Air Traffic Flow Prediction And Analysis Of Influencing Factors Based On Weighted Combination

With the continuous development of social economy,civil aviation transportation continues to maintain high-speed growth,resulting in increased air traffic flow and increased air traffic congestion.By predicting the flight flow of some specific airspace,some aviation management departments and airlines can prepare in advance according to the predicted results And deployment,to flexibly and reasonably formulate control strategies and rationally allocate airspace to improve operational efficiency.For this task,based on the characteristics of the predicted objects,this paper uses historical statistical data on the number of takeoffs and landings and flight flow in the Jinan control area from 2007 to 2016 to analyze the influencing factors based on the gray correlation and consider the impact of the higher correlation Scientifically predict the flight flow in Jinan control area and establish a quantitative prediction model.With reference to relevant research results,this paper selects the GM(1,1)model,exponential smoothing model,neural network model,multiple linear regression model and geometric average coefficient model,in a single prediction Based on the advantages of the method and optimizing it,a Simpson formula with higher accuracy is used to optimize the background value of the GM(1,1)model.The series gray BP neural network model and the parallel gray neural network model are based on the correlation coefficient and the average error.Screening,selecting appropriate models for combination,and giving different model weights;combining Critic and entropy weights to combine weighting models and combining the advantages of weighted Markov chains to reduce the influence of interference factors to optimize them and determine the Jinan control area The weight of the optimal weighted combined model of flow;the combined model of IOWA operator is introduced,and the accuracy of the model of IOWA operator in each period is optimized by Markov chain,and the optimal weighted combined model of approach flow of Jinan control area is determined Weights.Finally,based on the state transition matrix of the Markov chain,the predicted flight flow of the Jinan control area from 2019 to 2023 is obtained,which will provide some reference for the future airspace planning and sector division of the Jinan control area.