基于AQI的文山市空气质量统计预报研究

Research on Air Quality Statistical Forecast of Wenshan City Based on AQI

本文利用2015-2019年文山市气象监测站常规地面观测资料和文山市环境监测站空气质量指数(AQI)逐日平均值、各类污染物逐日平均浓度值等观测资料,通过统计学方法对不同季节气象要素和AQI做相关性分析,选取相关性好的因子做回归分析并分季节建立预报方程。结果表明,不同季节的AQI与各类气象要素相关程度不同,相较全年回归预报方程,分季节的回归预报方程的预报准确率更高;由于AQI的升高是污染物积累的过程,前一天的AQI与当天AQI相关系数高达0.795,在回归模型中加入前一天的AQI这项因子后,预测精度明显提高。

In this paper,by using the regular ground observation data of Wenshan City Meteorological Monitoring Station and the observation data of Wenshan City Environmental Monitoring Station from 2015 to 2019,such as the daily average value of air quality index(AQI)and the daily average concentration value of various pollutants,the correlation analysis of different seasonal meteorological elements and AQI is carried out by statistical methods,and the factors with good correlation are selected for regression analysis,and the forecast equation is established by season.The results show that the AQI in different seasons has different degrees of correlation with various meteorological elements,compared with the annual regression forecast equation,the forecast accuracy of the seasonal regression forecast equation is higher.Since the increase of AQI is the process of pollutant accumulation,the correlation coefficient between the AQI of the previous day and the AQI of the current day is as high as 0.795,after adding the AQI of the previous day into the regression model,the prediction accuracy is significantly improved.