【作者】 宁敢；

【导师】 郑新奇；

【作者基本信息】 中国地质大学（北京） ， 地图制图学与地理信息工程， 2011， 硕士

【摘要】 土地是人类赖以生存的物质基础,城市土地承担着人类社会经济活动的功能,而城市土地利用系统本质上则是一个复杂的、开放的巨系统,为准确的掌握土地利用数据和土地利用变更动态,控制城市化进程,实现城市土地资源的可持续利用和城市发展的可持续性,需要利用城市土地利用时空动力学模型来模拟城市土地利用的动态变化。论文依托于国家公益性行业(土地)科研专项经费项目:中国多目标国土空间规划支撑技术研究与应用(201011018)对城市土地利用时空动力学模型进行研究。本文综述了目前国内外城市土地利用演化模拟与预测的研究方法,提出了构建基于群体智能及CA的城市土地利用时空动力学模型的设想;概述了复杂系统理论、元胞自动机理论和群体智能理论,归纳了各种理论应用到城市土地利用系统中的结合点与优势,阐述了基于群体智能与CA模型结合的复合模型模拟城市发展变化的可行性;构建了基于群体智能及CA的城市土地利用时空动力学模型。该模型以土地利用历史数据为数据源,应用群体智能中的蚁群算法对城市土地利用变化的转换规则进行了挖掘,与标准CA结合进行模拟预测。研究可知,通过蚁群算法挖掘出的转换规则明确、清晰,同时还有较强的自适应能力。本文选取1988年、1998年北京市为研究区域,对基于群体智能及CA的城市土地利用时空动力学模型进行了实证研究,并以2008年北京市土地利用空间结构对该模型进行了验证。结果显示,基于群体智能及CA的城市土地利用时空动力学模型模拟预测效果较好,能够较为客观、准确地预测出2008年北京市土地利用各地类数量及空间分布状况。研究表明,本文构建的基于群体智能及CA的城市土地利用时空动力学模型预测出的结果能够为土地管理部门制定土地利用总体规划、合理利用和开发土地资源提供可靠依据,同时还能够为各级政府部门提供即时准确的各类土地信息,以此作为国民经济、区域规划、农业区划的科学依据。更多还原

【Abstract】 Land is the basic resources for the survival of human race, and urban land is the basic resources for the social and economic activity of human race. The essence of urban land use system is an open complex giant system. In order to possess land use data and land use change accurately, control the process of urbanization, and accomplish the sustainable use of urban land resources and the sustainability of urban development, it’s a spatial-temporal dynamics model of urban land use that is needed to be used for simulation of urban land use change. This paper is supported by the National Public Benefit (Land) Research Foundation of China (No. 201011018).This paper reviews the simulation and prediction methods of the evolution of urban land use from scholars at home and abroad, puts forward the idea of building a spatial-temporal dynamics model of urban land use based on swarm intelligence and CA; overviews the theory of complex system, cellular automata and swarm intelligence, summarizes the conjoint points and advantages of using these theories in the research of urban land use system, analyzes the feasibility of using a compound model based on swarm intelligence and CA to simulate the change of urban land use; and builds a spatial-temporal dynamics model of urban land use based on swarm intelligence and CA. Based on the historical land use data, this model uses ant colony algorithm, one method of swarm intelligence, to mine the rules of urban land use change, which will be applied in standard CA, to simulate and predict. Research shows that, the transformation rules mined by ant colony algorithm are clear and have a strong adaptive ability.This paper takes Beijing, in the year of 1988, 1998 and 2008, as the study area to empirical research the model, and then compares to the real land use spatial structure of Beijing in 2008. The comparative study shows that by using the model based on swarm intelligence and CA can predict better, more objective and more precise in the prediction of the number and spatial distribution of land use classes of Beijing in 2008. The study shows that the prediction results of spatial-temporal dynamics model of urban land use based on swarm intelligence and CA can provide a reliable method to establish the general land use planning, use and develop land resources rationally for land management department, and also can supply an immediate and accurate information of land to government, which can help them make decisions on national economy, regional planning, and agricultural regionalization.更多还原