【作者】 张乐珊；

【导师】 陈戈；

【作者基本信息】 中国海洋大学 ， 地图学与地理信息系统， 2010， 博士

【摘要】 随着城市化进程和城市发展速度的不断加快,城市现象和城市问题日益复杂化,并对国民经济、社会和谐和自然环境的健康可持续发展产生深刻影响。因此,对城市系统演化过程的复杂性、动态性探索受到越来越广泛的重视,成为本世纪的重要研究课题之一。基于复杂系统理论和自组织理论的城市空间增长动态模拟是以城市空间系统为研究对象,以城市空间的时空扩展变化为主要研究内容,以探索城市空间发展演化的客观规律和动力机制为主要研究目标,是进行城市系统复杂性研究的主要方法手段。但是,一直以来,大多数的城市模型都是将城市空间看作是平面城市,忽略立体空间变化的影响,仅对二维城市空间的发展变化进行建模。这对研究今天高楼林立的现代城市而言,已经面临越来越多的不足和局限性。因为,现代城市的发展不仅仅是二维平面的扩展,立体空间的增长越来越显著,而且对整个城市空间系统变化产生的影响已经不容忽视。在本论文中,我们将以三维城市空间系统为研究对象,建立可以描述三维城市空间动态演化过程的动态模型,并在此基础上,进行模拟分析和和应用实践,从而为城市空间复杂性、动态性研究向高维扩展做出有益尝试。本论文的研究工作可以概括为：(1)引介复杂系统理论和自组织理论,认识城市发展的复杂性、动态性和自组织性,理解城市空间发展的内在机制,建立自组织的城市发展观。在此基础上,以元胞自动机为城市动态建模工具,定义三维城市空间增长的动力因子和评价规则,建立三维城市空间增长的理论模型。通过对模拟结果进行静态分析、参数敏感度分析和动态分析,证明该模型可以丰富的反映城市空间增长的自组织特性,能够揭示三维城市发展的动力机制和一般规律；(2)基于面向对象的思想,提出了紧密融合空间数据模型与空间过程模型的具体机制,将元胞自动机的动力学机制融合到VRGIS一体化平台系统中,构建了基于虚拟现实地理信息系统环境的城市动态建模框架。实验证明,两者的融合可以实现静态、离散的空间数据模型与动态、连续的空间过程模型的紧密融合,为探索多维城市空间现象和空间过程问题提供有利的平台环境；(3)以真实城市的地理空间数据为基础,通过集成多个多维动态子模型,将三维城市空间增长理论模型进行应用扩展。实验结果证明,集成模型可以有效、正确的量化对未来各种增长情景的假设,提供定量分析与定性解释相结合的方法,对研究区域在未来各种情景下的扩展方向、规模、特点、速度、路径进行预测和分析,为研究区域的规划决策提供了有利的实验环境；(4)进行基于情景开发的城市发展预测,生成了不同社会经济发展背景下不同的城市空间格局,通过对预测结果进行比较分析,为决策者及规划人员理解不同城市发展背景、不同政策倾向下未来可能形成的空间格局,及其对周围环境产生的影响进行评估提供科学的定量分析依据。更多还原

【Abstract】 With the high speed of urbanization and urban growth process, the urban phenomenon and problems are more complex than before, and have a more deeper influence on the health and sustainable development of economic, society and natural envionment. So, the exploration of the complexity, dynamics of urban evolution has been attached more importance, which has been becoming a significant subject of this century.Complexity and self-organization theory based simulation of urban dynamic growth takes urban space system as the study object, the temporal-spatial changes of urban space as the main research content, and the exploration of the natural law and dynamic factors of the urban evolution as the final goal, which has been the main tools used in studying the complexity of urban system. However, to date, most of the urban models are modeling through compressing the urban space to two dimensional space, that ignoring the influence from the third dimensional space. This has been the main shortage and limitation in dealing with morden cities, which are charactered with high buildings,even skyscrapers. So the third dimension of the urban space can not be ignored any more, which affect the dynamics of the whole city. In this thesis, we take the three dimensional urban space as the study object, and build up a three dimensional urban model to descripe the evolution process of our cities. Then through the simulation, analysis and application, to contribute the high dimension expanded exploration of the complexity of urban space.The main research work of this thesis could be summarized as following four points:(1) Introduce the complexity and self-organization theory, understanding the complexity, dynamics of urban evolution process and its inner laws, to study the urban space as a self-organization system;(2) According to the previous study, on the basis of self-organization theory, using cellular automation as the modeling tools, and defining the dynamic factors and rules of three dimensional space growth to build up a three dimensional urban space growth simulation model. According to the simulation results, to describe the formation and development of three dimensional space pattern, and analysis the sensity of the model parameters;(3) Integrating the cellular automation with VRGIS platform, to construct a virtual geographical space based dyanmic modeling framework, and using this framework as the simulation test tools. For one side, this is an application trail of the high dimensional modeling, for the other side, it is also a new attempt to further the close integration of VR and GIS to a higher level via application;(4) On the basis of the real city geographical data, expand the CA model to satisfy the demand of real city simulationm through coupling with an agent-based residential demand model and a CA based two dimensional development suitability model as the urban dynamic simulation engine to applied in the real city simulation. According to the project demand and the alternative hypothesis of future urban growth trends, to simulate and predict the generated spacial patterns of different future scenarios, and provide the scientific quantitative analysis for decision-makers and planners to understand the possible spatial changes and their effects to around environment under different future scenarios and policy objectives.更多还原