基于K-shell的城际轨道网络关键节点识别研究  

Research on identification of key nodes in intercity rail networks based on K-shell

作　　者：刘星雨 刘杰 王喆[2] 黎浩东[3] LIU Xingyu;LIU Jie;WANG Zhe;LI Haodong(School of Transportation and Logistics Engineering,Shandong Jiaotong University,Jinan 250357,China;Jinan City Planning and Design Institute,Jinan 250101,China;School of Traffic and Transportation,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]山东交通学院交通与物流工程学院,济南250357 [2]济南市规划设计研究院,济南250101 [3]北京交通大学交通运输学院,北京100044

出　　处：《北京交通大学学报》2024年第4期181-190,共10页JOURNAL OF BEIJING JIAOTONG UNIVERSITY

基　　金：国家高端外国专家引进计划(G2022023003);山东省人文社会科学课题(2021-YYGL-15);山东省重点研发计划(软科学)重点项目(2023RZB06052)。

摘　　要：针对城际轨道交通网络中关键节点较难识别的问题,对网络节点分析方法进行研究.基于K-shell分层法,结合邻居节点的影响以及动静态网络指标,建立了关键节点识别的Ks^(+)法.该模型综合考虑节点度、节点最短路径等静态物理指标与枢纽客流量、运营强度等动态运营指标,计算节点的综合评价值.通过K-shell分层算法评估节点的全局核心位置,结合邻居节点的影响力评估节点的局部重要性,最终计算节点k_(s)^(+)值以衡量其在网络中的影响力.利用传染病(Susceptible-Infectious-Recvered,SIR)模型与长三角轨道网络数据为实例,检测了算法的有效性.研究结果表明:识别出的关键节点与城市影响力基本一致,如前4个节点均是长三角地区的直辖市节点和省会城市节点;准确识别了网络中的核心节点与非核心节点,并将线路更重要、客流更多的核心节点排在前列;Ks^(+)识别出的关键节点在SIR仿真中传播速率较其他算法快1~3次迭代,客流损失峰值增加7%.This study investigates methods for analyzing network nodes to address the challenge of identifying key nodes in intercity rail transit networks.A key node identification method,termed Ks~+,is developed.This method integrates the K-shell decomposition method with the influence of neighboring nodes and both dynamic and static network indicators.The model considers static physical indicators such as node degree and shortest path,as well as dynamic operational indicators like hub passenger flow and operational intensity,to compute a comprehensive evaluation value for each node.The k_s+ value of nodes,indicating their influence within the network,is determined by assessing the global core position through the K-shell decomposition algorithm and evaluating local importance with the influence of neighboring nodes.The effectiveness of this algorithm is demonstrated using the Susceptible-Infectious-Recvered(SIR) model and data from the Yangtze River Delta rail network.The results indicate that the identified key nodes closely correspond to city influence,with the top four nodes being direct-administered municipalities and provincial capitals of the Yangtze River Delta region.Furthermore,the algorithm accurately distinguishes core from non-core nodes,ranking nodes with more critical lines and higher passenger flows higher.Key nodes identified by Ks~+ exhibit a faster propagation rate in the SIR simulation by 1~3 iterations compared to other algorithms,with a peak passenger loss exceeding by 7%.

关 键 词：铁路运输 关键节点 Ks^(+)算法 复杂轨道网络 梯度提升机算法 

分 类 号：U292.18[交通运输工程—交通运输规划与管理]