自然资源学报 ›› 2012, Vol. 27 ›› Issue (3): 497-509.doi: 10.11849/zrzyxb.2012.03.016

• 资源研究方法 • 上一篇    下一篇

基于GECM与CA+ANN模型的土地资源优化配置与模拟

郑荣宝1, 董玉祥2, 陈梅英3   

  1. 1. 广东工业大学 管理学院,广州 510520;
    2. 中山大学 地理科学与规划学院,广州 510275;
    3. 广州市国土资源局和房屋管理局花都区分局,广州 510030
  • 收稿日期:2011-04-19 修回日期:2011-06-20 出版日期:2012-03-20 发布日期:2012-03-20
  • 作者简介:郑荣宝(1975- ),男,福建永定人,博士,主要研究土地资源开发利用与保护。E-mail:zhengrongbao@163.com
  • 基金资助:

    国家自然科学基金项目(40771003);广东省哲学社科基金(090-23)。

Simulation on Optimized Allocation of Land Resource Based on GECM and CA+ANN Model

ZHENG Rong-bao1, DONG Yu-xiang2, CHEN Mei-ying3   

  1. 1. College of Management, Guangdong University of Technology, Guangzhou 510520, China;
    2. College of Geographic Science and Planning, Zhongshan University, Guangzhou 510275, China;
    3. Guangzhou Land and Resources and the Housing Authority Huadu Branch, Guangzhou 510030, China
  • Received:2011-04-19 Revised:2011-06-20 Online:2012-03-20 Published:2012-03-20

摘要: 土地资源优化配置是土地科学研究的热点。论文在前人研究成果的基础上,以多目标动态规划方法中的逐步宽容约束(GECM)方法为基础,以社会、经济和生态效益为系统目标,设置8个方面的GECM约束条件,对广州市花都区2020年土地利用结构进行优化配置研究,并将土地利用结构优化配置结果作为约束条件,结合人工神经网络与元胞自动机(ANN+CA)模型对研究区土地利用空间布局进行模拟。研究结果表明:①GECM是一种较好的土地利用结构优化方法,花都区2008年建设用地规模(23 097 hm2)接近于2020年的合理用地水平(28 800 hm2);②ANN+CA可以实现土地资源的空间配置与模拟,通过GECM与ANN+CA的耦合进行土地资源优化配置的结果比较符合花都区的实际情况,模拟的方案更具合理性,能实现区域土地资源优化配置的时间与空间的统一。

关键词: 土地资源, 逐步宽容法, 元胞自动机模拟, 花都区

Abstract: Optimal allocation of land resources is becoming a hot spot in land scientific research. Based on the results of previous studies, the authors build up the model of the Gradual Easy Constraint Method (GECM) with the multi-objective dynamic programming method. This model aims on social, economic and ecological benefits, and sets up eight aspects as GECM constraints to research on the optimized allocation of land use structure in Huadu District in 2020. Then through the results of structural optimization as constraints, we can use the artificial neural networks and cellular automata (ANN+CA) model to simulate and predict the change of land use from 2008 to 2020. The results indicate that: 1) GECM is a good method on land use structure optimization research, the built-up area in 2008 (23097 hm2) will reach the rational level in 2020 (28800 hm2), and this district must select the direction of land intensive use in the future years. 2) CA+ANN can achieve the target of spatial allocation and simulation on land resources. Coupling with GECM and ANN+CA models, the research results are more suitable and reasonable to the actual situation in Huadu District. It also can achieve the unity of time and space in optimal allocation of regional land use.

Key words: land resource, gradual easy constraint method, CA simulation, Huadu District

中图分类号: 

  • F301.2