地理学报 ›› 2008, Vol. 63 ›› Issue (8): 869-881.doi: 10.11821/xb200808008

• GIS应用 • 上一篇    下一篇

多智能体城市土地扩张模型及其应用

张鸿辉1,2, 曾永年1, 金晓斌3, 尹长林1,2, 邹滨1   

  1. 1. 中南大学信息物理工程学院, 长沙410083;
    2. 长沙市规划信息服务中心, 长沙410013;
    3. 南京大学地理与海洋科学学院, 南京210093
  • 收稿日期:2007-11-12 修回日期:2008-06-27 出版日期:2008-08-25 发布日期:2008-08-25
  • 通讯作者: 曾永年, E-mail: ynzeng@mail.csu.edu.cn
  • 作者简介:张鸿辉(1980-), 男, 博士生, 湖南长沙人, 主要从事土地利用变化与模拟研究。E-mail: zhhgis@163.com
  • 基金资助:

    国家自然科学基金项目(40771198)

Urban Land Expansion Model Based on Multi- agent System and Application

ZHANG Hnghui1,2, ZENG Yongnian1, JIN X aobin3, YIN Changlin1,2, ZOU Bing1   

  1. 1. School of Info-Physics and Geomatics Engineering,Central South University,Changsha 410083,China;
    2. Changsha Planning Information Service Center,Changsha 410013,China;
    3. School of Geographic and Oceanographic Sciences,Nanjing University,Nanjing 210093,China
  • Received:2007-11-12 Revised:2008-06-27 Online:2008-08-25 Published:2008-08-25
  • Supported by:

    National Natural Science Foundation of China, No.40771198

摘要:

传统的城市土地扩张模型多为静态模型, 无法呈现空间上每一时间点的土地利用状况, 以元胞自动机(Cellular Automata) 模型为代表的新型城市土地扩张模型虽然具有动态特性, 但其无法描述影响城市土地扩张的智能体(Agent) 之间所产生的多元变化结果。以多智能体系统(Multi-Agent System) 理论为基础, 建立城市土地资源时间和空间配置规则, 构建了动态且能描述影响城市土地扩张的智能体(Agent) 间互动关系的城市土地扩张模型, 并以长沙市 区为例, 应用所构建之模型进行了城市土地扩张的实证分析。结果表明: 该模型可以反映城市土地扩张的基本特征和规律, 对于解释城市土地扩张的成因、理解智能体行为对城市土地 扩张过程的影响是合适的。并且将模拟结果与遥感土地利用解译结果对比, 1998 年、2001 年、2005 年城市土地扩张模拟的点对点精度均达到68%以上, 从而能够为政府和城市规划者 制定用地政策提供辅助决策支持。

关键词: 多智能体系统, 城市土地扩张, 土地资源分配, 长沙

Abstract:

The traditional urban land expansion model has paid m uch attention on a staticstate, and few scholars have focused research on land use change situations through tim e. A lthough the urban land expansion model based on cellular autom ata perform s dynam ic feature, the m odel fails to display the m ultivariate change results produced from different agents' behaviors. O n account of the above reasons, this research has built up a set of spatial-tem poral land resource allocation rule and developed a dynam ic urban land expansion m odel based on m ulti-agent system , w hich can sim ulate the interactions am ong agents. W hat's m ore, this m odel is applied to analyze and sim ulate urban land expansion process taking dow ntow n of C hangsha city as a study area. The results show that this m odel can not only reflect basic characteristics of urban land expansion, but also help to explain the reasons for urban land expansion process and understand the effect of agents' behavior on urban land expansion,w hat's m ore,in contrast to sim ulation results w ith land use classification m ap from rem ote sensing im ages, the accuracies of the sim ulation reached over 68% according to the cell-by-cell com parison,w hich m akes it possible to provide land use decision-m ake support to governm ent and urban planners.

Key words: multi-agent system, urbanland expansion, land resource allocation, Changsha city