基于贝叶斯结构学习的航班保障Petri 网构建

taptap下载安装安卓学报 ›› 2020, Vol. 38 ›› Issue (3): 28-33.

基于贝叶斯结构学习的航班保障Petri 网构建

邢志伟¹，陈哲¹，夏欢²，罗谦²，丛婉²，陈丰华³

（1. taptap下载安装安卓电子信息与自动化学院，天津300300；2. 中国民用航空局第二研究所机场运行与控制工程技术研究中心，成都610041；3. 广州白云国际机场股份有限公司信息科技部，广州510410）

出版日期:2020-06-26 发布日期:2020-06-22
基金资助:
国家自然科学基金项目（U1533203）；民航安全能力建设资金项目（PSDSA201801，PSDSA201802）；成都市战略性新型产品研发补贴项目
（2015-CP01-00158-GX）；成都市产业集群协同创新项目（2016-XT00-00015-GX）

Flight support Petri net construction based on Bayesian network structure learning

XING Zhiwei¹, CHEN Zhe¹, XIA Huan², LUO Qian², CONG Wan², CHEN Fenghua³

（1. College of Electronic Information and Automation, CAUC, Tianjin 300300, China; 2. Second Research Institute, CAAC,Chengdu 610041, China; 3. Guangzhou Baiyun International Airport Incorporated, Guangzhou 510410, China）

Online:2020-06-26 Published:2020-06-22
About author:邢志伟（1970—），男，辽宁沈阳人，教授，博士，研究方向为民航装备与系统、民航智能规划与调度、机场交通信息与控制.
Supported by:

摘要/Abstract

摘要： 高峰时段受机场资源和时间窗口限制，航班保障实际过程较计划流程会产生一定偏差。将贝叶斯结构学习K2 算法应用到航班保障业务场景，提出航班保障的贝叶斯网络结构学习模型；结合贝叶斯网与Petri 网，形成贝叶斯无环Petri 网，以精准刻画航空器在保障流程中的状态切换；最后利用国内某枢纽机场的实际数据，构建航班保障网络结构。实验分析表明，相比计划流程，从历史数据得到的航班保障网络更符合实际流程。

关键词: Petri 网, 航班保障, 贝叶斯结构学习, K2 算法

Abstract: Due to the limitation of airport resource and time window during peak hours, the actual process of flight support will be biased compared to the planned process. K2 algorithm of Bayesian network structure learning is employed into flight support working scenarios, proposing the Bayesian network structure learning model of flight support.Bayesian network and Petri net are combined to build a Bayesian acyclic Petri net in order to accurately describe the status alternation of aircraft during flight support procedures. Finally, the actual data of one domestic hub airport are used to construct flight support network structure. Experimental analyses show that the flight support network obtained from historical data can better fit the actual working flow.

Key words: Petri net, flight support, Bayesian network structure learning, K2 algorithm

中图分类号:

V352
TP391.9

邢志伟, 陈哲, 夏欢, 罗谦, 丛婉, 陈丰华. 基于贝叶斯结构学习的航班保障Petri 网构建[J]. taptap下载安装安卓学报, 2020, 38(3): 28-33.

XING Zhiwei, CHEN Zhe, XIA Huan, LUO Qian, CONG Wan, CHEN Fenghua. Flight support Petri net construction based on Bayesian network structure learning[J]. Journal of Civil Aviation University of China, 2020, 38(3): 28-33.

参考文献 10

[1]	VIDOSAVJEVIC A, TOSIC V. Modeling of turnaround process using Petri nets[C]//Proceedings of the World Conference of Air Transport Research Society, Porto, 2010: 1-13.
[2]	TANG FEI, LIU SHUAN. Research of aircraft ground service scheduling problems[C]//Proceedings of the Chinese Control and Decision Conference,Yinchuan, Ningxia, May 28-30, 2016: 3405-3410.
[3]	LIU CHANGYOU, QIN YING. Modeling the aircraft ground operation on airport based on timed petri net[C]//Proceedings of the 6th World Congress on Intelligent Control and Automation, Dalian, Liaoning, June
21	-23, 2006: 1304-1308.
[4]	游婷婷, 朱新平. 基于Petri 网的过站航班机坪保障作业过程建模[J].中国民航飞行学院学报, 2016, 27(2): 10-14.
[5]	邢志伟, 魏志强. 基于着色Petri 网的航班保障服务建模方法[J]. 系统工程与电子技术, 2018, 40(5): 1064-1069.
[6]	IOANNIS T, LAURA E B, CONSTANTIN F A. The max-min hillclimbing Bayesian network structure learning algorithm[J]. Machine Learning, 2006,65(1)：31-78.
[7]	TERENT’YEV A N, BIDYUK P I. Method of probabilistic inference from learning data in Bayesian networks[J]. Cybernetics and Systems Analysis, 2007, 43(3): 391-396.
[8]	COOPER G F, HERSKOVITS E. A Bayesian method for the induction of probabilistic networks from data[J]. Machine Learning, 1992, 9(4):309-347.
[9]	吴哲辉. Petri 网导论[M]. 北京: 机械工业出版社, 2006.
[10]	ROBINSON R W. Counting unlabeled acyclic digraphs[M]//Combinatiorial Mathematics, 1977: 28-43.