Forecasting PM_(2.5) and PM_(10) concentrations using GMCN(1,N) model with the similar meteorological condition: Case of Shijiazhuang in China

摘要

PM10 and PM2.5 (particulate matter with particle size is not greater than 10 and 2.5 mu m respectively) are major components of environmental pollution, which can seriously affect people's health. Therefore, the prevention of these pollutants is very necessary. In this paper, the daily concentrations of PM10, and PM2.5 are predicted by considering the meteorological conditions. Firstly, the similarity of meteorological conditions is calculated by grey relation analysis. Then, based on the similarity, grey multivariable convolution model with new information priority accumulation (GMCN(1,N)), grey multiple linear regression (GM(0,N)) and multiple linear regression (MLR) models are established to forecast the daily concentrations of PM2.5 and PM10, in Shijiazhuang. The results show that the GMCN(1,N) model is the best among three models and can provide reliable prediction results and important guide to air control.