Analysis And Inference Of Air Quality In Jinan Based On AQI

With the development of science and technology and the progress of society,people enjoy the improvement of life quality,but also suffer from air pollution.Nowadays,the whole society attaches great importance to air quality,but it is far from enough to rely only on subjective judgment of air quality changes Therefore,it is particularly important to quantitatively analyze air quality and make accurate prediction of future air quality.AQI is an index that quantitatively describes air quality,and it is of great practical significance to analyze and infer air quality based on AQI.For the research on AQI,most scholars use BP neural network model,Grey prediction model and ARIMA model,but there are relatively few researchers using ARIMA seasonal model to research AQI according to its seasonalityFirstly,this paper use the octopus collector to obtain AQI and six major pollutant concentration data of Jinan from 2014 to 2019,and using ggplot2 in the R software to visually analyze the air quality in the past six years.The results show that good days increased in recent years,the air quality showed an overall improvement trend and the characteristics of AQI and pollutants showed obvious seasonality.Then,the multivariate regression equation of AQI and pollutants during heating and non-heating period was established by using the multivariate statistical analysis method,the results show that the pollutant with greater influence on AQI was PM2.5,which providing a solid theoretical basis for targeted air pollution control.The regression model of two periods is used to predict AQI,and the prediction effect is good,but this model cannot predict the AQI when the independent variables are unknown,so a time series model is further selected.Finally,using Additive seasonal model based on ARIMA to predict AQI,it is found that this model has a better prediction effect.The prediction results show that the relative error between the predicted value and the true value is controlled within 12%,so the model has higher accuracy.Therefore,the additive seasonal model is used to predict the AQI in the next five months,the prediction results are in line with the seasonal characteristics of AQI,which is reasonable and credible,and provides a basis for the future treatment of air pollution.