中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

重庆交通大学学报(自然科学版) ›› 2021, Vol. 40 ›› Issue (03): 1-6.DOI: 10.3969/j.issn.1674-0696.2021.03.01

• 交通+大数据人工智能 •    下一篇

离场航空器滑行时间预测研究

李楠1,焦庆宇1,张连东2,樊瑞1   

  1. (1. 中国民航大学 空中交通管理学院,天津 300300; 2. 南京航空航天大学 民航学院,江苏 南京 211106)
  • 收稿日期:2019-07-22 修回日期:2020-05-04 出版日期:2021-03-15 发布日期:2021-03-15
  • 作者简介:李楠(1978—),女,辽宁抚顺人,副教授,主要从事空中交通运行规划与仿真技术方面的研究。Email:lily_cauc@163.com
  • 基金资助:
    国家重点研发项目(2020YFB1600101);国家自然科学基金项目(U1833103,71801215);中国民航大学民航航班广域监视与安全管控技术重点实验室基金项目(202008)

Taxi Time Prediction of Departure Aircraft

LI Nan1, JIAO Qingyu1, ZHANG Liandong2, FAN Rui1   

  1. (1. College of Air Traffic Management, Civil Aviation University of China, Tianjin 300300, China; 2. College of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, Jiangsu, China)
  • Received:2019-07-22 Revised:2020-05-04 Online:2021-03-15 Published:2021-03-15
  • Supported by:
     

摘要: 为准确预测离港航班滑行时间,结合北京首都国际机场实际运行情况,分析航空器滑行距离、场面滑行航空器数量(进,离港)、跑道运行模式对航班滑行时间的影响;并运用DBSCAN算法按每小时航班流量对机场运行时间段进行分类;根据分类结果建立多元回归模型,分别采用传统统计学和机器学习(Lasso回归)预测航空器离场滑行时间。结果表明:与传统统计学多元线性回归模型相比,机器学习交叉训练集下模型的预测准确度较高,预测与实际误差值在5 min内的占87%。研究结果可用于大型机场实际运行航班滑行时间预测。

 

关键词: 交通运输工程, 航空运输, 滑行时间, 多元回归, 场面运行, 机器学习

Abstract: In order to accurately predict the taxi time of departing flights, combined with the actual operation condition of Beijing Capital International Airport, a multiple regression model was established according to the classification results obtained by using DBSCAN algorithm to classify the airport operation time according to the flight flow per hour. The impact of the taxiing distance of aircrafts, the number of taxiing aircrafts on the ground (departure and arrive), the runway operation mode on the taxiing time of flights was analyzed. And two methods, traditional statistics and machine learning (Lasso regression), were used to predict the taxiing time of departure aircraft. The results show that: compared with the traditional statistical multiple regression model, the prediction accuracy of the proposed model under machine learning crosstraining set is higher and 87% of errors between the predicted and actual values are within 5 minutes, which can be used to predict the actual flight taxiing time of large airports.

Key words: traffic and transportation engineering, air transportation, taxi time, multiple regression, surface movement, machine learning

中图分类号: