Stacking Model for Photovoltaic-Power-Generation Prediction

Despite the clean and renewable advantages of solar energy, instability photovoltaic power generation limits its wide applicability. In order to ensure stable power-grid operations safe dispatching grid, it is necessary develop a model that can accurately predict generation. As widely used prediction method, stacking has been applied in many fields. However, few studies have models research, we four different are based on extreme gradient boosting, random forest, light boosting decision tree generation, by using two datasets. The results show accuracy higher than single ensemble-learning model, Stacking-GBDT other models. proposed this research provides reference for accurate