Design Of Air Quality Forecasting And Warning System Based On Ensemble Forecast

With the rapid development of industrialization in China,the demand for fossil energy is increasing,which leads to the increase of harmful substances discharged into the atmosphere.Various environmental pollution incidents emerge in an endless stream.The scale of development of environmental pollution incidents and the consequences of damage are expanding day by day,which have had an impact on China's economic development.In many areas of China,the type of air pollution has changed from single coal smoke pollution to multiple pollutant interaction complex pollution.How to effectively control the deterioration of ambient air quality and accurately forecast the air quality is an urgent problem to be solved.Numerical forecast model plays an important role in atmospheric forecast.However,due to the uncertainty of numerical forecast model itself and the great influence of initial disturbance on numerical forecast model,the prediction results of single numerical forecast model are inaccurate.Aiming at the above problems,this paper presents a ensemble forecasting method based on the optimal fixed weight combination forecasting method,and designs and develops an air quality early warning and forecasting system based on ensemble forecasting method.The system realizes the functions of forecasting the ambient air quality of every region in Yunnan Province in the next 24 hours,information sharing and consultation with environmental protection departments,and effectively helps environmental protection departments to monitor and control the ambient air quality.The main contents of this paper include the following aspects:(1)An ensemble forecasting method based on optimal fixed weight combination forecasting method is proposed.Based on the prediction results of several single air quality numerical forecast models and the principle of minimizingthe sum of squares of combined forecasting errors,an optimal fixed weight combined forecasting model is established.Compared with the existing multiple linear regression method and dynamic weight updating method,the effectiveness of the proposed method is verified.(2)An air quality early warning and forecasting platform is designed.The functional and non-functional requirements of the system are given.The related technologies for designing the system are introduced.The overall structure of the system,main functional modules,databases and related interfaces are designed.(3)Implementation and testing of the system platform.The realization of the main functional modules of the system is given,and the system functions,compatibility,performance and other aspects are tested in detail to ensure that the system can run continuously and reliably.