自然灾害下的山区道路网络关键节点识别与抗毁性分析  

Identification of Critical Nodes and Damage Resistance Analysis of Road Network in Mountainous Areas under Natural Disasters

作　　者：杨扬[1] 李广远 YANG Yang;LI Guangyuan(Faculty of Transportation Engineering,Kunming University of Science and Technology,Kunming 650500,China)

机构地区：[1]昆明理工大学交通工程学院,云南昆明650500

出　　处：《灾害学》2024年第3期87-93,124,共8页Journal of Catastrophology

基　　金：国家自然科学基金项目“跨境供应链协同行为与动态网络演化机理研究”(72264017)。

摘　　要：为预防自然灾害对山区道路节点结构和性能造成影响,对山区道路网络进行关键节点识别与抗毁性分析。考虑山区道路特性,将道路坡度比和道路等级纳入到关键节点识别因素中,提出考虑山区特性的L空间下山区道路无向加权网络构建方法,改进原有引力模型,考虑度中心值、介数中心值、节点核心位置和特征向量中心值的综合评估法,以网络最大连通度和全局效率作为网络抗毁性的评估指标,采用随机攻击和蓄意攻击两种策略模拟山区道路遭受灾害时的直接和间接影响。为避免灾害造成的级联失效影响,以网络节点破坏规模为评估指标,通过不同初始负载参数下关键阈值变化探究所需的最佳额外负载。选取昆明山区道路为研究对象,研究表明:改进的引力模型在山区网络关键节点评估中具有较高适用性,综合评估法提高了关键节点识别的全面性和准确度,所提出的级联失效模型能够有效界定初始负载参数的合理区间,最大限度地减少了山区网络预防灾害所需的附加负载。To prevent natural disasters from affecting the structure and performance of mountainous road nodes,key node identification and anti-destruction analysis are conducted on the mountain road network.Consid⁃ering the characteristics of mountainous roads,road gradient ratio and road grade are included in the factors of key node identification.A method for constructing an undirected weighted network of mountainous roads in L space con⁃sidering mountainous characteristics is proposed.The original gravity model is improved by considering the compre⁃hensive evaluation method of degree centrality value,betweenness centrality value,node core position,and char⁃acteristic vector centrality value.The maximum connectivity and global efficiency of the network are taken as the evaluation indicators of the network's anti-destruction ability.Random and deliberate attack strategies are used to simulate the direct and indirect impacts when mountain roads are affected by disasters.To avoid the cascading fail⁃ure effects caused by disasters,the network node destruction scale is used as the evaluation index to explore the optimal additional load required under different initial load parameters by varying the critical threshold.Based on the study of the Kunming mountainous area network,the results show that the improved gravity model is applicable for evaluating key nodes in the mountain network,and the comprehensive evaluation method enhances the compre⁃hensiveness and accuracy of key node identification.The proposed cascading failure model effectively defines the reasonable range of initial load parameters,minimizing the additional load required for preventing disasters in the mountainous network.

关 键 词：公路运输 山区道路安全 抗毁性仿真 山区道路网络 关键节点识别 级联失效 

分 类 号：X43[环境科学与工程—灾害防治] X915.5