地理研究 ›› 2014, Vol. 33 ›› Issue (3): 520-531.doi: 10.11821/dlyj201403011

• 土地利用与生态环境 • 上一篇    下一篇

基于多智能体模型与建筑物信息的高空间分辨率人口分布模拟

卓莉1, 黄信锐1, 陶海燕1, 王芳2, 谢育航1   

  1. 1. 中山大学地理科学与规划学院, 综合地理信息研究中心, 广东省城市化与地理环境空间模拟重点实验室, 广州510275;
    2. 广州大学地理科学学院, 广州510006
  • 收稿日期:2013-03-22 修回日期:2013-09-23 出版日期:2014-03-10 发布日期:2014-03-10
  • 作者简介:卓莉,女,湖南慈利县人,副教授,主要从事地理信息科学与资源环境遥感研究。E-mail:zhuoli@mail.sysu.edu.cn
  • 基金资助:
    国家自然科学基金项目(41371499);广东省自然科学基金项目(S2012010010517);中山大学柳林教授千人计划科研启动项目(2011-2014)

High spatial resolution population distribution simulation based on building information and multi-agent

ZHUO Li1, HUANG Xinrui1, TAO Haiyan1, WANG Fang2, XIE Yuhang1   

  1. 1. Center of Integrated Geographic Information Analysis, Guangdong Key Laboratory for Urbanization and Geo-simulation, School of Geography and Planning, Sun Yat-sen University, Guangzhou 510275, China;
    2. School of Geography Sciences, Guangzhou University, Guangzhou 510006, China
  • Received:2013-03-22 Revised:2013-09-23 Online:2014-03-10 Published:2014-03-10

摘要: 自上而下的人口分布模拟模型自动化程度较低,难以分析人口分布成因,且因精细尺度的人口样本较难获取而不太适用于高空间分辨率人口分布模拟。提出了一种基于多智能体模型和建筑物信息的高空间分辨率人口分布模拟模型。首先利用建筑物三维分布数据提取住宅建筑,构建智能体人口分布模拟模型的环境;然后基于统计、调研数据定义智能体属性,确定智能体居住选择行为规则;最后以泰日社区为例进行了居住人口分布仿真模拟。研究结果表明,基于建筑物信息的人口分布多智能体模型,可以获取每栋建筑物上的人口,改进了当前高分辨率人口模拟主要只模拟小区或者居委会人口的不足;多智能体模型具有较高的自动化程度,不仅能获得较好的模拟结果,而且可在一定程度上从微观机理解释宏观居住分布模式,是对传统统计模型的有益补充。

关键词: 人口分布, 人口密度, 高空间分辨率, 多智能体, 建筑物

Abstract: Knowledge of population distribution is essential for understanding and responding to many social, political, economical and environmental problems. Many studies have been done to estimate population distribution. However, most of the existing methods cannot produce high spatial resolution results or require data which are difficult to acquire. In this paper, we present a method of simulating high spatial resolution population distribution, which uses a multi-agent model based on building information. Firstly, simulation environment of the multi-agent model is constructed based on building information, such as height, area, density; Secondly, the property and behavior rules are defined, based on statistical data and survey reports; Finally, the multi-agent model is used to simulate population distribution of a community named Tairi, which is located in Fengxian District, Shanghai. Results prove that our multi-agent population model is capable of producing high spatial resolution population distribution results at acceptable accuracy level. The method is highly automated and the data it requires are easily available, thereby it has great potential of application.

Key words: population distribution, population density, high spatial resolution, multi-agent, building