【作者】 滕龙妹；

【导师】 刘南；

【作者基本信息】 浙江大学 ， 地图学与地理信息系统， 2008， 博士

【摘要】 土地是人类生存环境演变和社会经济发展的重要决定因素。及时准确掌握各种土地资源的类型、数量和空间分布,是政府和各级土地管理部门制定政策和落实各项管理措施的重要基础和科学依据。随着信息化的不断深入,土地资源信息已经渗透到土地利用、土地产权产籍、土地利用规划等各个土地业务管理之中,并且信息量还在不断增长。然而,土地资源的整体利用率较低,分布在各个单位、地区的不同节点上的数据与信息之间难以共享,存在着不同程度的信息孤岛。传统的分布式GIS技术体系在海量时空数据、信息资源的共享与互操作等方面存在不足。网格技术的出现,为解决传统分布式GIS中存在的问题提供了新的解决思路,利用网格技术把分布的硬件资源、数据资源、应用软件形成一个庞大的虚拟计算机来消除信息孤岛,对促进土地资源数据共享、集成和利用具有重要的现实意义。本文在对网格、海量时空数据管理、土地管理等方面进行深入的理论研究和技术分析的基础上,尝试将GIS学科、计算机学科与土地学科的研究结合起来,提出适合网格环境下土地资源管理的服务模型、管理模式、网格服务体系、应用模型等,为土地资源海量时空数据共享及应用提供技术支撑。首先,根据土地资源管理及数据内容的特点,结合资源体的概念,提出了土地资源体概念。在土地资源体概念的基础上,针对数据组织提出了土地资源时空数据服务模型。该模型采用虚拟时空数据服务器来组织各个类型的土地资源体,对土地资源进行统一管理。文中重点探讨了该模型下的土地资源访问与组织机构模型和土地资源元数据模型。结合具体的土地资源管理业务需要,摸索了一套基于该模型的土地资源体多级变粒度数据组织管理模式。该模式对土地资源体横向按资源体、图库、图层集合、图层和地物等不同资源粒度,纵向按多级区域层次进行管理,并给出了该模式下的资源体的元数据组织、节点动态管理、元数据复制、资源体动态组合等技术实现。其次,在开放式GIS网格服务体系（OG₂SA）的研究基础上,将土地资源时空数据服务模型扩展到OG₂SA下,提出了土地资源时空数据网格服务模型（LRSTD-GSM）。同时在研究土地资源数据的时空分区、逻辑分幅、土地资源体多粒度等特点基础上,设计了适合土地资源管理的网格时空数据库,并在网格时空数据库基础上给出了基于区域、任务、Agent、服务和数据5个层次的土地资源管理的应用模型及其实现。最后,在土地资源管理的服务模型、管理模式、网格服务体系、应用模型等研究的基础上,构建面向网格的土地资源管理与应用试验系统GridLandRSMgr。研究、测试与应用实例表明,LRSTD-GSM的设计思想的合理性和技术方法的有效性,并提出了下一步的研究方向。更多还原

【Abstract】 Land resources are crucial elements in the environmental and socio-economic development of mankind. Information and data regarding the types, quantity and spatial distribution of varieties of land resources should be timely and accurate, which serves as the foundation for the formulation and implementation of government policies and project management strategies. Land resources information has already penetrated into many aspects of daily land resources management, including land use, cadastral, land planning and so on, and the volume of it is still expanding. However, the efficiency of land use remains low. The integration using and data sharing of land resources which are distributed at different locations remains difficult, which restrict good use of land resources.There are weaknesses in a traditional distributed Geographic Information System (GIS) technology framework, especially in the sharing and interoperation of massive spatio-temperal data and resources. The development of grid technologies has provided a new method for solving the problems inherent in the traditional distributed GIS. By using grid technologies, hardware, data, and application resources can be pooled together to form a huge, virtual super-computer to effectively get rid of the shortcomings and limitations, which will accelerate the sharing of data and raise the level of land resources utilization.This study presents a grid service model, operating method, grid services architecture and application model of land resources management through combining the study in the fields of GIS, computer and land sciences, which provides theoretical and technical support for the sharing and application of massive spatial-temperal data and resources. It is based on a thorough theoretical research and technical analysis on grid technologies, massive spatio-temperal data management and land use management.Firstly, Land Resources Object (LRO) concept is formulated by applying the concept of resources object to land resource management. Based on LRO concept, a land resources spatio-temperal data service model (LRSTD-SM) about data organization and a multi-level land resource management (ML2RM) method is put forward. Mechanism of LRSTD-SM is to use virtual spatio-temperal data computer to manage the various LRO. The Metadata Model and Resource Organization Model of LRSTD-SM are introduced emphatically. The content of ML2RM is that organization region is divided into multi-partition vertically and the LRO is divided into five units horizontally. The five units are resources object, library, layer collection, layer and feature. Some important technologies of ML2RM are discussed, such as metadata organization, nodes dynamic management, metadata data replication and dynamic combination of resources.Secondly, the Grid Service Model of Land Resources Spatio-Temperal Data(LRSTD-GSM) is developed by applying open grid service architecture to land resourcesmanagement. A grid spatio-temperal database is designed for the storing and managementland resources data, by analyzing the characteristic of land resources management, such as aspatio-temperal partition, logical sheet number and multi-units. Application model for the realization of land resources management, composed of region, tasks, agent, services and data, is developed and implemented based on grid spatio-temperal database.At last, a prototype platform of land use information system based on the model and method mentioned above was established. The test results and case study show that the design of LRSTD-GSM is logical and the new model is effective, efficient, and practical. Finally, a summary of the previous work is given, and future research plan is also pointed out.更多还原