Analysis And Prediction Of AQI In Jinan

In the 21st century,air pollution has become one of the top ten environmental problems in the world,which has aroused close attention in the world.Air pollution not only does serious harm to people’s health,but also causes certain losses to the production and development of industry and agriculture,and affects social and economic development.China’s economic strength and economic development speed are in a leading position in the world,but the process of development also have a certain impact on the environment.According to relevant research,the biggest environmental problem in China is air pollution.Our country is very concerned about this and has taken a series of measures.In the central working conference in 2020,it is clearly pointed out that we should fight against pollution,and The 14th Five-Year Plan also clearly points out that we should pay more attention to and strengthen the construction of ecological civilization.The air quality ranking of Jinan is relatively low.Based on the air quality data and meteorological data,this paper analyzes AQI and six representative pollutant lindexes and five meteorological factors are selected.Then,ARIMA,BP neural network and ARIMA-BP neural network combined models were established to predict AQI,and their fitting effects were compared.The prediction of AQI has two aspects of significance.One is to establish different AQI prediction models and compare the fitting effect of different models,so as to provide certain basis for relevant departments to predict AQI;Second,through the analysis of the change trend and time distribution characteristics of AQI in Jinan,it provides reference for comprehensive urban environmental management and planning,and puts forward corresponding suggestions for the environmental protection department to carry out pollutant emission control work.There are the following innovations in the establishment of the prediction model:first,five new meteorological factors are introduced in the establishment of BP neural network model,and a total of eleven influencing factors are introduced to predict AQI;Secondly,the traditional ARIMA model and BP neural network model are combined to predict AQI.The results are as follows:first,the AQI of Jinan has improved,but the proportion of polluted weather in 2020 is about 30%;Secondly,ARIMA model is more accurate in short-term fitting,but not accurate in long-term data fitting.In general,BP neural network model is better than ARIMA model in predicting AQI,and combination model is better than single model.On the basis of the research,this paper puts forward corresponding suggestions for the prevention and control of air pollution.It is hoped that through the joint efforts of the government and the public,the air quality of Jinan can be significantly improved and reach a better level.