交通运输系统工程与信息

交通运输系统工程与信息 ›› 2021, Vol. 21 ›› Issue (4): 30-39.DOI: 10.16097/j.cnki.1009-6744.2021.04.004

• 综合交通运输体系论坛 • 上一篇    下一篇

基于顺风车数据和聚类方法的都市圈 区域划分与层级结构研究

闫学东1，郭浩楠1，李永昌2，王云*1，官云林1   

  1. 1. 北京交通大学，综合交通运输大数据行业重点实验室，北京 100044； 2. 山东省交通规划设计院集团有限公司，济南 250031
  • 收稿日期:2021-03-27 修回日期:2021-05-22 接受日期:2021-06-04 出版日期:2021-08-25 发布日期:2021-08-23
  • 作者简介:闫学东(1975- )，男，河北沧州人，教授，博士。
  • 基金资助:
    国家自然科学基金重大研究计划 ；国家自然科学基金创新研究群体科学基金

Regional Division and Hierarchical Structure of Metropolitan Area Based on Carpooling Data and Clustering Method

YAN Xue-dong1 , GUO Hao-nan1 , LI Yong-chang2 , WANG Yun*1 , GUAN Yun-lin1   

  1. 1. Key Laboratory of Transport Industry of Big Data Application Technologies for Comprehensive Transport, Beijing Jiaotong University, Beijing 100044, China; 2. Shandong Provincial Communications Planning and Design Institute Group Co. LTD, Jinan 250031, China
  • Received:2021-03-27 Revised:2021-05-22 Accepted:2021-06-04 Online:2021-08-25 Published:2021-08-23
  • Supported by:
    Major Research Plan of the National Natural Science Foundation of China(91746201)；Science Fund for Creative Research Groups of the National Natural Science Foundation of China(71621001)

PDF

摘要： 都市圈已经逐渐成为国家新型城镇化发展的主体形态之一，在区域经济一体化建设中起 着十分重要的作用。本文基于顺风车数据，使用聚类分析方法，围绕北京都市圈区域划分与层级 结构展开相关研究。首先，通过网格模型将研究区域网格化处理并作为基本处理单元，匹配获取 的顺风车数据与POI数据到网格中，利用基于网格的改进K-means++聚类算法，并结合使用手肘 法与轮廓系数法确定最佳聚类数量，对北京都市圈主要功能区进行划分。通过分析不同功能区 域内的居民通勤出行特征，提出通勤强度、通勤时间、功能区独立性、功能区可达性等区域通勤特 征评价指标，结合上述指标使用层次聚类方法对北京都市圈层级结构划分展开进一步研究。研 究结果表明：本文采用的改进聚类方法能克服传统聚类算法随机选取聚类数目所带来的影响，有 效划分并得到19类北京都市圈主要功能区域，聚类效果更佳；聚类结果显示北京都市圈主要功能 区域与北京市现有行政区域划分存在差异性，在都市圈规划建设当中应当主动破除行政区域壁 垒，实施面向都市圈范围的整体规划；根据不同功能区域的居民通勤特征与地理区位特征，北京 都市圈还可进一步划分为核心层、近郊层、远郊层3个圈层；应当根据北京都市圈圈层特性与功能 区自身属性制定相应发展策略，通过规划建设市郊铁路或轨道交通改善不同圈层间的通勤现状， 提高北京都市圈整体通勤可达性。研究结果为制定相应规划与管理政策提供依据，有利于都市 圈功能与结构进一步完善，促进都市圈良性发展。

关键词: 城市交通, 都市圈区域划分, 聚类算法, 都市圈, 顺风车数据

Abstract: Metropolitan area has gradually become one of the main forms of our country's new-type urbanization, and it plays a particularly important role in promoting the development of regional economic integration. Based on the carpooling data, this paper uses the cluster analysis method for the regional definition and hierarchical structure division of Beijing metropolitan area. First, we divided the research area into grids as the basic processing unit, and matched the carpooling data and the Point of Interest (POI) data to the grids. Then we combined the sum of the squarederrors and silhouette coefficient to determine the optimal number of clusters, and used the grid- based K- means ++ clustering algorithm to identify the main functional areas of the Beijing metropolitan area. By analyzing the commuting characteristics in different functional areas, the evaluating indicators such as commuting intensity, commuting time, regional independence, and regional accessibility are proposed, which can be used to study the hierarchical structure of Beijing metropolitan area by using the hierarchical clustering method. The results show that the method in this paper can overcome the influence of the random selection of the number of clusters in the traditional clustering algorithm, and can effectively divide and obtain 19 types of main functional areas of the Beijing metropolitan area, which had better clustering effect. The clustering results of functional areas are different from the existing divisions of administrative areas in Beijing. Therefore, the administrative area barriers should be broken in the construction of the metropolitan area, and the overall planning should be implemented. According to the commuting characteristics and geographical characteristics of different functional areas, the Beijing metropolitan area can be further divided into three distinct circles, including the core layer, the suburban layer, and the outer suburban layer. Development strategies should be formulated according to regional characteristics, the commuting status of different circles can be improved through the development of suburban railways or rail transit, and the overall commuting accessibility of the Beijing metropolitan area should be improved. The results can help to formulate planning and management policies, which can improve the function and structure, and promote the benign development of the metropolitan area.

Key words: urban traffic, metropolitan area division, clustering algorithm, metropolitan area, carpooling data

中图分类号: