中国民航大学学报 ›› 2025, Vol. 43 ›› Issue (5): 37-43.

• 民机安全性与适航 • 上一篇    下一篇

飞机冲偏出跑道事故统计分析及后果预测研究


  

  1. 中国民航大学交通科学与工程学院,天津 300300
  • 收稿日期:2023-05-15 修回日期:2023-06-20 出版日期:2025-11-17 发布日期:2025-11-17
  • 作者简介:蔡靖(1975— ),女,河北唐山人,教授,博士,研究方向为机场跑道结构和安全运行理论
  • 基金资助:
    中央高校基本科研业务费专项(3122019103)

Statistical analysis and consequence prediction of aircraft runway excursion accidents

#br#   

  1. College of Transportation Science and Engineering, CAUC, Tianjin 300300, China
  • Received:2023-05-15 Revised:2023-06-20 Online:2025-11-17 Published:2025-11-17

摘要:

为了研究飞机冲偏出跑道事故发生及后果的主要特征,本文统计分析了 2010—2019 年间全球范围内发生的
100 起冲偏出跑道事故,从事故类型、事故致因、事故时空分布、事故机型及事故后果等方面深入剖析了飞机
冲偏出跑道事故的规律性;并基于反向传播(BP,back propagation)神经网络构建了飞机冲偏出跑道事故后果
预测模型,以预测不同条件下事故后果的严重程度。 结果表明:偏出跑道事故为飞机冲偏出跑道事故的主要
方面,湿滑污染道面及侧风是事故主要致因,飞机冲偏出跑道事故具有较强的时空分布规律性,且不同影响因
素会造成不同程度的后果损失。 本文所构建的预测模型对飞机冲偏出跑道事故后果的预测准确率达到 76.67%,
且测试集与训练集准确率达到同一水平,说明该模型预测效果好,可为飞机着陆安全管理提供预警依据。

关键词:

Abstract:

In order to study the main characteristics of the occurrence and consequences of aircraft runway excursion accidents, 100 global aircraft runway excursion accidents from 2010 to 2019 were statistically analyzed in this paper,
and the regularity of aircraft runway excursion accident was deeply analyzed from the aspects of accident types,
accident causes, accident spatiotemporal distribution, aircraft types involved in accidents, and accident consequences. Based on back propagation (BP) neural network, a prediction model of the consequences of aircraft
runway excursion accidents was constructed to predict the severity of accident consequences under different
conditions. The results showed that veering off runway accidents were the main aspect of aircraft runway excursion accidents, wet, slippery and polluted pavement and crosswind were the main causes of accidents. Aircraft
runway excursion accident had strong spatiotemporal distribution regularity, and different influencing factors can
cause varying degrees of consequences and losses. The prediction model constructed in this article achieved a
prediction accuracy of 76.67% for the consequences of aircraft runway excursion accidents, and the accuracy of
the test set and training set reach the same level, indicating that the model has good predictive performance and
can provide early-warning basis for aircraft landing safety management.

Key words:

中图分类号: