Functional Data Analysis Of Air Quality In Fen-Wei Plain

The Fen-Wei Plain is one of the four major plains in China.The regional energy structure is mainly coal,with more heavy chemical enterprises and relatively single industrial structure.During the heating period,the pressure of environmental protection in prefecture level cities in Fen-Wei Plain increases,and the emission of heavy vehicles on highways is large,coupled with the special topography and meteorology of the river valley,which is not conducive to the diffusion of pollutants.In recent years,China has formulated a plan to include the Fen-Wei Plain in the key areas of air pollution prevention and control.In order to better implement the national policy of sustainable development,prefecture level cities in Fen-Wei Plain have issued a number of environmental protection policies to vigorously carry out environmental pollution prevention and control while developing economy.Because the air quality problem in Fen-Wei Plain has the dual characteristics of time and space,the traditional statistical methods have lost their original advantages,and it is necessary to introduce a new analysis method,namely functional data analysis,according to the characteristics of the data.Based on the perspective of functional data,this paper uses the original data related to air quality in 11 prefecture level cities from 2020 to 2021 to explore the environmental quality problems in Fen-Wei Plain.The main works were as follows.Firstly,by analyzing the daily average AQI concentration data、environment good day data the occurrence days of the primary pollutants,the air quality characteristics of the Fen-Wei Plain in 2020 and 2021 were explored.It was found that the change trend of air quality in the Fen-Wei Plain had certain seasonal characteristics,and the primary pollutants were mainlyPM_2.5、PM₁₀and ₃.Secondly,the functional principal component analysis is carried out on the daily average AQI concentration data of prefecture-level cities,and it is found that the air pollution in the Fen-Wei Plain has a"concave"trend of high air pollution in spring and winter and low air pollution in summer and autumn.It should be noted that the use of functional principal component analysis can reduce the dimension of infinite dimensional data.Thirdly,the functional cluster analysis is performed using the reduced dimension data.Using K-means cluster analysis and hierarchical cluster analysis methods,the 11 cities in Fen-Wei Plain were divided into four categories,and the clustering results were visualized to intuitively reflect the impact of climate change,human activities and geographical location on air quality changes.Finally,functional regression analysis was performed.The average annual concentration of AQI was used as the response variable,and the average daily concentration ofPM_2.5was used as the predictor variable to establish a functional linear regression model,which showed that the change of average daily concentration ofPM_2.5had a significant impact on the air quality index.Based on functional data and data on six indicators affecting air quality,this paper selects air quality index data to study the spatial-temporal distribution characteristics,primary pollutants and influencing factors of air quality in 11 prefecture-level cities in the Fen-Wei Plain in 2020 and 2021,analyzes the air quality situation in the Fen-Wei Plain,and puts forward suggestions for pollution prevention and air quality improvement in the Fen-Wei Plain based on the full-text analysis.To provide theoretical support for the implementation of follow-up prevention and control measures.