDT), Light Gradient Boosting Machine (LightGBM), and Random Forest (RF), are established. Secondly, particle swarm optimization (PSO) is employed to optimize the weighted coefficients of the combination prediction model. Finally, a GBDT-LightGBM-RF combination prediction model is constructed. The empirical analysis demonstrates that the mean square error (MSE) value of the combination forecasting model is 0.216, which is lower than that of individual forecasting models. The machine learning-based combination forecasting model significantly enhances wind power generation forecast accuracy and holds valuable implications for wind power generation planning in diverse regions.