摘要：

基于多元视角下的居民城际出行网络空间结构测度可以较为全面地刻画出城市间的复杂联系特征。论文利用春运期间腾讯迁徙平台中的人口流动数据,采用复杂网络分析方法以及转变中心性和转变控制力等指标,对比分析了航空、铁路和公路3种交通方式下的中国居民城际出行网络结构特征。结果表明：3种出行方式下,航空联系的线路最少,平均出行距离最长;铁路出行人数最多,其次是公路,航空最少。最大优势流中,北京和上海在国内航空出行联系中起着最重要的控制作用,其次是成都和重庆;铁路出行中,北京和成都占据绝对优势;公路多表现为省级行政中心与周边城市的关联。根据转变中心性和转变控制力划分城市类型,在航空和铁路出行网络中,高中心性—高控制力城市较多;公路出行网络中以高中心性—低控制力城市为主。不同出行方式下的城市聚类得到的网络集群“社区”数量有一定差异,航空、铁路和公路出行依次聚类为7、8和10个“社区”。不同类型出行方式透视的城市网络特征存在较明显的差异：航空出行的城际人口流体现出以全国性枢纽城市为核心分布的核心—边缘结构;铁路表现出以国家铁路大动脉沿线城市为核心,向腹地城市逐渐递减的核心—边缘结构;公路出行的城际人口流则表现为与人口规模匹配的局域强聚集的空间格局。

Abstract:

Spatial structure measurement of residents' intercity trip network based on multiple perspectives can be used to comprehensively describe the complex connection between cities. Based on the data of population flow from Tencent migration platform in 2018 during the Spring Festival in China and using the complex network analysis method and taking alter-based centrality and alter-based power as indicators, the structural characteristics of Chinese intercity trip network under the three traffic modes of aviation, railway, and highway were compared and analyzed. The results show that: 1) The number of routes connected by aviation is the smallest among the three modes of travel, and the average trip distance is the longest; the number of passengers by railway is the largest, followed by highway and aviation. 2) With regard to the maximum dominant flow, Beijing and Shanghai play the most important controlling role in the domestic aviation trip connection, followed by Chengdu and Chongqing. With regard to the railway trips, Beijing and Chengdu occupy absolute advantageous positions, and the highways mostly connect provincial administrative centers and the surrounding cities. 3) There are four types of cities based on the alter-based centrality and alter-based power: high alter-based centrality-high alter-based power cities, high alter-based centrality-low alter-based power cities, low alter-based centrality-high alter-based power cities, and low alter-based centrality-low alter-based power cities. The number of high alter-based centrality-high alter-based power cities is the largest in the aviation and railway travel network. The number of high alter-based centrality-low alter-based power cities is dominant in the highway travel network. 4) There are differences in the number of urban community detection clustering structures under different travel modes. Aviation, railway, and highway were clustered into 7, 8 and 10 urban communities respectively. Under the aviation travel mode, discontinuity between "communities" is clear. Under the mode of railway travel, agglomeration appears, and an obvious block distribution appears under the mode of highway travel. From the perspective of travel modes, the characteristics of urban network are also significantly different. The intercity flow of aviation travel showed a core-periphery structure with national hub cities as the core; train travel showed a core-periphery structure with the cities along the national railway artery as the core, and gradually decreasing to the hinterland cities; and the intercity flow of highway travel indicates the spatial pattern of local strong aggregation matching the population scale. The study of spatial structure of residents' intercity trip network under different travel modes can reveal the multiple spatial characteristics of population migration, residents' trip, and urban network from different perspectives, complement the results of existing studies based on single travel modes, and enrich the regional understanding of spatial relationships of Chinese cities.

Key words: intercity trip network, urban network structure, population flow, Spring Festival travel rush, Tencent migration platform, China

不同出行方式下的人口流动集散层级排名前10位的城市和流动线路"

不同出行方式下转变中心性(AC)、转变控制力(AP)前10名和后10名城市"