An Air Pollutant Forecast Correction Model Based on Ensemble Learning Algorithm

In recent years, air pollutants have become an important issue in meteorological research and indispensable part of quality forecasting. To improve the accuracy Chinese Unified Atmospheric Chemistry Environment (CUACE) model’s pollutant forecasts, this paper proposes a solution based on ensemble learning. Firstly, forecast results CUACE model corresponding monitoring data are extracted. Then, using feature analysis, we screen correction factors that affect quality. The random forest algorithm, XGBoost GBDT algorithm employed to correct prediction PM2.5, PM10, O3. further optimize model, introduce grid search method. Finally, compare analyze effect determine best for three pollutants. This approach enhances precision improves our understanding experimental show has better error than traditional machine learning statistical model. After correction, PM2.5 PM10 is increased by 60%, O3 70%.