Research On Air Pollution In Beijing Tianjin Hebei Region Based On Spatio-temporal Interaction Model

In recent years,the frequent occurrence of smog in the Beijing-Tianjin-Hebei region has drawn widespread attention to air pollution in the region.Fine particulate matter,known as PM2.5,is a major contributor to air pollution.The study of the relationship between PM2.5 and its influencing factors is helpful to determine the impact of air pollution on people’s production and life,so as to conduct more accurate research and treatment of air pollution.The spatio-temporal interaction model can reflect the influence of independent variables on dependent variables from three aspects: time effect,space effect and spatio-temporal interaction effect.The model includes two parts: spatio-temporal process and white noise process.Among them,the spatio-temporal process is composed of the mean value process and the spatio-temporal variation trend,which includes the effects of covariates,time and space effects and space-time interaction in the study.The white noise process represents the measurement error.Based on the spatio-temporal interaction model,this paper studies the air pollution in Beijing-Tianjin-Hebei region,and improves it in practical application by introducing the influencing factors of air pollutants.Using the model,hourly data of PM2.5concentrations at 78 air pollution monitoring stations in the Beijing-Tianjin-Hebei region in2018,as well as data on air pollutants and meteorological influencing factors during the same period were modeled and analyzed.In the study,firstly,FCM clustering algorithm was used to divide 78 air pollution monitoring stations in The Beijing-Tianjin-Hebei region into 7 categories according to longitude and latitude information.Secondly,the model of error term for heteroscedasticity and with variance distribution assumption,and the random sampling of consistency is estimated based on the robust regression algorithm and weighted least squares method to estimate the parameters in the model,and connecting with the actual meaning of the parameters,the distribution of the error term for heteroscedastic model is chosen as the final fitting model.Studies show that PM2.5 concentration is negatively correlated with temperature and wind speed,and positively correlated with relative humidity in The Beijing-Tianjin-Hebei region.The spatio-temporal interaction model of The Beijing-Tianjin-Hebei region,combined with the parameter estimation results,can give a reasonable explanation to the actual situation.At the same time,the model has a good fitting degree of 77.8% on the data of this region in 2018,which can well reflect the relationship between PM2.5 and its influencing factors,providing a new idea for the study of air pollution in this region and other regions.