基于空间聚类方法的京津冀城市群多层级空间结构研究

doi:10.18306/dlkxjz.2017.11.005

地理科学进展 ›› 2017, Vol. 36 ›› Issue (11): 1359-1367.doi: 10.18306/dlkxjz.2017.11.005

基于空间聚类方法的京津冀城市群多层级空间结构研究

张珣^1,³(), 陈健璋¹, 黄金川^2,^3,^4,^*(), 于重重¹, 陈秀新¹

1. 北京工商大学计算机与信息工程学院食品安全大数据技术北京市重点实验室,北京 100048
2. 中国科学院区域可持续发展分析与模拟重点实验室,北京 100101
3. 中国科学院地理科学与资源研究所,北京 100101
4. 中国科学院大学资源与环境学院,北京 100049

出版日期:2017-12-07 发布日期:2017-12-07
通讯作者: 黄金川
作者简介:
作者简介：张珣(1986-),男,吉林辽源人,副教授,硕导,从事商业地理分析、GIS软件技术研究,E-mail: zhangxun@btbu.edu.cn。
基金资助:
教育部人文社会科学研究青年基金项目(15YJCZH224);北京市自然科学基金青年项目(9164025);首都流通业研究基地内设课题一般项目(JD-YB-2017-024);北京市教委科研计划面上项目(KM201510011010)

Multi-level spatial structure analysis of urban agglomeration in the Beijing-Tianjin-Hebei region based on spatial clustering algorithms

Xun ZHANG^1,³(), Jianzhang CHEN¹, Jinchuan HUANG^2,^3,^4,^*(), Chongchong YU¹, Xiuxin CHEN¹

1. Beijing Key Laboratory of Big Data Technology for Food Safety, Beijing Technology and Business University, Beijing 100048, China
2. Key Laboratory of Regional Sustainable Development Modeling, CAS, Beijing 100101, China
3. Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China
4. College of Resources and Environmental Sciences, University of Chinese Academy of Sciences, Beijing 100049, China

Online:2017-12-07 Published:2017-12-07
Contact: Jinchuan HUANG
Supported by:
Research Project of Humanities and Social Science on Youth Fund of the Ministry of Education, No.15YJCZH224;Beijing Natural Science Foundation, No.9164025;Opening Fund of Capital Circulation Industry Research Base, No.JD-YB-2017-024;Scientific Research Project of Beijing Educational Committee, No.KM201510011010

摘要/Abstract

摘要：

在京津冀协同发展战略背景下,以建设京津冀世界级城市群为引领,遵循城市发展规律,优化城市空间布局,明确京津冀城市群等级结构及其空间特征具有重要意义。本文以京津冀城市群156个区县为研究对象,从经济中心性、交通中心性、信息中心性、人口中心性4个角度,利用4种空间聚类方法进行5个等级的聚类分析,并基于克氏中心地理论对京津冀城市群等级划分结果进行空间结构分析。结果显示,自组织特征映射神经网络算法(SOM)较适合京津冀城市群的等级划分;京津冀城市群正从以北京城区为单核心的圈层空间结构向3条带型空间结构转变,其中京津都市发展走廊发育成熟,沿海都市发展带也初具规模,而包括雄安新区在内的京石都市发展带正在孕育。

关键词: 空间聚类, 空间结构, 京津冀城市群

Abstract:

With the main focus of regional competition shifting from cities to urban agglomerations, it is important to analyze the spatial structure and direction of coordinated development in urban agglomerations. This is especially significant in the Beijing-Tianjin-Hebei urban agglomeration, which is a core urban agglomeration in China. Machine learning algorithms are relatively new methods for addressing geographical problems. Clustering method, as unsupervised learning, is useful for classifying geographical units without the need for priori knowledge. Using data from 156 counties in the Beijing-Tianjin-Hebei urban agglomeration, this study applied four clustering algorithms: the K-means, density-based spatial clustering of applications with noise (DBSCAN), Chameleon, and self-organizing map (SOM) methods, for classifying counties and districts in the Beijing-Tianjin-Hebei urban agglomeration from the perspectives of economic centrality, traffic centrality, information centrality, and population centrality. GDP of the counties in 2014 was used to represent economic centrality; density of road networks in counties and attraction factor, calculated by the unsold train tickets in different time periods of the year, represent traffic centrality; Sina Weibo check-in data were used to represent information centrality; and county/district population represents population centrality. The result classifies the urban agglomeration into several levels. Respectively, K-means algorithm classifies counties into five levels; DBSCAN algorithm classifies counties into six levels; Chameleon algorithm classifies counties into six levels; and SOM algorithm classifies counties into five levels. SOM is the most applicable algorithm for the division of the urban agglomeration because the structure of counties is stable. This study further analyzed the spatial structure of the urban agglomeration with the central place theory, which points out that an agglomeration should contain certain number of counties in every level. The result of the SOM algorithm matches the central place theory. This research shows that there were remarkable gaps between different levels of the urban agglomeration. The central area of Beijing, as the core of the region, has strong radiation effect on the surrounding areas, but its functions are shared by the nearby counties. Moreover, the second and third level central cities distribute evenly and play an important role in regional development.

Key words: spatial clustering, spatial structure, Beijing-Tianjin-Hebei Region Agglomeration

张珣, 陈健璋, 黄金川, 于重重, 陈秀新. 基于空间聚类方法的京津冀城市群多层级空间结构研究[J]. 地理科学进展, 2017, 36(11): 1359-1367.

Xun ZHANG, Jianzhang CHEN, Jinchuan HUANG, Chongchong YU, Xiuxin CHEN. Multi-level spatial structure analysis of urban agglomeration in the Beijing-Tianjin-Hebei region based on spatial clustering algorithms[J]. PROGRESS IN GEOGRAPHY, 2017, 36(11): 1359-1367.

图/表 9

表1

表2

表3

表4

图1

图2

图3

表5

图4

参考文献 31

[1]	陈园园, 李宁, 丁四保. 2011. 城市群空间联系能力与SOM 神经网络分级研究: 以辽中南城市群为例[J]. 地理科学, 31(12): 1461-1467.
	[Chen Y Y, Li N, Ding S B.2011. Spatial combination capacity and classification based on SOM network of urban agglomerations: A case study of central and southern Liaoning urban agglomerations[J]. Scientia Geographica Sinica, 31(12): 1461-1467.]
[2]	方碧琪, 汪永新. 1991. 北京市各区县经济分析的聚类方法[J]. 数理统计与管理, (1): 25-30. doi: 10.3969/j.issn.1002-1566.1999.01.006
	[Fang B Q, Wang Y X.1991. Clustering analysis on the economy of Beijing regions[J]. Application of Statistics and Management, (1): 25-30.] doi: 10.3969/j.issn.1002-1566.1999.01.006
[3]	顾朝林, 王颖, 邵园, 等. 2015. 基于功能区的行政区划调整研究: 以绍兴城市群为例[J]. 地理学报, 70(8): 1187-1201. doi: 10.11821/dlxb201508001
	[Gu C L, Wang Y, Shao Y, et al.2015. Research on administrative divisions based on functional areas analysis: A case of Shaoxing metropolitan area[J]. Acta Geographica Sinica, 70(8): 1187-1201.] doi: 10.11821/dlxb201508001
[4]	韩玉刚, 叶雷. 2016. 中国欠发达省际边缘区核心城市的选择与区域带动效应: 以豫皖省际边缘区为例[J]. 地理研究, 35(6): 1127-1140. doi: 10.11821/dlyj201606010
	[Han Y G, Ye L.2016. Core city selection in Chinese undeveloped provincial border-region and its promoting effects on regional development: A case study of Anhui-Henan Provincial Border-region[J]. Geographical Research, 35(6): 1127-1140.] doi: 10.11821/dlyj201606010
[5]	黄金川, 刘倩倩, 陈明. 2014. 基于GIS的中国城市群发育格局识别研究[J]. 城市规划学刊, (3): 37-44.
	[Huang J C, Liu Q Q, Chen M.2014. The identification of urban agglomeration distribution in China based on GIS analysis[J]. Urban Planning Forum, (3): 37-44.]
[6]	金建国. 2014. 聚类方法综述[J]. 计算机科学, 41(Z2): 288-293.
	[Jin J G.2014. Review of clustering method[J]. Computer Science, 41(Z2): 288-293.]
[7]	李国平, 罗心然. 2017. 京津冀地区人口与经济协调发展关系研究[J]. 地理科学进展, 36(1): 25-33. doi: 10.18306/dlkxjz.2017.01.003
	[Li G P, Luo X R.2017. Coordinated development between population and economy in the Beijing-Tianjin-Hebei region[J]. Progress in Geography, 36(1): 25-33.] doi: 10.18306/dlkxjz.2017.01.003
[8]	李新延, 李德仁. 2005. DBSCAN空间聚类算法及其在城市规划中的应用[J]. 测绘科学, 30(3): 51-53. doi: 10.3771/j.issn.1009-2307.2005.03.014
	[Li X Y, Li D R.2005. DBSCAN spatial clustering algorithm and its application in urban planning[J]. Science of Surveying and Mapping, 30(3): 51-53.] doi: 10.3771/j.issn.1009-2307.2005.03.014
[9]	刘卫东, 金凤君, 张文忠, 等. 2011. 中国经济地理学研究进展与展望[J]. 地理科学进展, 30(12): 1479-1487. doi: 10.11820/dlkxjz.2011.12.003
	[Liu W D, Jin F J, Zhang W Z, et al.2011. Progress in economic geography (2006-2011)[J]. Progress in Geography, 30(12): 1479-1487.] doi: 10.11820/dlkxjz.2011.12.003
[10]	马林靖, 周立群. 2010. 基于聚类分析法的县域均衡发展指数构建与测度应用研究: 以天津12个涉农区县为例[J]. 经济经纬, (2): 57-60. doi: 10.3969/j.issn.1006-1096.2010.02.014
	[Ma L J, Zhou L Q.2010. A research into the construction and measurement of county balanced development indexes and its application based on cluster analysis method[J]. Economic Survey, (2): 57-60.] doi: 10.3969/j.issn.1006-1096.2010.02.014
[11]	牛方曲, 刘卫东, 宋涛, 等. 2015. 城市群多层次空间结构分析算法及其应用:以京津冀城市群为例[J]. 地理研究, 34(8): 1447-1460. doi: 10.11821/dlyj201508004
	[Niu F Q, Liu W D, Song T, et al.2015. A multi-level spatial structure analysis algorithm for urban agglomeration study in China[J]. Geographical Research, 34(8): 1447-1460.] doi: 10.11821/dlyj201508004
[12]	王波, 甄峰. 2016. 网络社区交流中距离的作用: 以新浪微博为例[J]. 地理科学进展, 35(8): 983-989.
	[Wang B, Zhen F.2016. The role of distance in online social networks: A case study of Sina micro-blog[J]. Progress in Geography, 35(8): 983-989.]
[13]	闫友彪, 陈元琰. 2004. 机器学习的主要策略综述[J]. 计算机应用研究, 21(7): 4-10, 13. doi: 10.3969/j.issn.1001-3695.2004.07.002
	[Yan Y B, Chen Y Y.2004. A survey on machine learning and its main strategy[J]. Application Research of Computers, 21(7): 4-10, 13.] doi: 10.3969/j.issn.1001-3695.2004.07.002
[14]	杨志民, 化祥雨, 叶娅芬, 等. 2015. 金融空间联系及K-means聚类中心等级识别研究: 以长三角为例[J]. 地理科学, 35(2): 144-150.
	[Yang Z M, Hua X Y, Ye Y F, et al.2015. Spatial combination of finance and center level identify based on K-means clustering: A case study of the Changjiang river delta[J]. Scientia Geographica Sinica, 35(2): 144-150.]
[15]	袁媛, 王仰麟, 马晶, 等. 2014. 河北省县域贫困度多维评估[J]. 地理科学进展, 33(1): 124-133. doi: 10.11820/dlkxjz.2014.01.014
	[Yuan Y, Wang Y L, Ma J, et al.2014. Multidimensional evaluation of county poverty degree in Hebei Province[J]. Progress in Geography, 33(1): 124-133.] doi: 10.11820/dlkxjz.2014.01.014
[16]	张旺, 申玉铭. 2012. 京津冀都市圈生产性服务业空间集聚特征[J]. 地理科学进展, 31(6): 742-749. doi: 10.11820/dlkxjz.2012.06.010
	[Zhang W, Shen Y M.2012. The spatial characteristics of producer service agglomeration in Beijing-Tianjin-Hebei metropolitan region[J]. Progress in Geography, 31(6): 742-749.] doi: 10.11820/dlkxjz.2012.06.010
[17]	周一星, 张莉, 武悦. 2001. 城市中心性与我国城市中心性的等级体系[J]. 地域研究与开发, 20(4): 1-5. doi: 10.3969/j.issn.1003-2363.2001.04.001
	[Zhou Y X, Zhang L, Wu Y.2001. Study of China's urban centrality hierarchy[J]. Areal Research and Development, 20(4): 1-5.] doi: 10.3969/j.issn.1003-2363.2001.04.001
[18]	Bação F, Lobo V, Painho M.2005. Self-organizing maps as substitutes for K-means clustering[C]//Proceedings of the 5th international conference on computational science. Berlin, Heidelberg, Germany: Springer-Verlag: 476-483.
[19]	Baselice F, Coppolino L, D'Antonio S, et al.2015. A DBSCAN based approach for jointly segment and classify brain MR images[C]// Proceedings of the 37th International Conference of the IEEE Engineering in Medicine and Biology Society. Milan, Italy.
[20]	Bell T L, Lieber S R, Rushton G.2015. Clusterg of services in central places[J]. Annals of the Association of American Geographers, 64(2): 214-225. doi: 10.1111/j.1467-8306.1974.tb00972.x
[21]	Duda R O, Hart P E, Stork D G.2001. Pattern classification[M]. 2nd ed. New York, NY: A Wiley-Interscience Publication, 2001.
[22]	Ferrari D C, de Castro L N.2015. Clustering algorithm selection by meta-learning systems: A new distance-based problem characterization and ranking combination methods[J]. Information Sciences, 301: 181-194. doi: 10.1016/j.ins.2014.12.044
[23]	Gong W W.2010. ARMA-GRNN for passenger demand forecasting[C]//2010 Sixth international conference on natural computation (ICNC). Yantai, China: IEEE.
[24]	Gómeza D, Zarrazolab E, Yáñezc J, et al.2015. A Divide-and-Link algorithm for hierarchical clustering in networks[J]. Information Science, 316: 308-328. doi: 10.1016/j.ins.2015.04.011
[25]	Grubesic T H, Wei R, Murray A T.2014. Spatial clustering overview and comparison: Accuracy, sensitivity, and computational expense[J]. Annals of the Association of American Geographers, 104(6): 1134-1156. doi: 10.1080/00045608.2014.958389
[26]	Hagen L, Kahng A B.2006. New spectral methods for ratio cut partitioning and clustering[J]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 11(9): 1074-1085. doi: 10.1109/43.159993
[27]	Kohonen T.1990. The self-organizing map[J]. Neurocomputing, 21(1-3): 1-6.
[28]	Lin G F, Wu M C.2007. A SOM-based approach to estimating design hyetographs of ungauged sites[J]. Journal of Hydrology, 339(3-4): 216-226. doi: 10.1016/j.jhydrol.2007.03.016
[29]	Marshall J U.1989. The structure of urban systems[M]. Toronto, Canada: University of Toronto Press.
[30]	Preston R E.1970. Two centrality models[J]. Yearbook of the Association of Pacific Coast Geographers, 32: 59-78. doi: 10.1353/pcg.1970.0006
[31]	Zhou Y H, Guo H, Dong Y S, et al.2011. Cluster analysis for drilling-quality based on the modified algorithm of InDBSCAN[J]. Key Engineering Materials, 480-481: 877-882. doi: 10.4028/www.scientific.net/KEM.480-481.877

城区名称	包含区
北京城区	东城区、西城区、朝阳区、丰台区、海淀区、石景山区
天津城区	和平区、河东区、河西区、南开区、河北区、红桥区
保定城区	莲池区、竞秀区(含高新区)、徐水区、清苑区、满城区
沧州城区	新华区、运河区
承德城区	双桥区、双滦区、高新区、营子区
邢台城区	桥东区、桥西区
张家口城区	桥东区、桥西区、宣化区、下花园区、万全区、崇礼区
唐山城区	路北区、路南区、古冶区、开平区、丰润区、丰南区、曹妃甸区
石家庄城区	桥西区、新华区、长安区、裕华区、井陉矿区、藁城区、鹿泉区、栾城区
廊坊城区	安次区、广阳区
邯郸城区	丛台区、复兴区、邯山区、峰峰矿区
秦皇岛城区	海港区、山海关区、北戴河区、抚宁区

算法	算法效率	适合的数据类型	发现的聚类类型	对噪声数据或异常数据的敏感性	对输入顺序的敏感性	其他
K-means	高	数值	凸形或球形	不敏感	一般	易于理解和实现,时间复杂度低
Chameleon	一般	任意	任意形状	不敏感	不敏感	算法灵活但执行时间长
DBSCAN	一般	数值	任意形状	敏感	敏感	处理高纬度数据效果差
SOM	高	任意	任意形状	不敏感	敏感	结果有特征映射,能可视化

	第一类	第二类	第三类	第四类	第五类	第六类
K-means	北京市城区等3个区县	保定市城区等8个区县	北辰区等9个区县	霸州市等40个区县	安国县等96个区县
Chameleon	成安县等10个区县	霸州市等19个区县	安平县等25个区县	北京市城区等22个区县	安新县等35个区县	安国市等40个区县
DBSCAN	固安县等6个区县	隆化县等6个区县	广宗县等7个区县	滦平县等7个区县	大名县等15个区县	噪声数据(115个区县)
SOM	北京市城区	石家庄、天津市城区	保定市城区等21个区县	霸州市等42个区县	安国市等90个区县

	人口	GDP	路网密度	吸引因子	微博签到数
第一级	1.000000	1.000000	0.566022	0.925568	1.000000
第二级	0.187632	0.323590	0.712439	0.824197	0.089826
第三级	0.079778	0.066928	0.263885	0.202920	0.016356
第四级	0.015132	0.036220	0.096532	0.052529	0.003622
第五级	0.008050	0.025172	0.048087	0.013577	0.000620

基于空间聚类方法的京津冀城市群多层级空间结构研究

Multi-level spatial structure analysis of urban agglomeration in the Beijing-Tianjin-Hebei region based on spatial clustering algorithms

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 31

相关文章 15

Metrics

本文评价

推荐阅读 0

聚类算法	参数设置
K-means	质心个数k=5;迭代次数n=15
Chameleon	k最邻近值设定为k=3;度量函数距离阈值MinMetric=0.015
DBSCAN	中心点P的扫描邻域半径Eps=3;以P为中心的邻域内最小包含点数阈值min=3
SOM	输入层输入神经元k=5;迭代次数n=300

[1]	卢明华, 周悦颜, 刘汉初, 许欣. 北京企业对河北直接投资的时空动态特征及影响因素[J]. 地理科学进展, 2020, 39(3): 389-401.
[2]	罗奎, 李广东, 劳昕. 京津冀城市群产业空间重构与优化调控[J]. 地理科学进展, 2020, 39(2): 179-194.
[3]	段学军,邹辉,陈维肖,王雅竹,叶磊. 长江经济带形成演变的地理基础[J]. 地理科学进展, 2019, 38(8): 1217-1226.
[4]	段亚明, 刘勇, 刘秀华, 何东. 基于宜出行大数据的多中心空间结构分析——以重庆主城区为例[J]. 地理科学进展, 2019, 38(12): 1957-1967.
[5]	毛帅永, 焦利民, 许刚, 李泽慧. 基于多源数据的武汉市多中心空间结构识别[J]. 地理科学进展, 2019, 38(11): 1675-1683.
[6]	马冰滢, 黄姣, 李双成. 基于生态-经济权衡的京津冀城市群土地利用优化配置[J]. 地理科学进展, 2019, 38(1): 26-37.
[7]	徐银凤, 汪德根. 中国城市空间结构的高铁效应研究进展与展望[J]. 地理科学进展, 2018, 37(9): 1216-1230.
[8]	王晓梦, 王锦, 吴殿廷. “交通—产业”耦合背景下的京津冀城市群空间发育特征[J]. 地理科学进展, 2018, 37(9): 1231-1244.
[9]	宋琼, 赵新正, 李同昇, 刘静玉. 多重城市网络空间结构及影响因素——基于有向多值关系视角[J]. 地理科学进展, 2018, 37(9): 1257-1267.
[10]	叶雅玲, 杨博飞, 何紫云, 曾贝妮, 潘峰华, 方成. 中国城市网络空间结构研究——基于A股上市医药公司网络[J]. 地理科学进展, 2018, 37(8): 1096-1105.
[11]	杨喜平, 方志祥. 移动定位大数据视角下的人群移动模式及城市空间结构研究进展[J]. 地理科学进展, 2018, 37(7): 880-889.
[12]	王琦, 黄金川. 东京都市圈大气污染防治政策对京津冀的启示[J]. 地理科学进展, 2018, 37(6): 790-800.
[13]	陈明星, 郭莎莎, 陆大道. 新型城镇化背景下京津冀城市群流动人口特征与格局[J]. 地理科学进展, 2018, 37(3): 363-372.
[14]	李郇, 周金苗, 黄耀福, 黄玫瑜. 从巨型城市区域视角审视粤港澳大湾区空间结构[J]. 地理科学进展, 2018, 37(12): 1609-1622.
[15]	苏文松, 方创琳. 京津冀城市群高科技园区协同发展动力机制与合作共建模式——以中关村科技园为例[J]. 地理科学进展, 2017, 36(6): 657-666.