地理研究 ›› 2014, Vol. 33 ›› Issue (2): 349-357.doi: 10.11821/dlyj201402013.1

• 区域发展 • 上一篇    下一篇

基于空间自相关的东莞市主体功能区划分

林锦耀, 黎夏   

  1. 中山大学地理科学与规划学院, 广东省城市化与地理环境空间模拟重点实验室, 广州 510275
  • 收稿日期:2013-07-15 修回日期:2013-11-23 出版日期:2014-02-10 发布日期:2014-02-10
  • 通讯作者: 黎夏(1962- ),男,教授,博士生导师。E-mail:lixia@mail.sysu.edu.cn E-mail:lixia@mail.sysu.edu.cn
  • 作者简介:林锦耀(1989- ),男,广东广州人,硕士,从事地理模拟与优化等研究。E-mail:ljy2012@foxmail.com
  • 基金资助:
    国家自然科学基金项目(41371376)

MFOZ planning of Dongguan based on spatial autocorrelation by using genetic algorithms

LIN Jinyao, LI Xia   

  1. School of Geography and Planning, and Guangdong Key Laboratory for Urbanization and Geo-simulation, Sun Yat-sen University, Guangzhou 510275, China
  • Received:2013-07-15 Revised:2013-11-23 Online:2014-02-10 Published:2014-02-10

摘要: 中国“十一五”规划纲要明确提出,根据资源环境承载能力、现有开发密度和发展潜力,统筹考虑未来中国人口分布、经济布局、国土利用和城镇化格局,将国土空间划分为优化开发、重点开发、限制开发和禁止开发四类主体功能区。至今已有不少学者开展主体功能区划分研究,但所用方法工作量大,或带有一定的主观性,且未考虑同类功能区集聚或分散程度。因此提出利用遗传算法改进传统聚类方法,自动划分主体功能区,在划分过程中考虑区域的全局空间自相关特性,使同类功能区在空间上呈集聚分布的格局。以近年来城市快速扩张的东莞市为例,验证了此方法的可行性,能简单有效地进行主体功能区划分。与常用的K-means聚类方法相比,新方法划分结果更符合实际情况,能进一步推广应用到其它地区的主体功能区划分。

关键词: 主体功能区, 空间自相关, 遗传算法, 东莞

Abstract: According to the 11th Five-Year Plan(2006-2010), land should be divided into four types of Major Function Oriented Zones(MFOZ), namely optimal development zone,key development zone, restricted development zone and prohibited development zone, based on bearing capacity of resources and environment, existing development intensity and potential of development. So far, numerous researchers have already conducted some related studies. Still, these somewhat subjective and labor-consuming methods seldom concern the cluster or dispersion degree of the same certain zone. This research tried to improve traditional clustering-based method by using genetic algorithms. This automatic method which is then applied to MFOZ planning of Dongguan concerns global spatial autocorrelation of the region. The comparison demonstrates that this simple and effective method has better performance than commonly-used clustering-based methods. The proposed method can be further applied to MFOZ planning of other regions.

Key words: MFOZ, spatial autocorrelation, genetic algorithms, Dongguan

PACS: 

  • TU984.2