基于ADS-B时空数据挖掘的大终端区空域特征分析及其改善策略研究
Spatial Characteristics Analysis and Improvement Strategy of Large Terminal Area Based on ADS-B Spatio-Temporal Data Mining
摘要: 由于航班总量的迅速增加,国内每个空域都存在着日益严重的空域交通拥堵问题。空域热点包含大量能有效反映空中交通拥堵的信息。因此,空中交通管理领域必须研究如何从海量航线数据中定位空域热点和热门航线。本文主要阐述了目前中国国内关于航空飞行器运动轨迹聚类分析算法的主要研究状况;介绍了广播式自动相关监视系统(ADS-B, automatic dependent surveillance broadcast);对航迹数据采用剔除、过滤和模糊等航迹预处理方法;对预处理过后的航迹数据采用基于密度的DBSCAN聚类,对特征空域采用热点密度分类,并对热点拥堵地区进行识别。通过使用BMAP API提取航空地图(Aeronautical Chart)在python平台建立华北空域航路地图;利用DBSCAN对华北空域一天当中的航迹数据进行基于密度的聚类;将聚类结果写入空域程序进行显示得出华北机场群空域热点。研究成果表明该程序可显示华北机场群终端区空域的热点拥堵区域和冷点空闲区,依据给出的空域结构和航班时刻,可给出改善大终端区运行效率的方法。
Abstract:
Due to the rapid increase of the total number of flights, every airspace in China has an increasingly serious problem of airspace traffic congestion. Airspace hot spots contain a lot of information that can effectively reflect air traffic congestion. Therefore, the field of air traffic management must study how to locate airspace hot spots and hot routes from massive route data. In this paper, the main research status of the cluster analysis method of aircraft trajectory in China is described. This paper introduces the automatic dependent surveillance broadcast system (ADS-B). The track data are processed by eliminating, filtering and blurring. DBSCAN clustering based on density was used for the pre-processed track data, and hotspot density classification was used for the characteristic airspace, and hotspot congestion areas were identified. Based on the Python platform, By using BMAP API to extract Aeronautical Chart, the route map of North China airspace was established on Python platform. DBSCAN is used to cluster the daily track data based on density in north China airspace. The clustering results are written into the airspace program to display the airspace hot spots of North China airport group. The results show that the program can solve the hot spot congestion area and cold spot idle area in the terminal area of North China Airport group, and provide a method to improve the operation efficiency of large terminal area according to the given airspace structure and flight time.
参考文献
|
[1]
|
徐涛, 李永祥, 吕宗平. 基于航迹点法向距离的航迹聚类研究[J]. 系统工程与电子技术, 2015, 37(9): 2198-2204.
|
|
[2]
|
吴欣蓬, 汤新民, 毛继志, 郭鸿滨, 管祥民. 基于密度聚类与匹配算法的异常飞行行为挖掘[J]. 南京航空航天大学学报, 2021, 53(6): 863-871.
|
|
[3]
|
黄天镜. 基于多维特征的ADS-B数据异常检测方法研究[D]: [硕士学位论文]. 天津: 中国民航大学, 2020.
|
|
[4]
|
马兰, 高永胜. 基于ADS-B数据挖掘的4D航迹预测方法[J]. 中国民航大学学报, 2019, 37(4): 1-4.
|
|
[5]
|
刘继新, 董欣放, 徐晨, 杨光, 江灏. 基于密度峰值的终端区航迹聚类与异常识别[J]. 交通运输工程学报, 2021, 21(5): 214-226.
|
|
[6]
|
Zeng, W.L., Xu, Z.F., Cai Z.P., Chu, X. and Lu, X.B. (2021) Aircraft Trajectory Clustering in Terminal Airspace Based on Deep Autoencoder and Gaussian Mixture Model. Aerospace, 8, Article No. 266. [Google Scholar] [CrossRef]
|
|
[7]
|
Olive, X. and Morio, J. (2019) Trajectory Clustering of Air Traffic Flows around Airports. Aerospace Science and Technology, 84, 776-781. [Google Scholar] [CrossRef]
|
|
[8]
|
Popa, I.S., Zeitouni, K., et al. (2015) Spatio-Temporal Compression of Trajectories in Road Networks. GeoInformatica, 19, No. 1. [Google Scholar] [CrossRef]
|
|
[9]
|
朱剑明, 余朋. 基于百度地图API的全国机场天气警报系统设计与实现[J]. 气象科技, 2017, 45(2): 247-253
|
|
[10]
|
Chaffer, J. jQuery基础教程[M]. 北京: 人民邮电出版社, 2008.
|