Assessment And Forecast Of PM_2.5 Based On Conventional Atmospheric Monitoring Data

Fine particulate matter in the atmosphere is the particle size is less than or equal to 2μm(sometimes with less than 2.5μm, PM2.5) particles. but PM2.5 said the content of particles per cubic meter of air. The higher this value, on behalf of more severe air pollution. By the method of mathematical statistics analysis of the main influencing factors of PM2.5, and the use of these factors to establish the PM2.5 regression model, with scientific research value.This article from the provincial environmental monitoring center official website of Jiangsu, the provincial environmental monitoring center official website of Jilin and the weather network, including PM2.5 concentration, PM10 concentration, CO concentration, SO2 concentration, NO2 concentration, temperature, weather, wind data. Through principal component analysis, multiple regression model and other statistical methods, to study the relationship between PM2.5 value and other observation index, comparative analysis and typical pollutants in South and north of atmospheric pollution andpollutant formation. The final regression model PM2.5 other indicators, and the regression model is verified. In the guarantee is not related, the independent variable conditions, to test the model,choice of adjusting the R2 value of the largest model, using independent component analysis of the multivariate regression model as the final prediction model. Experimental results show that the estimation of the regression model, the root mean square error is within 1, can effectively estimate PM2.5. Through this study, reasonable prediction and evaluation put forward reasonable scientific advice in atmospheric environment PM2.5 in China, provide safe and comfortable living environment for the people.