Research On Atmospheric Pollution Concentration And Quality Evaluation Method Based On Grey System

In this paper, the main components of the air pollution in the capital city of five provinces in Northwest China, the main components of the cluster analysis method is used to establish a reasonable air quality evaluation model, using GM(1,1) analysis, BP neural network model and GNNM(1,1) to study the air quality of five cities in Northwest China.First of all, on the basis of the five major indicators of atmospheric environmental quality as the original data, including sulfur dioxide, nitrogen dioxide, carbon monoxide, inhalable particles(particle diameter less than or equal to) and fine particles(particle diameter less than or equal to). A qualitative analysis of the atmospheric environmental quality of five provincial capital cities in the North West of china. Using principal component analysis to extract three principal components of each index, F1 is the representative of industrial pollution factors. F2 is a response to civil heating. F3 reflects the dust factor. The comprehensive score of each city is analyzed, and the atmospheric environment quality is from 1(very good) to 3(very poor) to three levels. In cluster analysis, the conclusion is "noted" said Yinchuan scored the highest, was designated as the first class. Air quality of Lanzhou city and Xining city in second categories. The third category is Xi’an city and Urumqi City, the air quality is urgent to take measures to protect.Secondly, the GM(1,1) analysis method is used to forecast the atmospheric environmental quality of five provincial capitals in Northwest China. The gray neural network model GNNM(1,1) is introduced into the atmospheric environment quality forecast for the characteristics of the air quality and pollutant concentration. GNNM(1,1) model is a combination of gray and neural network, and it is a model to solve complex problem. The model is established by establishing a BP neural network to map the GM(1,1) model of the grey differential equation. GNNM(1,1) model is used to study the BP learning algorithm, and can be used to predict the atmospheric environment quality. The results show that the GNNM(1,1) model has better adaptability and higher prediction accuracy than the grey forecasting method, and can be applied to urban mass concentration prediction.In this paper, the principal component and clustering evaluation method is used to establish the evaluation model, and the objectivity and accuracy of the evaluation results are enhanced. Then, the gray model is established according to the main air pollutant concentration. Based on this, the grey neural network model is established, and the relationship between the concentration of air pollutant and meteorological parameters is discussed. After testing and analyzing the gray neural network model is a powerful tool to solve the nonlinear problem.