Analysis And Prediction Research On Influencing Factors Of Air Quality In Qingdao

Based on the data of air pollutant concentration and meteorological factors in Qingdao and its surrounding cities from spring 2014 to winter 2018,this paper explores the temporal and spatial distribution characteristics and evolution rules of air quality in Qingdao and its surrounding cities,and forecasts the concentration of air pollutants in Qingdao,and studies the temporal and spatial and meteorological factors affecting air quality in Qingdao.The main research contents and results are as follows:First of all,the data are classified by year,month and season,and a series of statistical analysis charts are drawn,so as to intuitively explore the characteristics and evolution rules of various air pollutants in the year,month and season.It is found that the six main pollutants in Qingdao have a strong dynamic relationship with the surrounding cities,and all of them have obvious seasonal and periodic characteristics,with low pollutant concentration in summer and autumn At the beginning of the season,the concentration of pollutants reaches the highest in winter,and then falls back in spring,which circulates in turn,forming a periodic fluctuation trend with four seasons as a cycle.Secondly,through Pearson correlation coefficient,this paper explores the correlation and intensity between various air pollutants and meteorological factors in Qingdao.The results show that Qingdao air quality index has obvious positive correlation with PM2.5,PM10,sulfur dioxide,nitrogen dioxide,carbon monoxide,and negative correlation with ozone;meteorological factors and air quality index and main air pollutants There is a significant correlation between the dyes,and there is a significant difference in different seasons.Finally,three kinds of models are established to compare and analyze the two kinds of representative air pollutants according to the wind direction of Qingdao on the day of prediction.It is found that the prediction effect is better after using the pollution sources of surrounding cities.In the prediction ability of the three kinds of models,multiple linear regression is the worst,neural network is the second,nonparametric regression is the second The result is better.