Research On Nonlinear Characteristics Of Air Traffic Flow

Insights into the temporal and spatial characteristics of air traffic flow is the prerequisite of air traffic flow simulation, prediction and management. By using the real trajectory data,nonlinear characteristics of air traffic flow were explored in the chaotic and fractal perspectives.Firstly, based on the characteristics of integrated trajectory and ADS-B trajectory database, the trajectory processing methods were proposed. And based on the characteristics of different kinds of wrong tracks, the method to identify and process the wrong tracks was proposed. To solve the problems existing in current traffic flow identification studies, a method to identify air traffic flow based on air route network was proposed. The series of flying by navigation points could be found while the traffic flow was identified. And the air traffic flow time series from macro and micro perspectives were constructed.Secondly, by using the Maximal Lyapunov Exponent and Recurrent Plot, the chaos of air traffic flow was investigated from the macro and micro perspectives. Chaos exists in air traffic flow, both from the macro and micro perspectives. The chaotic characteristics of air traffic flow system were not simply determined by the air traffic flow volume, the density of traffic flow or the type of air traffic system. For air traffic flow which showed more significant chaotic characteristics from the macro perspective, it would show more significant chaotic characteristics from micro perspective too.Thirdly, by calculating the Correlation Dimension, the fractal characteristics of traffic flow were studied from the macro and micro perspectives. When analyzed from different perspective, the same traffic flow had different fractal dimensions. When studied from the macro perspective, the fractal dimensions of different traffic flow from converging en-route air traffic varied a lot, and the fractal dimensions of different traffic flow from runway air traffic were close. When studied from the micro perspective, the fractal dimensions of different traffic flow from converging en-route air traffic varied little, and the fractal dimensions of different traffic flow from runway air traffic varied a lot.At last, the influence of time scale to nonlinear characteristics of air traffic flow was investigated from the macro perspective. The geometric characteristics of air traffic flowvaried a lot under different time scales, but the chaotic characteristics were found in all air traffic time series. With the increase of time scales, the uncertainty of traffic flow decreased,and the randomness of traffic flow increased. Meanwhile the complexity of the traffic flow system decreased. A method to choose the time scale for air traffic flow time series, which based on nonlinear characteristics analysis was propose.