Air Pollution Source Identification And Air Quality Study Based On Functional Data Analysis

As China's economic development has entered a new normal,urban air pollution has become increasingly serious,and the quality of the atmospheric environment has become an important constraint to the coordinated development of China's social and economic development.In recent years,in order to better implement the national policy of sustainable development,Jiangxi province has continued to issue many policies on environmental protection,and started to build a national ecological civilization pilot zone,striving to improve the environment of Jiangxi province.Based on the environmental quality,combining with the meteorological and economic fields,this paper makes use of the air quality,meteorological and economic data of 11 districts and cities in Jiangxi province,and studies the environmental quality of Jiangxi province from the perspective of functional data.First of all,analysis of Jiangxi province in 2018 the space-time characteristics and the main variation of PM2.5 index,the observation of the discrete data curve,the study found that Jiangxi PM2.5 concentrations significantly in time and space characteristics and regional characteristics,the province is one of the most polluted year period in winter,and the main pollution area of pingxiang and jiujiang,the local climate conditions and economic development are closely related.Secondly,the use of functional principal component analysis method,the detection of nanchang in 2015-2018 quarters of air pollution,identify each quarter of the main pollutants,quarterly analysis can more clearly different quarters of pollution sources,is conducive to improve air quality,better results for nanchang summer air quality best,main pollutants are O3 and CO,and pollution is relatively serious in winter and spring,pollutant mainly for PM2.5 and PM10;The mean test method of functional data was further adopted and the statistical quantity₍,?9? was used to compare the air quality changes in jiangxi province in recent years.After the implementation of various environmental protection policies and bans in recent years,the broad masses of nanchang city continued to support and make efforts,and nanchang city has achieved remarkable results in environmental improvement.Finally,this paper analyzes the linear regression analysis model of the functional data of multiple independent variables,and adds the penalty term when solving the parameter function,and USES this new method to estimate the parameter function of the functional multivariate linear model.Using this new method,considering the air quality,meteorological,environmental and economic aspects of the data fitting function type linear regression model to study the air quality in Jiangxi province,at the same time,considering the general linear regression model and the general function type linear regression model,and then to get the results of different methods were compared,the results show that the prediction error of the functional model with penalty term is the smallest,and the study finds that there is cross-pollution in the air pollution of Jiangxi province,the concentration of PM2.5 is significantly correlated with PM10,O3 and NO2,and the economic indicators also have an impact.This paper USES the functional data analysis method,through the analysis of the space-time characteristics of PM2.5 data in Jiangxi province,the recognition of air pollution,environmental quality improvement in Jiangxi province,study the main influencing factor of the main pollutants PM2.5 in Jiangxi province,objectively to grasp the current air quality situation of Jiangxi province,analyzed the relationship between the air condition,put forward the corresponding policies and Suggestions,in the future environmental governance has a certain reference significance of Jiangxi province,at the same time,the study found that functional data analysis method has certain advantages in the practical problem solving,deserves further research.