摘要
为降低飞机冲出跑道事故造成的人员伤亡和财产损失,以长着陆事件为研究对象,探讨1种有效的长着陆事件预警方法。提出应用随机森林模型对长着陆事件进行预警,给出模型的特征筛选、构建及评价方法,并使用实际飞行数据验证模型有效性。结果表明:基于随机森林的飞机长着陆预警模型精确率为88.41%,召回率为87.14%,模型可以有效对长着陆事件进行预警。
In order to reduce the casualties and property losses caused by the runway overrun accident of aircraft,taking the long landing events as the research object,an effective early-warning method of long landing events was discussed.It was put forward to apply the random forest model to carry out the early-warning of long landing events,then the feature selection,construction and evaluation methods of the model were given.The effectiveness of the early-warning method of long landing events based on the random forest model was verified by the actual flight data.The results showed that the accuracy rate of the model was 87.14%,and the recall rate was 88.41%.The model can effectively warn the long landing events.
作者
孙瑞山
李重锋
SUN Ruishan;LI Chongfeng(Research Institute of Civil Aviation Safety,Civil Aviation University of China,Tianjin 300300,China;Economics and Management College,Civil Aviation University of China,Tianjin 300300,China)
出处
《中国安全生产科学技术》
CAS
CSCD
北大核心
2021年第5期182-186,共5页
Journal of Safety Science and Technology
关键词
飞行安全
飞行数据
长着陆
随机森林
预警方法
flight safety
flight data
long landing
random forest
early-warning method
