Research on Wind Power Generation Forecasting Based on Explainable Machine Learning

With the development of renewable energy, the grid connection operation of wind power generation has brought huge challenges to the safety and stability of the power system. This paper constructs three wind power generation prediction models, namely artificial neural network (ANN), long short-term memory network (LSTM), and extreme gradient boosting (XGBoost), and proposes to use the Shapley additive explanation (SHAP) method to interpret these models, in order to reveal the influence of each feature on the wind power generation prediction results. Simulations results carried out in Python demonstrate the superior performance of prediction models and conduct a thorough interpretability analysis on the XGBoost model.