摘要
目前航班延误是民航亟待解决的问题,航班延误受多种因素的交叉影响,导致延误时间分布不规律。为了进一步提高航班延误预测精度,提出了基于ARIMA模型与BP神经网络的航班延误组合预测模型。模型考虑了航班延误线性和非线性因素,结合了ARIMA模型与BP神经网络的优势,分别从最优权重和残差优化角度对航班延误进行组合预测。选取某航空公司运行数据进行实例仿真验证,结果表明:组合预测模型的误差均小于单一预测模型,能够进一步提高航班延误的预测精度。
At present,flight delay is an urgent problem to be solved in civil aviation.Flight delay is affected by many factors,which leads to the irregular distribution of delay time.In order to further improve the prediction accuracy of flight delay,a combined prediction model of flight delay based on ARIMA model and BP neural network is proposed.The model takes linear and nonlinear factors of flight delay into consideration and combines the advantages of ARIMA model and BP neural network,which makes combined prediction of flight delay from the perspective of optimal weight and residual optimization respectively.Finally,a operating data of an airline company is selected for example simulation and verification.The results show that the errors of the combined prediction model are all smaller than that of the single prediction model,which can further improve the accuracy of flight delay prediction.
作者
罗军
李鹏飞
罗凤娥
LUO Jun;LI Peng-fei;LUO Feng-e(Civil Aviation Flight University of China,Guanghan 618000,China)
出处
《航空计算技术》
2021年第5期51-54,59,共5页
Aeronautical Computing Technique
基金
中国民用航空飞行学院科研创新团队建设计划项目资助(JG2019-35)。
关键词
航班延误
ARIMA模型
BP神经网络
组合预测
flight delay
ARIMA model
back propagation neural network
combination forecast
