Air Quality Prediction And Influencing Factors In Shijiazhuang

With the rapid development of social economy and the rapid expansion of population,a series of problems caused by environmental pollution are becoming more and more obvious.As the most polluted city in Beijing,Tianjin and Hebei,Shijiazhuang is the provincial city of Hebei Province.How to effectively improve air quality and prevent air pollution has become a necessary condition to promote the coordinated development of Beijing,Tianjin and Hebei.Therefore,the use of advanced and scientific methods to accurately predict air quality can improve the ability to deal with air pollution,make people more intuitive understanding of air pollution and timely take effective prevention and control measures.Based on the air quality data of 11 monitoring stations in Shijiazhuang city from 2016 to2019,this paper first uses descriptive statistical analysis to study the spatial and temporal distribution characteristics of air quality of each station;secondly,uses multiple linear regression analysis and decision tree to establish air quality prediction model,and analyzes the primary factors affecting air quality,and then uses them The time series model predicts the air quality index?AQI?of each station in the next six months in a short term,and analyzes the range of its prediction error.Finally,based on correlation analysis and regression analysis,the correlation between AQI and meteorological factors and socio-economic factors is studied,and the most influential factors of AQI are obtained.The analysis results are as follows:?1?From the trend point of view,the air quality of Shijiazhuang is getting better and better on the whole,but the situation of some monitoring stations is still not optimistic.The concentrations of AQI,PM_2.5,PM₁₀,NO₂,SO₂ and CO of stations in Shijiazhuang city from2016 to 2019 all showed a downward trend,while the concentration of O₃ was in an upward trend as a whole,and some monitoring stations showed fluctuations.?2?From the perspective of quantitative prediction and qualitative prediction,this paper uses multiple linear regression model and decision tree model to carry out empirical analysis and verify the feasibility.According to the regression coefficient and the importance of variables,the main pollutant is PM_2.5.Aiming at a series of problems caused by PM_2.5.5 on the air quality of Shijiazhuang City,this paper puts forward several prevention and control measures.In this way,the government is making governance policies When making a policy,we should aim at the target.?3?Based on the AQI data of each monitoring station,ARIMA model was established by R software to predict the short-term,and finally ARIMA?0,1,2?×?1,1,0?₁₂2 was established.All the models passed the test.Arima is used to predict AQI in the next six months.Comparing the predicted data with the actual data,it is found that the errors are all within the controllable range,which indicates that the model fits well.?4?Using correlation analysis to study the relationship between AQI and meteorological factors,it is found that AQI has a negative correlation with wind speed and air temperature,that is,with the increase of air temperature or wind speed,AQI will also decrease,and air quality will be relatively better;based on the stepwise regression analysis model,the social and economic factors affecting AQI are empirically analyzed,and the results show that car ownership is the most influential factor for AQI,In view of the increase of the number of cars,some control measures are put forward.This paper not only describes the change of air quality of each station,but also shows the development trend of AQI in the next six months.It also reveals the existing air quality problems in Shijiazhuang,which has important policy guidance and practical significance for the prediction and sustainable development of air quality in Shijiazhuang city.