《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (10): 3283-3291.DOI: 10.11772/j.issn.1001-9081.2022010002

• 前沿与综合应用 • 上一篇    

基于深度全连接神经网络的离港航班延误预测模型

徐海文1, 史家财2, 汪腾2   

  1. 1.中国民用航空飞行学院 理学院,四川 广汉 618307
    2.中国民用航空飞行学院 民航安全工程学院,四川 广汉 618307
  • 收稿日期:2022-01-06 修回日期:2022-04-25 接受日期:2022-04-27 发布日期:2022-10-14 出版日期:2022-10-10
  • 通讯作者: 徐海文
  • 作者简介:第一联系人:徐海文(1978—),男,山东菏泽人,教授,博士,主要研究方向:优化理论与算法、交通运输规划与管理; xuhaiwen_dream@163.com
    史家财(1996—),男,山东济南人,硕士研究生,主要研究方向:安全科学、交通运输规划与管理
    汪腾(1998—),男,广东江门人,硕士研究生,主要研究方向:安全科学、交通运输规划与管理。
  • 基金资助:
    中央高校基本科研业务费专项资金资助项目(J2021-057)

Departure flight delay prediction model based on deep fully connected neural network

Haiwen XU1, Jiacai SHI2, Teng WANG2   

  1. 1.School of Science,Civil Aviation Flight University of China,Guanghan Sichuan 618307,China
    2.College of Civil Aviation Safety Engineering,Civil Aviation Flight University of China,Guanghan Sichuan 618307,China
  • Received:2022-01-06 Revised:2022-04-25 Accepted:2022-04-27 Online:2022-10-14 Published:2022-10-10
  • Contact: Haiwen XU
  • About author:XU Haiwen, born in 1978, Ph. D. , professor. His research interests include optimization theory and algorithms, transportation planning and management.
    SHI Jiacai, born in 1996, M. S. candidate. His research interests include safety science, transportation planning and management.
    WANG Teng, born in 1998, M. S. candidate. His research interests include safety science, transportation planning and management.
  • Supported by:
    Fundamental Research Funds for Central Universities(J2021-057)

摘要:

针对提升离港航班延误预测精确度困难的问题,提出一种基于深度全连接神经网络(DFCNN)的离港航班延误预测模型。首先,在考虑航班信息、机场气象与航班延误历史的基础上,考虑航班网络结构对预测模型的影响;然后,从激活函数、输入数据项及延误时间阈值三个维度进行实验,以对模型抑制梯度弥散与提升学习表现能力的能力进行了优化与验证;最后,通过调控神经网络层数的纵向拓展方式与随机丢失层的Dropout参数,提升模型的泛化能力。实验结果表明:所提模型使用tanh、指数线性函数(ELU),预测精确度比使用线性整流函数(ReLU)分别提升了1.26、1.28个百分点;考虑航班网络结构后,所提模型采用ELU函数计算时,预测精确度比未考虑航班网络结构时提升了3.12个百分点;在时间阈值为60 min时,通过调控Dropout参数,模型的损失值不断降低;在5层隐含层网络和Dropout参数为0.3时,所提模型可以取得92.39%的预测精确度。因此,所提模型能够对国内航班延误做出较为准确的判断。

关键词: 深度学习, 航班延误预测, 航班网络, 数据融合, 模型参数

Abstract:

Aiming at the problem that it is difficult to improve the accuracy of departure flight delay prediction, a departure flight delay prediction model based on Deep Fully Connected Neural Network (DFCNN) was proposed. Firstly, on the basis of considering flight information, airport weather and flight delay history, the influence of flight network structure on prediction model was considered. Secondly, experiments were carried out from three dimensions of activation function, input data item and delay time threshold to optimize and verify the model ability to suppress gradient dispersion and improve the learning performance. Finally, through adjusting the vertical expansion method of the number of neural network layers and the Dropout parameters of the random loss layers, the generalization ability of the model was improved. The results of experiments indicate that the prediction accuracy of the proposed model can be improved by 1.26 percentage points and 1.28 percentage points respectively after using tanh and Exponential Linear Unit (ELU) functions in the proposed model than using Rectified Linear Unit (ReLU). After considering the flight network structure, the prediction accuracy calculated by the proposed model using ELU function is improved by 3.12 percentage points than without considering the flight network structure. When the Dropout parameters are adjusted, the loss value of the model is continuously reduced with 60 min time threshold. With a 5-layer hidden layer network and a Dropout parameter of 0.3, the prediction accuracy of 92.39% can be achieved by the proposed model. Therefore, the proposed model can make more accurate judgments on domestic flight delays.

Key words: deep learning, flight delay prediction, flight network, data fusion, model parameter

中图分类号: