Comprehensive Evaluation Of Ambient Air Quality Based On Multiple Models In Anhui Province

This paper constructs an improved grey clustering model and applies it to the study of air quality evaluation in Anhui province.On the basis of taking the air quality of Anhui during 2015²017 as the clustering object,using the annual average concentration of six main air pollutants(SO₂,NO₂,CO,PM₁₀,O₃,PM_2.5)as clustering indicators,and dividing into six ashes,we use the improved normal distribution grey clustering model to cluster analysis.The results show that:during 2015²017,the air quality in Anhui province generally rose first and then declined;the grade of overall air quality were Class II;the primary pollutant was PM_2.5;The order of primary and secondary pollution factors is PM_2.5>PM₁₀>O₃>NO₂>CO>SO₂.In this paper,the basic principles and methods of artificial neural network are applied to air quality assessment of Anhui province.Firstly,we interpolate a large number of training and test sample data between concentration limit standards by linspace function;then,comprehensively considering six indicators and establishing RBF neural network model by the ANN toolbox in MATLAB to study the classification of air quality in the evaluation level,so as to determine the results of air quality quickly and accurately.The air quality data of Anhui province is effectively simulated by the powerful function of machine learning,and compare it with the results of neural network evaluation.The results show that:the conclusions of the two methods are very similar,and the results of air quality in Anhui during 2015²017 are Grade II.Based on the development characteristics and natural geographical conditions of Anhui Province,a comprehensive statistical analysis and vertical and longitudinal comparison are carried out on the weekly,monthly,seasonal and annual values of PM_2.5 and PM_10-2.5 in 16 cities of Anhui province during 2015²017.The results show that:in 2017,the annual average concentrations of PM_10-2.5 and PM_2.5 in Anhui were35?g/m³ and 57?g/m³,respectively,which increased slightly compared with the previous year,and decreased from 2015 to 2017;particulate matter showed different weekly effects,monthly rules and seasonal characteristics at different time scales.The spatial distribution pattern of particulate pollutants in Anhui province was preliminarily obtained by using the statistical analysis software such as ArcGIS and SPSS,and the relationship between particulate matter and other air mass index was discussed.This study has a certain reference value for the evaluation of regional ambient air quality.