基于WebGIS的特色林果预警系统关键技术研究

发布时间：2018-01-16 18:37

本文关键词：基于WebGIS的特色林果预警系统关键技术研究　出处：《南京信息工程大学》2014年硕士论文　论文类型：学位论文

【摘要】：WebGIS具有良好的用户体验、数据共享等优势,正越来越广泛地应用于各类灾害预警监测系统中。本文以特色林果农业气象灾害监测预警系统为依托,以灾害监测预警处理流程为主线,围绕基础数据集成、灾害预警方法、预警结果WebGIS建模展示等问题开展研究。提出了基于XML的全局数据视图结构以实现多源气象报文的统一访问；基于线性索引结构解决海量气象报文快速定位问题；提出了基于模糊理论的预警模型,提高了预警结果评判的准确性；并基于GIS空间分析实现了预警结果的基本WebGIS输出。针对三维空间模型建立,提出了基于聚类的动态LOD (Levels of Detail)三维地形建模方法。本文的研究内容与成果主要包括： (1)多源数据集成研究。针对多数据源、多格式的气象报文数据整合问题,提出基于XML的全局数据视图模型,用于数据的统一访问,从而为预警结果评判提供了数据支撑。而线性索引结构的引入解决了海量气象报文的快速定位难题,提高了报文文件的管理效率。 (2)基于模糊理论的预警研究。针对当前结构化预警指标体系中,多种致灾因子的综合评价问题,基于模糊数学的相关理论,建立了模糊集的隶属函数,依据关系合成与最大隶属度原则,对多种气候因子进行综合研判,产生预警结果。在此基础上,基于GIS空间分析方法,形成WebGIS环境下的预警机制。 (3)三维地形建模。通过分析DEM (Digital Elevation Model)高程数据集的空间自相关性,本文基于K－means聚类分析方法,将数据采样点与地貌特征关联在一起。在此基础上,基于地理网格重构的方法对地形进行简化建模,通过实验验证了本方法的可行性。对上述方法进一步优化,依据地形的微观区域地貌特征,实现了多分辨率、动态LOD的地形简化方法,实验说明本方法可以很好的保持地形特征,提高了图形显示效率,为预警结果的立体化展示奠定了基础。在以上研究基础上,结合RIA(富客户端应用)技术的WebGIS设计思想,设计了特色林果原型系统,解决了特色林果预警系统的数据处理、预警评价方法、三维可视化输出等关键问题。
[Abstract]:WebGIS has a good user experience, data sharing and other advantages, is widely used in all kinds of disaster early warning and monitoring system. Based on the characteristics of forestry and agricultural meteorological disaster monitoring and early warning system based on disaster monitoring and early warning process as the main line, on the basis of data integration, disaster warning, warning results to carry out research on WebGIS modeling show the proposed unified access to achieve multi-source meteorological global view of data structure based on XML; rapid positioning of massive meteorological index structure based on linear solution; put forward early-warning model based on fuzzy theory to improve the accuracy of early warning evaluation results; and the GIS spatial analysis to achieve early warning results based on basic WebGIS output for three-dimensional space model, proposed a dynamic clustering based on LOD (Levels of Detail) 3D terrain modeling method.
The main contents and results of this paper are as follows:
(1) research on multi-source data integration. Based on multiple data sources, meteorological data integration of multi format, the global view of data model based on XML for unified access to data, so as to provide data support for the evaluation of the early warning results. And the linear index structure is introduced to solve the problem of fast location of massive meteorological message. To improve the efficiency of management of the message file.
(2) early warning research based on fuzzy theory. In view of the current structure of early-warning index system, comprehensive evaluation of various hazard factors, based on fuzzy mathematics theory, a fuzzy set membership function, according to the relationship between the synthesis and the principle of maximum degree of membership, carry out a comprehensive study of various climate factors, this produced the warning results. Based on GIS spatial analysis method based on the formation of early warning mechanism in WebGIS environment.
(3) 3D terrain modeling. Through the analysis of the DEM (Digital Elevation Model) spatial autocorrelation of elevation data sets, this paper based on K-means clustering analysis method, data sampling points associated with the features together. Based on the method of geographic grid reconstruction based on the terrain model. Through the experiment to verify the feasibility of this method. To further optimize the method, based on micro regional geomorphic features of the terrain, realize multi-resolution terrain simplification method, dynamic LOD, the experimental results show that this method can well maintain topographic features, improve the efficiency of the graphics display, laid the foundation for three-dimensional display of the warning results.
On the basis of the above research, combined with the WebGIS design idea of RIA (rich client application) technology, a prototype system of characteristic forest fruit is designed, which solves the key problems of data processing, early warning evaluation method and 3D visual output of characteristic forest early warning system.

【学位授予单位】：南京信息工程大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：S126;P208

【参考文献】