基于级联BP神经网络的航班撤轮挡时刻预测  

FLIGHT OFF-BLOCK TIME PREDICTION BASED ON CASCADED BP NEURAL NETWORK

作　　者：徐涛[1,2,3] 丁杨 卢敏 Xu Tao;Ding Yang;Lu Min(College of Computer Science and Technology, Civil Aviation University of China, Tianjin 300300, China;Information Technology Research Base of Civil Aviation Administration of China, Tianjin 300300, China;Key Laboratory of Intelligent Passenger Service of Civil Aviation, CAAC, Beijing 101318, China)

机构地区：[1]中国民航大学计算机科学与技术学院,天津300300 [2]中国民航信息技术科研基地,天津300300 [3]民航旅客服务智能化应用技术重点实验室,北京101318

出　　处：《计算机应用与软件》2019年第6期226-232,共7页Computer Applications and Software

基　　金：国家自然科学基金项目(61502499);民航旅客服务智能化应用技术重点实验室项目;中央高校基本科研业务费科研专项(3122015Z007)

摘　　要：合理的航班协同离场前排序可以提高机场、航空公司、空管等部门的运行效率和可预测性,减少航班起飞前的等待时间。准确地预测航班撤轮挡时刻是建立航班起飞顺序的先决条件,对调整起飞前航班排序和计算航班起飞时间具有重要的决策意义。提出一个基于级联BP神经网络的航班撤轮挡时刻预测模型。该模型分别在航班过站过程的不同时刻进行航班撤轮挡时刻的预测,并进行过拟合研究。实验结果表明,与目前采用的经验统计预测模型相比,在相同时刻,该预测模型具有更高的预测准确率。A reasonable arrangement of pre-departure sequence of flights can improve the efficiency and predictability of airport,airline and blank pipe,and reduce the waiting time before the aircrafts take off.Accurate prediction of the flight off-block time is a prerequisite for the establishment of a pre-departure sequence,which has important decision significance for adjusting the flight departure order and calculating the flight departure time.This paper proposed a flight off-block time prediction model based on cascaded BP neural network.The model predicted the flight off-block time at different times of the flight turnaround process,and made over-fitting study.The experimental results show that compared with the empirical statistical prediction model currently used,the model has higher prediction accuracy at the same time.

关 键 词：航班撤轮挡时刻预测 BP神经网络 级联模型 里程碑事件 过拟合 协同决策 

分 类 号：TP339[自动化与计算机技术—计算机系统结构]