【作者】 马立山；

【导师】 陶建华；

【作者基本信息】 天津大学 ， 流体力学， 2011， 博士

【摘要】 我国正处于城市化深入发展的时期,城市化使城市规模不断扩大,促进了经济繁荣,但同时也对城市水资源和环境产生了一些负面影响,如水资源短缺,水环境污染和雨洪灾害等。城市雨洪系统是复杂系统,不同地域有着不同的特征,城市功能定位不同对雨洪研究的要求不同,需要综合各方面的因素,采用有效的方法开展研究。现代信息技术的发展使水环境空间数据获取范围、数据量都在飞快地增长,对水环境问题的认识和推动相关数学模型的发展提供了良好的条件。将现代信息技术与数值模拟、实验和监测手段相结合产生的水信息学和水信息技术成为城市水环境研究的有效手段。本文以我国北方某干旱少雨城镇为背景,将水信息技术应用到雨洪系统研究中,提出了城市雨洪空间信息系统(Urban Stormwater Spatial Information System,USSIS)的基本构架;确定了USSIS的功能框架;以GIS为平台,建立了USSIS的体系结构。在GIS平台上,以地理数据库(Geodatabase)为基础,利用面向对象技术和统一建模语言(UML)建立了城市雨洪空间信息系统数据库,对城市雨洪系统各环节的要素及其相互关系进行建模,实现了多源数据融合,为城市雨洪系统研究建立知识基础。利用数字高程模型(DEM)将研究自然流域水系的方法用于城市道路与排水管网中,确定“排水路线”,划分城市汇水区域,提取城市地表水文特征参数,为城市雨洪系统计算和分析运行提供支撑。信息科学中的软计算方法差异演化算法(DE)与人工神经网络(ANN)结合,改进了人工神经网络初始联结权重的确定方式。基于对城市地下水位及水质的预测,对DE-BP网络结构和影响因素进行了分析,特别对训练样本和预测样本相关性对预测结果的影响进行了分析,结果表明样本相关性与预测精度不存在明显正相关,而网络结构和样本代表性则有重要影响。在GIS平台上,实现了对城市雨洪系统的模拟计算。本文建立的城市雨洪空间信息系统,可为城市规划设计、雨洪资源合理利用、防洪减灾、决策支持和数字化城市建设等提供基础支撑。更多还原

【Abstract】 China is experiencing a rapid process of urbanization, which results in increasingly severe water environment problems, such as water resource shortage, water environment depravation, flood disaster and so on. The urban water environment is a complicated system. According to the different urban function orientation, the different effective research methods need to be adopted based on an overall consideration of the factors. With the wide application of modern technologies, both the quantity and the quality of the available data increase rapidly. Benefit from the high quality and huge number of data, the urnderstanding of the water environment is increasing and this provides a good opportunity for the development of the mathematical models. Hydroinformatics, which combines the modern information technologies, traditional numerical simulation, experiment and monitoring methods has being becoming a effective method for the study of urban water environment.Take a small town in the arid region in north China for instance, this thesis applied hydroinformatics and the related technologies to develop the Urban Stormwater Spatial Information System (USSIS) and established the frame and functions of the USSIS based on the Geographic Information System (GIS). The USSIS is a multivariate data fusion system, with the integration of database and models.Based on the Geodatabase, the USSIS takes the advantage of object oriented technology and build a logical model of urban stormwater and the relative elements with unified modeling language (UML). The multivariate data fusion was realized using the advanced spatial data models, and the knowledge foundation of the urban stormwater system was established. This paper presents an application of DEM, in which the methods used in natural river basin system were applied to urban road and wastewater pipelines. This method was used to determe“drainage route”, to divide urban runoff catchments, to extract city surface hydrological features parameters and to provide support for calculation and analysis of urban stormwater system.Soft computation is an important part of the information science system. This thesis uses differential evolution(DE)algorithm to determine the initial connection weights for artificial neural networks(ANN)to improve the computing result. Based on prediction of urban ground water level and water quality, the network structure and factors affecting the performance of the DE-BP method are analyzed, especially the effect of the correlation between training samples and prediction samples on the prediction result. The analysis shows that there is not a positive correlation between sample correlation and prediction accuracy, while the network structure and sample representativeness are significant important.Based on the GIS, the SWMM method and database technology is used together in urban runoff simulation, and the flooding area is determined by non-source flooding method.The urban stormwater spatial information system(USSIS) developed in this thesis can be used as an effective tool for urban planning, rainwater harvesting, flood mitigation, and it is also an important part of decision support system and digital urban system.更多还原