基于混沌理论的区域航路网络交通状态预测  

Traffic Status Prediction Method of Regional Air Route Network Based on Chaos Theory

作　　者：李桂毅[1] 郭铭宇 张洪海[1] 罗一帆 LI Gui-yi;GUO Ming-yu;ZHANG Hong-hai;LUO Yi-fan(College of Civil Aviation/College of Flight,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China;College of Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China)

机构地区：[1]南京航空航天大学民航/飞行学院 [2]南京航空航天大学航空学院,江苏南京211106

出　　处：《航空计算技术》2020年第1期61-66,共6页Aeronautical Computing Technique

基　　金：中央高校基本科研业务经费专项资金项目资助(NJ20140016);国家级大学生创新创业训练计划项目资助(2019CX00713);南京航空航天大学2018年度实验技术与开发项目资助(2018N03).

摘　　要：研究区域航路网络交通状态预测问题,可为航路网络系统的规划管理与交通综合管控提供重要支持。基于区域航路网络中航空器ADS B数据计算路网交通量时间序列;对区域航路网络交通量时间序列进行相空间重构,判定区域航路网络交通量时间序列的混沌特性;分别构建基于RBF神经网络和Volterra级数的航路网络交通量混沌预测模型,预测区域航路网络交通量变化趋势;基于k均值聚类算法预测识别区域航路网络交通运行状态等级,最后进行实验验证。研究结果表明:区域航路网络交通量时间序列具有混沌特性,Volterra级数混沌预测模型预测精度优于RBF神经网络模型,k均值聚类算法可较好实现区域航路网络交通运行状态预测识别,提出的交通状态预测方法可为航路网络规划管理以及拥挤管控提供技术支持。This paper studies the traffic status prediction of regional air route network,which aims at providing important support for integrated planning and traffic control of regional air route network system.Firstly,based on aircraft ADS B data in regional air route network,a traffic volume time series of regional air route network is calculated.Secondly,the phase space reconstruction of the traffic volume time series of regional air route network is carried out,and the chaotic characteristics of the traffic volume time series of regional air route network are determined.Thirdly,the chaotic prediction model of traffic volume based on RBF neural network and Volterra series is constructed to predict the trend of regional air route network traffic volume,the traffic status of regional air route network is predicted and identified based on k means clustering algorithm.Finally,experimental verification is carried out.The results demonstrate that the time series of regional air route network traffic volume has chaotic characteristics.The prediction performance of Volterra series chaotic prediction model is better than RBF neural network model.k means clustering algorithm can better realize the prediction and recognition of regional air route network traffic status.The traffic status prediction method proposed in this paper can provide technical support for route network planning,management and congestion control.

关 键 词：航空运输 交通量预测 混沌时间序列预测 区域航路网络 交通状态预测 

分 类 号：V355[航空宇航科学与技术—人机与环境工程] U268.2[机械工程—车辆工程]