基于门架数据的高速公路拥堵识别研究  

Research on Expressway Congestion Discrimination Based on Toll Gantry Data

作　　者：余畅 苏跃江 韦清波 YU Chang;SU Yuejiang;WEI Qingbo(Guangzhou Transport Research Institute Co.Ltd.,Guangzhou 510635,China;School of Civil Engineering and Transportation,South China University of Technology,Guangzhou 510614,China;Guangzhou Public Transport Research Institute Co.Ltd,Guangzhou 510620,China)

机构地区：[1]广州市交通运输研究院有限公司,广州510635 [2]华南理工大学土木与交通学院,广州510614 [3]广州市公共交通数据管理中心有限公司,广州510620

出　　处：《科技和产业》2023年第22期113-118,共6页Science Technology and Industry

基　　金：广州市科技计划项目(202206010056)。

摘　　要：有效识别高速公路拥堵情况,能为高速公路的交通管控及建设规划提供有效依据。基于高速公路门架数据,提出了一套基于行程时间阈值的拥堵识别分析方法,提取车辆的行程时间集合,利用聚类算法对路段的不同拥堵等级的行程时间阈值进行标定,从而对高速公路拥堵情况进行判别。在实例分析中,以广州市高速公路网为研究对象,开展了拥堵识别分析。结果表明,广州市高速公路总体运行状态良好,80%的时段路段处于顺畅状态。It can provide effective evidence for expressway traffic control and construction planning by effectively identifying congestion of expressway.A methodology of congestion discrimination was proposed based on travel time threshold.The travel time collections of all vehicles were extracted based on toll gantry data.The travel time collection was clustered and analyzed by clustering algorithm,and the travel time threshold of different congestion levels was calibrated according to the clustering results.The congestion condition will be discriminated based on the congestion travel time thresholds.The results indicates that the overall operation of Guangzhou expressways is good,with 80%of periods and sections being in a smooth state.

关 键 词：智能交通 门架数据 聚类算法 行程时间阈值 拥堵识别 

分 类 号：U491[交通运输工程—交通运输规划与管理]