基于WRF-CMAQ的空气质量等级预报模型

Air Quality Index Prediction Model Based on WRF-CMAQ

摘要: 空气质量与气象条件之间存在较强的相关关系，目前我国常用的WRF-CMAQ模型预测效果并不理想，本文在此一次预报数据的基础上，结合实测气象数据、实测主要污染物浓度数据，建立了空气质量二次预报模型。利用中国环境监测总站发布的相关数据，对污染物浓度进行K均值聚类，将天气分型，再按类别逐步回归拟合模型，最后分析二次预报模型的预测准确性。结果表明相较于一次预报，二次预报准确率显著提升。

Abstract: There is a strong correlation between air quality and meteorological conditions, and the predic-tion effect of the commonly used WRF-CMAQ model is not ideal. Based on this first prediction data, combined with the measured meteorological data and the measured pollutant concentration data, this paper establishes a secondary prediction model for air quality. Using the relevant data released by China Environmental Monitoring Station, the pollutant concentration is clustered by K-means, the weather is classified, and then the model is fitted by stepwise regression according to categories. Finally, the prediction accuracy of the secondary prediction model is analyzed. The results show that the accuracy of the second prediction is significantly improved compared with the first prediction.