基于改进支持向量机的空域交通态势识别方法

朱承元; 张澈; 管建华

doi:10.3963/j.jssn.1674-4861.2023.02.008

基于改进支持向量机的空域交通态势识别方法

doi: 10.3963/j.jssn.1674-4861.2023.02.008

中国民航大学空中交通管理学院天津 300300

基金项目:

国家自然科学基金项目 U1833103

国家自然科学基金项目 62173332

工信部民用飞机专项 MJ-2020-S-03

详细信息

通讯作者:
朱承元（1965—），博士，副教授. 研究方向：空域规划与仿真. E-mail：cyzhu@cauc.edu.cn

中图分类号: V355.1
计量
- 文章访问数: 778
- HTML全文浏览量: 345
- PDF下载量: 29
- 被引次数: 0
出版历程
- 收稿日期: 2022-03-07
- 网络出版日期: 2023-06-19

A Method for Monitoring Traffic State in the Airspace Based on an Improved Support Vector Machine

College of Air Traffic Management, Civil Aviation University of China, Tianjin 300300, China

摘要

摘要: 为了准确分析空域交通态势，针对当前空域交通态势感知中管制行为因素难以度量的问题，从多种管制员工作负荷的角度进行量化分析。借助全空域及机场模型软件（Total Airspace and Airport Modeller，TAAM）建立空域仿真模型，研究了栅格化空域场景下基于改进支持向量机（support vector machine，SVM）的空域交通态势识别方法。通过对比不同的栅格化方案，结合管制员实际运行经验确定栅格形状和尺寸，以边长为25 km的六边形为最小单元，对目标空域进行栅格化处理。在引入多种管制负荷的基础上同时考虑导航设施的分布情况，建立空域交通态势识别指标体系，运用K-means聚类算法对降维后的仿真样本数据进行聚类分析，获取先验分类数据。在支持向量机模型的基础上，引入麻雀搜索算法（sparrow search algorithm，SSA），构建基于麻雀搜索算法优化支持向量机（SSA-SVM）的空域交通态势识别模型。依据适应度对解集进行划分，对模型关键参数核函数参数σ和惩罚系数C进行优化，确定了1组泛化能力强同时避免过拟合问题的参数组合，并将栅格化空域交通态势划分为4个等级。以西安区域管制区为对象开展仿真实验，结果表明：与基于遗传算法优化支持向量机（GA-SVM）的空域交通态势识别模型相比，SSA-SVM模型克服了GA-SVM模型确定的的过拟合问题，平均分类识别准确率提高2.50%，最佳分类识别准确率提高1.73%；在176个栅格中，拥堵态、拥挤态和平稳态栅格个数分别为26、18和51，模型识别结果与基于管制员经验划分的复杂空域相比，覆盖率可达95%，验证了提出方法对空域交通态势识别及降低管制员工工作负荷的有效性。
- 空中交通管理 /
- 栅格化 /
- 交通态势识别 /
- TAAM /
- 麻雀搜索算法 /
- 支持向量机
Abstract: This paper quantitatively analyzes the methods for monitoring airspace traffic state from the perspective of the workload of air traffic controllers, due to the difficult measurement of such a factor in current studies.In response to the need of monitoring airspace traffic state more efficiently, an airspace simulation model is developedbased onTotal Airspace and Airport Modeller (TAAM) software and a method for identifying traffic state in the rasterized airspace scenario is proposed based on an improved support vector machine (SVM). Based on real-world operation experience of controllers, the shape and size of grids are determined by comparing different rasterization schemes. Target airspace is rasterized by taking the hexagon with a side length of 25 km as the smallest unit. Considering a variety of controller's workloads and the distribution of navigation facilities, a set of indicators for describing traffic states of the airspace is developed.Ak-means clustering algorithm is used to generate prior classified data by aggregating simulated sample data. A traffic state model for the airspace, developed based on the sparrow-search algorithm (SSA) and SVM, iscalled SSA-SVM.The solution set is divided according to the fitness. Moreover, key parameters of the model, including kernel parameters σ and penalty coefficients C, are optimized to determine a combination of parameters, which can increase the generalization capability of the model and avoid overfitting. Traffic states in the rasterized airspace are divided into four levels. Simulations are conductedfor the control airspace ofthe City of Xi'an. Study results show that the proposed SSA-SVM model can mitigate the overfitting problem, but not by the proposed genetic algorithm and support vector machine (GA-SVM) model.The average accuracy of classification is improved by 2.50%, and the accuracy of classification is improved by 1.73%. Among the tested 176 grids, the number of congested, crowded, and steady grids are 26, 18, and 51, respectively. Compared with the partition method for the complex airspace based on controller experience, the convergencerate of the proposed model is as high as 95%, which verifies the effectiveness of the proposed method for identifying airspace traffic state and reducing the workload of air traffic controllers.
- air traffic management /
- rasterized airspace scenario /
- traffic situation recognition /
- TAAM /
- sparrow search algorithm /
- support vector machine

HTML全文

图 1 不同形状的SEL对比

Figure 1. Comparison of SEL in different shapes

下载: 全尺寸图片幻灯片

图 2 SEL生成过程

Figure 2. SEL generation process

下载: 全尺寸图片幻灯片

图 3 参数C和σ的编码方式

Figure 3. Encoding method for parameters of C and σ

下载: 全尺寸图片幻灯片

图 4 栅格化空域交通态势识别流程

Figure 4. Rasterized airspace traffic situation identification process

下载: 全尺寸图片幻灯片

图 5 目标空域SEL尺寸

Figure 5. SELsize of target airspace

下载: 全尺寸图片幻灯片

图 6 西安区域管制空域栅格化及冲突点表征

Figure 6. Rasterization and conflict point characterization of Xi'an area control airspace

下载: 全尺寸图片幻灯片

图 7 目标空域栅格化仿真模型

Figure 7. Rasterization simulation model of target airspace

下载: 全尺寸图片幻灯片

图 8 模型指标数据

Figure 8. Index data of model

下载: 全尺寸图片幻灯片

图 9 基于K-means算法的样本数据

Figure 9. Sample data based on K-means algorithm

下载: 全尺寸图片幻灯片

图 10 适应度变化曲线对比

Figure 10. Comparison of fitness curve

下载: 全尺寸图片幻灯片

图 11 模型测试集样本预测分类与实际分类对比

Figure 11. Comparison between predicted and actual classification of model test set samples

下载: 全尺寸图片幻灯片

图 12 栅格化空域态势识别结果与基于管制经验的拥堵空域对比

Figure 12. Comparison between rasterized airspace situation recognition results and congested airspace based on control experience

下载: 全尺寸图片幻灯片

表 1 指标载荷系数

Table 1. Load factor of each index

主成分	态势评估指标的重要程度
主成分	V₁	V₂	V₃	V₄	V₅	V₆	V₇
Y₁	0.764	0.614	0.191	0.016	0.035	0.045	0.062
Y₂	0.154	0.088	0.079	0.029	0.219	0.110	0.949

下载: 导出CSV

表 2 模型分类结果比较

Table 2. Comparison of model classification results

识别模型	参数组合	样本数	准确点数
GA-SVM	C =63.783 5	120	115
GA-SVM	σ =0.292 3	120	115
SSA-SVM	C =30.099 4	120	119
SSA-SVM	σ =3.364 7	120	119

下载: 导出CSV

参考文献(21)

[1]	MENON P K, SWERIDUK G D, BILIMORIA K D. A new approach for modeling, analysis and control of air traffic flow[J]. AIAA Journal of Guidance, Control and Dynamics, 2004, 27 (5): 737-744. doi: 10.2514/1.2556
[2]	程承旗, 吴飞龙, 王嵘, 等. 地球空间参考网格系统建设初探[J]. 北京大学学报(自然科学版), 2016, 52 (6): 1041-1049. https://www.cnki.com.cn/Article/CJFDTOTAL-BJDZ201606009.htm CHENG C Q, WU F L, WANG R, et al. Study on globe spatial grid reference system construction[J]. Journal of Peking University (Natural Science Edition), 2016, 52 (6) : 1041-1049. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-BJDZ201606009.htm
[3]	徐鑫宇, 万路军, 陈平, 等. 基于GeoSOT网格的空域栅格化表征方法[J]. 空军工程大学学报(自然科学版), 2021, 22(2): 15-22. https://www.cnki.com.cn/Article/CJFDTOTAL-KJGC202102003.htm XU X Y, WAN L J, CHEN P, et al. An airspace raster repre sentation method based on GeoSOT grid[J]. Journal of Air Force Engineering University(Natural Science Edition), 2021, 22 (2): 15-22. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-KJGC202102003.htm
[4]	张兆宁, 黎新华, 王莉莉. 基于飞行跟驰模型的纵向安全间隔计算方法[J]. 交通运输工程学报, 2008 (3): 73-76. https://www.cnki.com.cn/Article/CJFDTOTAL-JYGC200803019.htm ZHANG Z N, LI X H, WANG L L. Computational method of longitudinal safety separation based on flight following theory[J]. Journal of Traffic and Transportation Engineering, 2008(3): 73-76. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JYGC200803019.htm
[5]	王灵丽, 黄敏, 高亮. 基于聚类算法的交通网络节点重要性评价方法研究[J]. 交通信息与安全, 2020, 38 (2): 80-88. doi: 10.3963/j.jssn.1674-4861.2020.02.010 WANG L L, HUANG M, GAO L. Methods of importance evaluation of traffic network node based on clustering algorithms[J]. Journal of Transport Information and Safety, 2020, 38 (2): 80-88. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2020.02.010
[6]	LEE K, FERON E, PRITCHETT A. Describing airspace complexity: Airspace response to disturbances[J]. Journal of Guidance, Control and Dynamics, 2009, 32 (1): 210-222. doi: 10.2514/1.36308
[7]	徐肖豪, 任杰, 李楠. 基于FCM的终端区交通态势识别[J]. 航空计算技术, 2014, 44 (1): 1-8. https://www.cnki.com.cn/Article/CJFDTOTAL-HKJJ201401001.htm XU X H, REN J, LI N. Identification of terminal area traffic situation based on FCM[J]. Aeronautical Computing Technique, 2014, 44 (1): 1-8. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HKJJ201401001.htm
[8]	李楠, 任杰, 徐肖豪. 终端区交通态势识别研究[J]. 科学技术与工程, 2014, 14 (11): 256-261. https://www.cnki.com.cn/Article/CJFDTOTAL-KXJS201411057.htm LI N, REN J, XU X H. Identification of terminal area traffic situation[J]. Science Technology and Engineering, 2014, 14(11): 256-261. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-KXJS201411057.htm
[9]	赵嶷飞, 吕立萱, 张勰. 空域扇区运行状态评估[J]. 科学技术与工程, 2014, 14 (32): 105-109. https://www.cnki.com.cn/Article/CJFDTOTAL-KXJS201432023.htm ZHAO Y F, LYU L X, ZHANG X. Status evaluation of airspace sector operation[J]. Science Technology and Engineering, 2014, 14 (32): 105-109. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-KXJS201432023.htm
[10]	李桂毅. 基于航迹数据的航路网络交通运行态势识别与预测技术研究[D]. 南京: 南京航空航天大学, 2018. LI G Y. Research on traffic operation status identification and prediction technology of air route network based on flight trajectory data[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2018. (in Chinese)
[11]	王利航. 终端区交通态势研究[D]. 天津: 中国民航大学, 2020. WANG L H. Research on traffic situation in terminal control area[D]. Tianjin: Civil Aviation University of China, 2020. (in Chinese)
[12]	杜婧涵, 胡明华, 张魏宁, 等. 基于度量学习的机场交通态势弱监督评估[J/OL]. 北京航空航天大学学报: (1-10)[2022-01-13]. https://kns. cnki. net/kcms/detail/11.2625. V. 20211222. 0902. 001. html
[13]	岳仁田, 赵嶷飞, 罗云. 空中交通拥挤判别指标的建立与应用[J]. 中国民航大学学报, 2008 (3): 30-35. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGMH200803008.htm YUE R T, ZHAO Y F, LUO Y. Construction and application of congestion discrimination index for air traffic[J]. Journal of Civil Aviation University of China, 2008(3): 30-35. (inChinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGMH200803008.htm
[14]	袁立罡, 胡明华, 张洪海, 等. 融合先验经验聚类的终端区交通流相态识别[J]. 交通运输工程学报, 2016, 16(5): 83-94. https://www.cnki.com.cn/Article/CJFDTOTAL-JYGC201605010.htm YUAN L G, HU M H, ZHANG H H, et al. Phase-state identification of traffic flow in terminal area incorporated with prior experience clustering[J]. Journal of Traffic and Transportation Engineering, 2016, 16 (5): 83-94. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JYGC201605010.htm
[15]	李桂毅, 胡明华, 张洪海. 基于FCM-SVM方法的时空航路网交通状态识别研究[J]. 武汉理工大学学报(交通科学与工程版), 2018, 42 (4): 569-573, 578. https://www.cnki.com.cn/Article/CJFDTOTAL-JTKJ201804008.htm LI G Y, HU M H, ZHANG H H. Research on traffic state identification of spatio-temporal air route network based on FCM-SVM method[J]. Journal of Wuhan University of Technology(Transportation science&Engineering), 2018, 42(4): 569-573, 578. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JTKJ201804008.htm
[16]	李桂毅, 胡明华, 郑哲. 基于FCM-粗糙集的多扇区交通拥挤识别方法研究[J]. 交通运输系统工程与信息, 2017, 17(6): 141-146. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201706021.htm LI G Y, HU M H, ZHENG Z. Multi-sector traffic congestion identification method based on FCM-rough sets[J]. Journal of Transportation Systems Engineering and Information Technology, 2017, 17 (6): 141-146. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201706021.htm
[17]	YOUSEFI A, DONOHUE G. Temporal and spatial distribution of airspace complexity for air traffic controller workload-based sectorization[C]. AIAA 4th Aviation Technology, Integration and Operation Forum. Chicago, USA: AIAA, 2004.
[18]	宁亚美. 基于复杂度指标的终端管制区划分研究[D]. 天津: 中国民航大学, 2020. NING Y M. Research on the division of terminal control area based on complexity index[D]. Tianjin: Civil Aviation University of China, 2020. (in Chinese)
[19]	YIN Z, HOU J. Recent advances on SVM based fault diagnosis and process monitoring in complicated industrial processes[J]. Neurocomputing, 2016, 174 (9): 643-650.
[20]	李巧茹, 郝恩强, 陈亮, 等. 遗传算法优化支持向量机的城市交通状态识别[J]. 重庆交通大学学报(自然科学版), 2020, 39 (8): 1-5, 13. https://www.cnki.com.cn/Article/CJFDTOTAL-CQJT202008001.htm LI Q R, HAO E Q, CHEN L, et al. Urban traffic state recognition based on genetic algorithm optimized support vector machine[J]. Journal of Chongqing Jiaotong University(Natural Science Edition), 2020, 39 (8): 1-5, 13. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-CQJT202008001.htm
[21]	薛建凯. 1种新型的群智能优化技术的研究与应用[D]. 上海: 东华大学, 2020. XUE J K. Research and application of a noval swarm intelligence optimization technique: sparrow search algorithm[D]. Shanghai: Donghua University, 2020. (in Chinese)