Model And Analysis To The Data Of PM_2.5 Daily Concentration In Huangshi

PM_2.5 is a diameter of less than or equal to 2.5 micron particles which can be directly lunged.The accurate prediction of PM_2.5 plays a key role in atmospheric pollutant and management,and many experts have researched on this issue.But most models are about the air pollution index and the air quality evaluation,the model of PM_2.5 and the air quality prediction are less studied.In specilally,it is very little studied about the PM_2.5 model with missing data and the method is simple.This paper mainly discusses the building of PM_2.5 time series model,fitting the optimal of PM_2.5 model and the effective forecast to the future of PM_2.5.Where,We use the supplement of missing values and the appropriate replacement to the missing values.The article include six chapters as follows:The first chapter introduces the background,significance and current situation of the study and the main works.The second chapter introduces some usual methods of the supplement of missing values and the characteristics.In the third chapter,introduce the three classic time model,modeling process and forecasting.In the fourth chapter,introduce the based on key point technology,a method for time series similarity matching is proposed.In the fifth chapter,on the base of the daily PM_2.5 concentration data of Huangshi city in 2015,the missing date is filled by EM method.we chooses the different models in diffrent periods in order to improve the accuracy of forecast.Compared with the Tie shan forecast results and consider the feasibility of the model.In the last chapter,we summarize our results and give some improved advices.