The Prediction And Optimization For PM2.5 Based On Neural Network Method

With the rapid development of China’s economy,the problem of air pollution has become increasingly serious,and the ecological environment has become a matter of great concern in today’s society.As an important air pollutant,atmospheric particulate matter PM2.5 has adversely affected people’s normal production,life and work,and harmed people’s health.Therefore,it is of great practical significance to analyze and study the prediction of the concentration of PM2.5.First of all,in order to prove that the prediction model has wide applicability,the selected research areas are Beijing,Chengdu,Guangzhou,Shanghai and Shenyang,which are five representative cities.In order to explore the factors affecting PM2.5 concentration,this thesis explores the influence of meteorological and ambient air conditions on PM2.5 concentration through the correlation analysis of season,temperature,wind speed,air pressure,humidity and PM2.5 concentration.Secondly,the ARIMA model and the gray prediction model are used to calculate the concentration prediction value of PM2.5 in Shenyang.In order to obtain a better prediction model,the LSTM neural network prediction model is established and fitted to Beijing,Chengdu,The trend of PM2.5 concentration in Guangzhou,Shanghai and Shenyang in one year,the PM2.5 concentration was predicted.The advantage of the LSTM neural network prediction model in this thesis is that the input variables are selected objectively,the influencing factors are comprehensive and the research is considered.The area targeted by the object is more universal.Finally,the results of various prediction models are evaluated,and the validity and applicability of the model are analyzed to provide a basis for the selection of models in practical applications.