The Research On Air Traffic Flow Statistics And Prediction Technology

With the rapid development of the air transportation in our country, the increasing air traffic flow conflicted with the air traffic service capacity, which is growing more obviously. The conflict brings traffic congestion that not only threatens the safety of the aircraft, but also leads to amounts of flight delay and unnecessary economic loss. Therefore, the air traffic flow monitoring and management have become a significant problem which needs to be resolved. The statistics and prediction of the air traffic flow is the foundation of Air Traffic Flow Management, it can be offered as the technical support and data reference for scientific strategies of Air Traffic Flow Management.Referred to the domestic and oversea researches, this paper started with introduction of the methods for air traffic flow prediction. Then, this paper studied the multi-models forecasting methods based on combination in order to predict the strategic air traffic flow with the historical data. The tracks correlation method was proposed after studying various prediction models and analyzing the inaccuracy of the results. The algorithm was validated by using the practical operation data from ZGGG. The result was derived from comparing fix point time predicted with the actual flight data, its error is within±30sec. Further, the trajectory prediction model was established, which can make effective statistics and prediction for tactical air traffic flow in a given interval and space condition. Besides, this paper optimized and improved the existing Air Traffic Flow Management System with redesigns of database and program structure. Finally, Air Traffic Flow Statistics and Prediction System for DALIAN was designed and implemented, and then the results of test run were analyzed.