Research On The Methods Of En Route Traffic Flow Management Based On Network Traffic Demand Prediction

Global air traffic has grown dramatically during the last few decades.Advanced and effective air traffic flow management(ATFM)in the en route airspace with increasing air traffic demandis thus important to support the development of a safe and efficient air transportation system,which is an urgent problem to be resolved by civil aviation industry.Motivated by this observation,this paper proposes a novel prediction method for traffic flow demand,based on which an en route traffic flow management method is studied with time granularities of both short/medium-term and short-time.The achievements of this paper can provide methods and technologies for the development of the safe,convenient,efficient and green civil aviation system.Based on a study of basic conceptions and main technologies and a literature survey of en route traffic flow management,considering the problems in the practice,the main topics of this thesis focus on short/medium-term prediction of en route traffic demand,key technologies in strategicen route flow management,short-time prediction of en route traffic demand as well as dynamic optimization of en route traffic flow,.The main content of this thesis is as follows:(1)A dynamic network for en route traffic flow is developed to be regarded as the basis of traffic flow characteristic analysis as well as traffic flow modeling.Adynamic network is constructed to characterize both the static topology of theairspace and the dynamics of the traffic flow in practice,which provides a mechanism to capturethe variation of the aircraft speed both in time and space in the airspace system of interest.(2)A short/medium-term prediction method for traffic demand in en route airspace is proposed to forecast the trends of air traffic flow between the city-pairs as well as in the en route areas.This thesis develops a dynamic linear model(DLM)of the state-space models and its improved model for en route traffic demand prediction.DLM models do not require the stationary assumption for the time series,which makes it advantageous to model air traffic demand sensitive to factors like heavy weather,important events,etc.Additionally,the uncertainty of the prediction results caused by the accumulated error as the forecast time horizon gets larger can be quantified by recursive computational method based on the Bayesian state estimation and forecasting theory,therefore providing the corresponding confidence interval of the prediction results.The case study shows that the accuracy and stability of theprediction results based on the proposed models are superior over those based on the existing ones.(3)This thesis studies the main methods and the key technologies of en route strategic flow management,and proposes a capacity demand prediction method as well as an aviation emissions estimation model.On the one side,a spatial-temporal prediction model for en route traffic demand is proposed,based on which a capacity demand prediction model is developed to quantify the trend of capacity demand of en route airspace.On the other side,a novel aviation emissions prediction model is proposed based on the en route traffic demand prediction results and the emissions estimations with the BFFM2,which is suitable for non-standard atmosphere conditions during aircraft cruise phase in the en route airspace.The model also incorporates the operational information extracted from the real radar data and the field flight plan and the emissions reduction efficiency due to the potential technological improvements for airframes and aviation engines,thus a more practical and more accurate prediction method for the distribution of en route emissions is achieved.The case study shows that the proposed method can well track the trend of capacity demand of en route sectors as well as the dynamics and uncertainty of emissions generated in the en route airspace.(4)A short-time traffic demand prediction method is proposed based on the developed dynamic air traffic flowmodel,which can be applied to forecast traffic demand in various airspace units,such as air route,en route sector,etc.The continuity equation in fluid mechanics is adopted to describe the continuous behavior of the en route traffic.Building on the network-based continuity equation,the space division concept in cell transmission model is introduced to discretize the proposed model both in space and time.On this basis,a dynamic en route traffic flow model is proposed to predict the traffic demand and its uncertainty interval for air routes as well as en route sectors.The model parameters are sequentially updated based on the statistical properties of the recent radar data and the new predicting results.The proposed method is applied to a real data set for the short-term air traffic flow prediction at both air route and en route sector level.The analysis of the case study shows that the developed method can characterize well the dynamics of the en route traffic flow,thereby providing satisfactory prediction results with accuracy and stabilitysuperior over those based on theexisting ones.(5)An optimization framework for en route traffic flow is proposed to address demand-capacity imbalances in en route airspace from the perspective of ATFM.This thesis presents a novel dynamic network air traffic flow model,based on which two efficient en route optimization methodsare proposed for air traffic on a particular route and in a network of such routes,respectively.This provides a more practical and flexible mechanism to optimally control air traffic flow at both air route and en route network levels.The proposed methods are applied to a real-work radar data set to enforce capacity constraints.The results show that our methods can effectively balance the demand and capacity of air traffic system at both air route and en route network levels.The optimization at the network level is more efficient to address the imbalanced demand and capacity,as well as to reduce flight time,operational costs and aviation emissions.This can be attributed to the system-wide air traffic flow modeling and optimization from a global perspective.