基于网格划分的城市交通常发拥堵识别方法  

Identification of Frequent Congestion Method in Urban Transportation Based on Grid Division

作　　者：冯强 李丽 赵训婷 李杨 云旭 FENG Qiang;LI Li;ZHAO Xunting;LI Yang;YUN Xu(Beijing Traffic Infrastructure Construction Project Management Center,Beijing 100166,China;BTI Smart Tech Co.,Ltd,Beijing 100073,China)

机构地区：[1]北京市交通基础设施建设项目管理中心,北京100166 [2]北京交研智慧科技有限公司,北京100073

出　　处：《交通工程》2023年第5期1-6,25,共7页Journal of Transportation Engineering

摘　　要：为解决常发性拥堵识别中合流点较多、存在环路导致识别困难、路段值缺失使拥堵难以溯源等问题,提出了基于网格模型的城市交通网络常发性拥堵识别方法,基于空间自相关理论,确定以3 km为识别拥堵扩散的显著范围;定义常发性拥堵路段识别规则,计算常发性拥堵路段在拥堵扩散范围内与邻近路段的时间相关系数,得到强相关路段对应的时序偏移量,最终以北京市朝阳区为例,识别常发性拥堵瓶颈37个,并以健翔桥为例证明方法的准确性.研究表明,本文提出的常发性拥堵路段和拥堵瓶颈识别方法,简化了复杂的路网拓扑关系,为拥堵传播机理的准确分析奠定了基础,结果能为制定疏堵方案提供理论依据和数据支撑.In order to solve the problems of frequent congestion identification,such as too many conjunctions,difficulty in identification due to loop,and difficulty in tracing congestion due to missing road segment value,this paper proposes a method of frequent congestion identification in urban traffic network based on grid model.Based on spatial autocorrelation theory,3 km is determined as the significant range of congestion diffusion identification.The identification rules of frequent congestion sections were defined,the time correlation coefficients between frequent congestion sections and adjacent sections within the congestion diffusion range were calculated,and the corresponding time sequence offsets of strongly correlated sections were obtained.Finally,37 frequent congestion bottlenecks were identified in Chaoyang District of Beijing,and the Jianxiang Bridge was taken as an example to prove the accuracy of the method.The research shows that the identification method of frequent congestion sections and congestion bottlenecks proposed in this paper simplifies the complex road network topology,lays a foundation for the accurate analysis of congestion propagation mechanism,and the results can provide theoretical basis and data support for the formulation of congestion schemes.

关 键 词：常发性拥堵瓶颈 拥堵扩散 浮动车数据 时空相关性 网格划分 

分 类 号：U491.4[交通运输工程—交通运输规划与管理]