Stochastic Air Traffic Flow Management for Demand and Capacity Balancing Under Capacity Uncertainty

This paper introduces an innovative approach to the synchronized demand-capacity balance with special focus on sector capacity uncertainty within a centrally controlled collaborative air traffic flow management(ATFM)framework.Further with previous study,the uncertainty in capacity is considered as a non-negligible issue regarding multiple reasons,like the impact of weather,the strike of air traffic controllers(ATCOs),the military use of airspace and the spatiotemporal distribution of nonscheduled flights,etc.These recessive factors affect the outcome of traffic flow optimization.In this research,the focus is placed on the impact of sector capacity uncertainty on demand and capacity balancing(DCB)optimization and ATFM,and multiple options,such as delay assignment and rerouting,are intended for regulating the traffic flow.A scenario optimization method for sector capacity in the presence of uncertainties is used to find the approximately optimal solution.The results show that the proposed approach can achieve better demand and capacity balancing and determine perfect integer solutions to ATFM problems,solving large-scale instances(24 h on seven capacity scenarios,with 6255 flights and 8949 trajectories)in 5-15 min.To the best of our knowledge,our experiment is the first to tackle large-scale instances of stochastic ATFM problems within the collaborative ATFM framework.

Terminal Traffic Flow Prediction Method Under Convective Weather Using Deep Learning Approaches

In order to improve the accuracy and stability of terminal traffic flow prediction in convective weather,a multi-input deep learning(MICL)model is proposed.On the basis of previous studies,this paper expands the set of weather characteristics affecting the traffic flow in the terminal area,including weather forecast data and Meteorological Report of Aerodrome Conditions(METAR)data.The terminal airspace is divided into smaller areas based on function and the weather severity index(WSI)characteristics extracted from weather forecast data are established to better quantify the impact of weather.MICL model preserves the advantages of the convolution neural network(CNN)and the long short-term memory(LSTM)model,and adopts two channels to input WSI and METAR information,respectively,which can fully reflect the temporal and spatial distribution characteristics of weather in the terminal area.Multi-scene experiments are designed based on the real historical data of Guangzhou Terminal Area operating in typical convective weather.The results show that the MICL model has excellent performance in mean squared error(MSE),root MSE(RMSE),mean absolute error(MAE)and other performance indicators compared with the existing machine learning models or deep learning models,such as Knearest neighbor(KNN),support vector regression(SVR),CNN and LSTM.In the forecast period ranging from 30 min to 6 h,the MICL model has the best prediction accuracy and stability.

Review on artificial intelligence techniques for improving representative air traffic management capability

The use of artificial intelligence(AI)has increased since the middle of the 20th century,as evidenced by its applications to a wide range of engineering and science problems.Air traffic management(ATM)is becoming increasingly automated and autonomous,making it lucrative for AI applications.This paper presents a systematic review of studies that employ AI techniques for improving ATM capability.A brief account of the history,structure,and advantages of these methods is provided,followed by the description of their applications to several representative ATM tasks,such as air traffic services(ATS),airspace management(AM),air traffic flow management(ATFM),and flight operations(FO).The major contribution of the current review is the professional survey of the AI application to ATM alongside with the description of their specific advantages:(i)these methods provide alternative approaches to conventional physical modeling techniques,(ii)these methods do not require knowing relevant internal system parameters,(iii)these methods are computationally more efficient,and(iv)these methods offer compact solutions to multivariable problems.In addition,this review offers a fresh outlook on future research.One is providing a clear rationale for the model type and structure selection for a given ATM mission.Another is to understand what makes a specific architecture or algorithm effective for a given ATM mission.These are among the most important issues that will continue to attract the attention of the AI research community and ATM work teams in the future.

Assessing Air Traffic Management Performance Interdependencies through Bayesian Networks： Preliminary Applications and Results

The performance model proposed by this study represents an innovative approach to deal with performance assessment in ATM （air traffic management）. It is based on Bayesian networks methodology, which presents several advantages but also some drawbacks as highlighted along the paper. We illustrate the main steps required for building the model and present a number of interesting results. The contribution of the paper is two-fold： （1） It presents a new methodological approach to deal with a problem which is of strategic importance for ANSPs （air navigation service providers）; （2） It provides insights on the interdependencies between factors influencing performance. Both results are considered particularly important nowadays, due to the SES （Single European Sky） performance scheme and its related target setting process.

OPTIMAL FLOW CONTROL PROBLEM IN AIR TRAFFIC CONTROL:PARTⅠ

The fundamental case is considered in which flights from many destinations must be scheduled for arrival at a single congested airport having limited capacities.An air traffic control(ATC)model is developed in this case.A new and efficient algorithm for the optimal solution of ground holding strategy problem(GHSP)is put forward and verified by a numerical example.

Recognition of Similar Weather Scenarios in Terminal Area Based on Contrastive Learning

In order to improve the recognition accuracy of similar weather scenarios(SWSs)in terminal area,a recognition model for SWS based on contrastive learning(SWS-CL)is proposed.Firstly,a data augmentation method is designed to improve the number and quality of weather scenarios samples according to the characteristics of convective weather images.Secondly,in the pre-trained recognition model of SWS-CL,a loss function is formulated to minimize the distance between the anchor and positive samples,and maximize the distance between the anchor and the negative samples in the latent space.Finally,the pre-trained SWS-CL model is fine-tuned with labeled samples to improve the recognition accuracy of SWS.The comparative experiments on the weather images of Guangzhou terminal area show that the proposed data augmentation method can effectively improve the quality of weather image dataset,and the proposed SWS-CL model can achieve satisfactory recognition accuracy.It is also verified that the fine-tuned SWS-CL model has obvious advantages in datasets with sparse labels.

Analysis on Potential Conflict Frequency of Intersected Air Routes in Terminal Airspace Design

In order to obtain accurate conflict risks in terminal airspace design,the concept and calculation model of potential conflict frequency for intersected routes are proposed.Conflict frequency is represented by the product of horizontal conflict frequency and vertical conflict probability.The horizontal conflict frequency is derived from the probability density distribution of conflicts in a period of time.Based on the recorded radar trajectory data,the concept and model of ROUTE distance are proposed,and the probability density function of aircraft height at a specified ROUTE distance is deduced by kernel density estimation.Furthermore,vertical conflict probability and its horizontal distribution are achieved.Examples of three intersected arrival and departure route design schemes are studied.Compared with scheme 1,the conflict frequency values of the other two improved schemes decrease to53% and 24%,respectively.The results show that the model can quantify potential conflict frequency of intersected routes.

基于AirTOp的终端区扇区负荷校准方法研究

管制员工作负荷的评估是扇区运行的一个重要论题,它对提高机场终端区容量值,进而增加航班量和经济收益具有重要意义。结合北京终端区的管制员实际工作负荷评估和校准工作,提出了一种基于AirTOp仿真软件的终端区进近扇区负荷校准的新思路和方法。通过对比同等运行环境下的实际管制负荷与仿真输出,校准管制负荷权重设置,获得一个可用于仿真管制员负荷评估的负荷权值汇总表,使用交叉验证方法将校准后的权值作为输入进行面向扇区的计算机仿真。结果显示仿真负荷与实际负荷偏差值小于±2,证明该校准方法可以适用于对复杂终端区扇区负荷的评估。

AUCTIONING METHOD FOR AIRSPACE CONGESTING RESOURCE ALLOCATION AND GAME EQUILIBRIUM ANALYSIS

The airspace congestion is becoming more and more severe.Although there are traffic flow management（TFM）initiatives based on CDM widely applied,how to reschedule these disrupted flights of different airlines integrating TFM initiatives and allocate the limited airspace resources to these airlines equitably and efficiently is still a problem.The air traffic management（ATM）authority aims to minimizing the systemic costs of congested airspaces.And the airlines are self-interested and profit-oriented.Being incorporated into the collaborative decision making（CDM）process,the airlines can influence the rescheduling decisions to profit themselves.The airlines maybe hide the flight information that is disadvantageous to them,but is necessary to the optimal system decision.To realize the coincidence goal between the ATM authority and airlines for the efficient,and equitable allocation of airspace resources,this paper provides an auction-based market method to solve the congestion airspace problem under the pre-tactic and tactic stage of air traffic flow management.Through a simulation experiment,the rationing results show that the auction method can decrease the total delay costs of flights in the congested airspace compared with both the first schedule first service（FSFS）tactic and the ration by schedule（RBS）tactic.Finally,the analysis results indicate that if reallocate the charges from the auction to the airlines according to the proportion of their disrupted flights,the auction mechanism can allocate the airspace resource in economy equitably and decrease the delay losses of the airlines compared with the results of the FSFS tactic.

基于ARMA-AE-LSTM模型的进场交通流预测方法

为建立准确有效的空中交通短期流量预测模型,提高终端区管理效率,以进场交通流为对象进行研究。首先采用自回归移动平均(autoregressive moving average,ARMA)模型对流量时间序列进行初步线性预测,然后通过长短期记忆网络(long short term memory,LSTM)模型对线性预测后的残差序列进行非线性修正预测。考虑到冗余特征会降低LSTM模型预测精度的问题,采用自编码器(autoencoder,AE)模型对LSTM模型的天气以及流量特征输入进行自适应压缩优化,最后设置对比实验对ARMA-AE-LSTM模型的准确性、鲁棒性以及时效性进行验证。实验结果表明:预测绝对误差在1.3架以内的占比达到75%;LSTM模型的平均每轮迭代时间降低为1.014 s;与其他常用深度学习预测模型相比,ARMA-AE-LSTM模型的均方根误差(root mean square error,RMSE)、平均绝对误差(mean absolute error,MAE)以及决定系数(r-squared,R2)评价指标分别改善了45.98%~67.66%、48.56%~67.35%、5.18%~21.07%;恶劣天气影响下,ARMA-AE-LSTM模型的鲁棒性更好。由此可见,该方法能够准确有效快速的预测空中交通流量。

机场航班时刻资源管理研究进展

繁忙机场都会面临严重的交通拥堵和航班延误问题。解决该问题的一个有效手段是对机场航班时刻资源进行优化管理。全面介绍了航班时刻资源管理相关的研究进展,系统回顾了机场公布容量设置、机场航班时刻配置的主要方法和手段。从公布容量确定、单机场、机场网络层面和机场群航班时刻配置及航班时刻配置技术复杂性等方面,得出机场航班时刻资源优化配置的关键技术问题。展望了机场航班时刻资源管理的主要方向,并对下一步研究提出建议。

终端区交通复杂性评估及演变分析

终端区内飞行密集、环境多变,已成为国家空域系统运行能力提升的重要瓶颈。精准量化终端区内交通复杂性并把握其演化规律是实现空中交通精细化管理的基础,对于提升终端区运行能力有重要意义。终端区空中交通复杂性描述了某一时间点或时间段内空中交通流特征的复杂程度。基于动态密度思想,针对终端区运行特征,构建了多维度的空中交通复杂性评价指标,并运用熵权法确定指标权重,反映评估指标的无序化程度,评定信息量的大小。构建复杂度计算模型,从客观上利用多指标反映航空器实时复杂性演变。采用多智能体仿真建模工具NetLogo搭建了终端区仿真环境,以广州终端区的实际进场数据搭建了基础场景,模拟了终端区交通复杂性的宏观演化过程。通过调节进场移交点交通流需求比例、设计不同天气情景,分析了不同场景下交通复杂性演化特征。结果表明,该评估体系使用较少的指标和计算时间,能够较为准确、客观地评估时间片内的复杂性,符合实际情况。NetLogo建模工具能够仿真终端区交通流,直观地观察交通流运行行为。当进场移交点交通流流量分布更为均匀时,复杂性较低。终端区受恶劣天气影响时,其受影响程度取决于恶劣天气覆盖航段的航班流量分布情况、影响的范围以及受影响航段到跑道距离。部分交通流存在迟滞现象,反映了复杂性的蔓延时空特征。该复杂性评估方法能够有效评估航段、航线、终端区复杂性及演变规律,为开展终端区精细化空中交通管理提供参考。

TBO模式下终端区容量评估与优化

随着航空运输业的快速发展,航班量逐渐增加,造成了终端区可用空域资源饱和、空中交通拥堵、航班延误等现象。基于未来空中交通管理模式向着基于航迹运行(TBO)模式转变的趋势,考虑了机场、管制部门、航空公司这三个容量管理参与方,即在不同的协同决策下,建立终端区容量优化模型。运用基于滑动时间窗的优化-反馈算法和元启发式模拟退火算法,并以广州终端区作为实验案例进行验证。研究结果表明,模型在不同的协同决策下可以有效提高空域资源的利用率,对于增加终端区的运行容量具有一定效果。

基于任务优先权的军民航飞行活动协同排序模型

随着空域资源需求的不断增大,军民航间飞行矛盾日益突显。为解决此问题,本文以国务院、中央军事委员会空中交通管制委员会提出的“军民航空管联合运行”为背景,引入军民航共享空域的概念,重点研究了在此类空域中军民航飞行活动协同排序(CMFCS,civil-military aviation flight activity collaborative sequencing)问题。首先,基于军民航各自飞行任务特点与差异,对军民航飞行任务的种类进行划分,并使用层次分析法确定各类飞行任务的优先权原则;其次,以军民航飞行活动总延误时间成本最小为目标,建立CMFCS模型;最后,使用遗传算法对模型进行求解,确定军民航飞行活动批准进入共享空域的时间序列。研究结果表明,与经典的先到先服务(FCFS,first come first service)策略相比,协同排序策略得到的总延误时间成本降低了72.17%,优化效果显著且更符合实际,能够实现军民航共同使用国家空域资源,保障飞行活动安全、有序、高效地运行。

基于二次分解集成的机场流量短期预测

为实现准确的机场流量短期预测,本文建立了基于二次分解方法的分解集成预测模型。首先,应用局部加权回归周期趋势分解(STL,seasonal and trend decomposition procedure based on Loess)算法将原始时间序列分解为趋势项、季节项和余项3个分量,并计算其样本熵。其次,应用遗传算法(GA,genetic algorithm)优化变分模态分解(VMD,variational mode decomposition)参数,对熵值较大的分量进行二次分解。再次,使用极端梯度提升(XGBoost,extreme gradient boosting)对二次分解后的所有分量进行预测,采用加和集成得到最终的预测值。最后,采集国内典型机场实际运行数据进行实例分析。针对北京首都国际机场60 min进场、离场流量时序,本文模型预测的均等系数(EC,equal coefficient)值分别为0.9703、0.9959,相比其他常用模型均有所提高。此外,对于上海浦东、上海虹桥、广州白云3个大型国际机场,本文模型在60 min、30 min统计尺度下进场和离场流量预测的EC值均在0.9700以上,15 min统计尺度下预测的EC值均在0.9500以上。结果表明,本文建立的二次分解集成预测模型具有良好的准确性和普适性,用于机场流量短期预测是可行和有效的。

基于GWO-HMM的空中交通网络流系统态势预测研究

针对空中交通流量管理部门如何更高效地实施流量管理的问题,本文将态势感知理论应用于空中交通网络流系统(ATNFS,air traffic network flow system),建立空中交通网络流系统的运行态势预测模型。首先,给出了空中交通网络流系统的态势感知过程,从节点和航线的角度筛选出航线饱和度、不正常航班率、节点饱和度、节点延误架次比、节点航班取消率5个态势要素,使用态势值作为态势理解的指标;其次,分析隐马尔可夫模型(HMM,hidden Markov model)的优势与不足,建立了基于灰狼优化(GWO,grey wolf optimization)算法和改进隐马尔可夫模型的态势预测模型;最后,使用某空中交通网络流系统的实际运行数据进行算例验证。结果表明,改进后的预测模型相较于原本的隐马尔可夫预测模型精度更高,预测结果更准确。

增强鲁棒性的点融合系统进场排序与调度

现有进场航班排序与调度模型较少考虑不确定因素的影响,本文基于点融合系统(PMS,point merge system)连续下降进近时间符合高斯分布的特点,建立增强鲁棒性的两阶段进场排序与调度模型。第一阶段,描述随机变量带来的不确定性,基于机会约束规划理论设计带有额外缓冲区的静态排序与调度模型,以飞行时间和次序变更量为指标确定最佳额外缓冲区;第二阶段,补偿不可预知原因导致飞行时间超出缓冲区的情况,采用基于滑动时间窗的启发式算法对第一阶段的方案进行动态调整。通过Monte-Carlo模拟不确定性场景下的连续下降进近时间,仿真结果表明,在高、中、低3种交通密度下,所提出的两阶段模型相较于其他模型在鲁棒性和公平性的平衡方面表现出了良好性能。

A data-driven approach to modeling high-density terminal areas：A scenario analysis of the new Beijing,China airspace

Airports are being developed and expanded rapidly in China to accommodate and promote a growing aviation market.The future Beijing Daxing International Airport（DAX） will serve as the central airport of the JingJinJi megaregion,knitting the Beijing,Tianjin,and Hebei regions together.DAX will be a busy airport from its inception,relieving congestion and accommodating growth from Beijing Capital International Airport（PEK）,currently the second busiest airport in the world in passengers moved.We aim to model terminal airspace designs and possible conflicts in the future Beijing Multi-Airport System（MAS）.We investigate standard arrival procedures and mathematically model current and future arrival trajectories into PEK and DAX by collecting large quantities of publicly available track data from historical arrivals operating within the Beijing terminal airspace.We find that（1） trajectory models constructed from real data capture aberrations and deviations from standard arrival procedures,validating the need to incorporate data on historical trajectories with standard procedures when evaluating the airspace and（2） given all existing constraints,DAX may be restricted to using north and east arrival flows,constraining the capacity required to handle the increases in air traffic demand to Beijing.The results indicate that the terminal airspace above Beijing,and the future JingJinJi region,requires careful consideration if the full capacity benefits of the two major airports are to be realized.

基于VOSviewer和CiteSpace的空中交通流量管理研究综述

以中国知网(CNKI)和Web of Science (WOS)数据库内的文献作为数据来源,以2013—2023年为检索年限,将CNKI中以“空中交通流量管理”为主题的956篇文献和WOS中以“air traffic flow management”为主题的379篇文献作为研究对象,使用VOSviewer和CiteSpace文献计量工具生成了作者合作网络图,研究机构共现图、关键词共现图以及关键词突变图,以此来了解并分析有关空中交通流量管理的研究内容。结果表明,对空中交通流量管理的研究关注度较高,发文量总体呈现上升趋势,研究机构大多集中在高校。关键词研究显示,空中交通流量管理研究的方向包括流量管理、深度学习等方面。结论为空中交通流量管理研究提供理论依据。

基于集成学习的航空器滑出时间预测研究

为提升航空器离场滑出时间的预测精度,分析了影响滑出时间的各类因素,引入场面运行状况和气象条件两类特征,基于装袋方法、随机森林、自适应增强和梯度提升等四种典型的集成学习方法,构建了滑行时间预测模型。以美国肯尼迪机场为算例,采用判决系数、RMSE和MAE等性能度量指标验证算法预测性能。实验结果表明:气象特征的引入能够提升滑出时间预测精度;与其它回归算法对比,集成学习的预测误差较小;分析集成方法下的学习曲线发现自适应增强和梯度提升方法能够有效避免过拟合现象。研究结果可用于集成化场面管理软件的开发应用。