The Traffic Capacity Of Airspace Based On Dynamic Network Flow Method

In recent years,CAAC has paid great attention to "traffic capacity".Its core problem is how to allocate the available level,so as to meet the traffic demand and optimize the allocation of airspace resources.Due to the complex and changeable weather conditions,the influence of military aviation activities and other factors,the traffic capacity often can not reach the assessed capacity.Therefore,it is necessary to study the optimization of traffic capacity under the dangerous weather and military aviation activities,so that every airspace user can use the airspace fairly and effectively,avoid the waste and waste of airspace resources,and alleviate the airspace congestion.The difference and connection between traffic capacity and capacity is the basis of this thesis.Firstly,this thesis analyzes the concepts of capacity.Secondly,from the two aspects of normal operation and emergency costs,this thesis defines the cost function to be used in enroute network flow allocation.Finally,according to the characteristics of the route network and its level,the mathematical description of the network and level is given,and the multi-commodity flow theory is used to describe the aircraft type.In view of the shortcomings of the traditional operational research model and the road traffic flow allocation model applied to the aviation system,two models are improved respectively.In the aspect of model building,multiple edges between nodes are used to describe the level,and variable cost function is used to replace the constant cost.In addition,a single flow is extended to multiple flows,and route capacity constraints and controller load constraints are added.In the aspect of model solving algorithm,firstly,the algorithm of minimum cost and maximum flow problem is improved,and an approximate algorithm is designed to distribute the flow in stages,which makes the flow distribution result more close to the actual operation;secondly,the Hungarian algorithm is improved from the aspects of iteration mode and efficiency matrix transformation,which increases the accuracy of the result and reduces the time complexity of the operation.Thirdly,the concepts of connectivity and tolerance are introduced to improve the connectivity and reroute of dial algorithm.In the aspect of example verification,the two models and their solution algorithms are verified respectively.In view of the difference between airspace system and its elements on capacity,this thesis discusses it from three aspects.Based on the maximum flow of route network,a key route identification algorithm is designed to measure the impact of different route blocking on capacity.In terms of the impact of system reliability changes on the capacity,the probability of dangerous weather in airspace is transformed into the reliability of enroute network,so as to measure the impact of weather changes on the capacity of airspace.In terms of the influence of the division of regional management center on the traffic capacity,taking the airspace planning of East China as an example,the traffic capacity changes under different division of regional management are calculated.In view of the problem that dangerous weather and military aviation activities have a great impact on the control operation,the dynamic network flow method is used to study the short-term air traffic flow managemnt problem under emergencies.Firstly,according to the characteristics of network congestion varying with time and flow,the characteristics of random change of dangerous weather,the characteristics of difference of air waiting and ground waiting costs,three optimization objectives are constructed,and a multi-objective optimization model is established considering the constraints of airport capacity,sector capacity,flight continuity and sector continuity.Thirdly,according to the characteristics of coordination of airspace in military aviation activities,the airspace is required to be coordinated Because of the high time complexity of solving the multi-commodity flow model,it can not adapt to the defects of short-term flow management.Finally,taking the airspace controlled by Southwest Air Traffic Control Bureau as an example,three scenarios are designed for simulation using the actual traffic data.The results show that the proposed model and algorithm can effectively solve the short-term traffic management problem in emergencies,and the efficiency of the algorithm is better than the traditional algorithm in large traffic.To sum up,this thesis studies three topics most related to airspace capacity by improving dynamic network flow method,which provides a theoretical solution to alleviate the problems of "high risk in operation,high load in work and high pressure in support" of air traffic control,gives the calculation method of capacity under normal operation and emergency,and expands the application scope of dynamic network flow method.