中国安全科学学报 ›› 2017, Vol. 27 ›› Issue (8): 132-137.doi: 10.16265/j.cnki.issn1003-3033.2017.08.023

• 安全工程技术科学 • 上一篇    下一篇

基于BA-WNN的滑行道安全风险预警方法

刘俊勇1, 高曙1,2 教授, 罗帆**3, 魏万淇1   

  1. 1 武汉理工大学 计算机科学与技术学院,湖北 武汉 430063;
    2 武汉理工大学 交通物联网技术湖北省重点实验室,湖北 武汉 430063;
    3 武汉理工大学 管理学院,湖北 武汉 430070
  • 收稿日期:2017-04-28 修回日期:2017-07-03 出版日期:2017-08-20 发布日期:2020-10-13
  • 通讯作者: **罗帆(1963—),女,湖南益阳人,博士,教授,博士生导师,主要从事安全风险管理、人力资源管理等方面的研究。E-mail:sailluof@126.com。
  • 作者简介:刘俊勇(1992—),男,湖北应城人,硕士研究生,主要研究方向为智能计算。E-mail: ljy120261693@163.com。
  • 基金资助:
    国家自然科学基金资助(71271163)。

Research on BA-WNN based safety risk early warning method of taxiway in airport

LIU Junyong1, GAO Shu1,2, LUO Fan3, WEI Wanqi1   

  1. 1 School of Computer Science and Technology, Wuhan University of Technology, Wuhan Hubei 430063, China;
    2 Hubei Key Laboratory of Transportation Internet of Things, Wuhan University of Technology,Wuhan Hubei 430063,China;
    3 School of Management, Wuhan University of Technology, Wuhan Hubei 430070, China
  • Received:2017-04-28 Revised:2017-07-03 Online:2017-08-20 Published:2020-10-13

摘要: 为更有效地实现具有复杂性、时变性及非线性的机场滑行道安全风险预警,降低事故发生率,针对小波神经网络(WNN)训练过程易陷入局部最优以及训练不稳定等影响预测准确性问题,采用蝙蝠算法(BA)优化WNN,设计和实现基于BA-WNN的滑行道安全风险预警方法,并将其与BP神经网络(BPNN)、WNN、遗传算法优化小波网络(GA-WNN)等3种方法进行有效性对比。结果表明:BA-WNN方法的预警准确率最高(约为84%),在所有工况下误警率都较低。

关键词: 风险预警, 预警指标, 滑行道, 蝙蝠算法(BA), 小波神经网络(WNN)

Abstract: For the sake of finding a more effective solution to safety risk early warning of taxiway in the airport, WNN was chosen as the main method for realizing the safety risk early warning of taxiway. Seeing that the training process of WNN is easy to fall into local optimum and the training is unstable, BA was used to optimize WNN. A BA-WNN based safety risk early warning method of taxiway in the airport was worked out. An effectiveness comparison was made between BPNN, WNN and GA-WNN and BA-WNN method. The results show that BA-WNN has the highest accuracy rate of 84%, and a low false alarm rate under all working conditions.

Key words: risk early warning, early warning indicators, taxiway, bat algorithm (BA), wavelet neural network (WNN)

中图分类号: