Grey Multivariable New Accumulation Prediction Model And Its Application

In recent years,air pollution has become a global concern,and it seriously affects people's production and living activities.Therefore,air pollution is a problem that people urgently need to solve.In the process of solving the problem of air pollution,the prediction of air pollutant concentration is an important part,which can provide scientific basis for the formulation of air pollution control measures.Therefore,prediction is a necessary and meaningful work in the process of air pollution control.With the deepening of the prediction work,different models are applied to the field of air pollution control,and people's requirements for the model prediction accuracy are getting higher and higher.Considering that the daily concentration of pollutants varies greatly and is affected by meteorological conditions,a new grey multivariate prediction model is proposed in this paper,and the daily concentrations of PM_2.5 and PM₁₀ are predicted with the new model to improve the prediction accuracy.Firstly,from the perspective of changing the accumulation generation operator,GM?0,N?model with new accumulation is proposed by combining the new accumulation generation operator with parameter with GM?0,N?model in this paper.Through the study of its properties and the application in examples,it is proved that the new accumulation generation operator can not only make the new information play a greater role,but also improve the prediction accuracy of the model.Secondly,based on the advantages of the accumulation generation operator,the grey multivariable convolution model with new information priority accumulation is proposed by introducing a new accumulation generation operator into the grey multivariate prediction model with convolution integral,and the validity of the new model is verified by some cases.Finally,the new grey multivariable accumulative convolution model with similar information priority is built by using grey relational degree to select the data that are highly similar to the weather conditions to be predicted,and the superiority of the model is verified by comparing with other grey models.Then,the daily concentrations of PM_2.5 and PM₁₀ in Xingtai city are predicted with this model.