水旱灾害发展趋势的灰色拓扑预测与应用

根据青岛市 65 a的降水量序列 ,利用灰色拓扑预测方法 ,建立了降水量指标的灰色预测模型群 ,并对未来降水过程及水旱灾害发展趋势进行了预测。     

The Properties of Topological Network Connectivity Measures and Their Application to U.S. Urban Freeway Networks

The mathematical concepts of graph theory were introduced into geography in the early 1960s, providing a means of conceptualizing transport networks as made up of nodes and links. Among the contributions were several measures of topological network connectivity known as α (alpha), β (beta), and γ (gamma). These ratio measures identified the level of connectivity present in a network. These measures attracted considerable attention before fading from view, although they have recently returned to prominence in several applications. Despite this renewed attention, the basic mathematical properties of these measures remain largely unexamined. This research investigates the properties of these three simple ratio measures and explores values and properties that might be typical in urban freeway networks. Although simple correlation reveals little relationship between connectivity and freeway structure, they can be used to identify several groups of freeway networks as well as identifying important details about the use and interpretation of these measures. It is likely that these simple measures will continue to be applied to new situations, and this article provides some guidelines for those applications.     

基于拓扑构建的多边形区域间运算的新方法

提出了一种新的直观的方法进行多边形区域之间的运算。首先将需要计算的多边形区域的边进行自动拓扑构建,利用多边形区域的边将平面划分为n个小多边形区域;然后生成这些多边形区域的内点,通过判断小多边形区域的内点是否在原始多边形区域内来确定小多边形区域是否选取;最后合并选取的小多边形即为所求。试验结果表明,该方法思路清晰、鲁棒性强,在GIS中得到了有效的运用。     

三维GIS场景与多路视频融合的对象快速确定法

虚实融合技术是视频增强三维虚拟地理场景的关键环节之一,对提高真实沉浸式视觉体验具有重要作用。现有基于视频投影的虚实融合方法,在三维场景动态更新时,通常需要对用户视角下新场景内所有渲染对象进行遍历,来筛选出摄像头范围内需融合的对象,导致融合过程耗时长、效率不高。本文提出一种顾及拓扑关系的多路视频与三维GIS场景融合的对象快速确定方法。首先对三维场景视锥体内视频对象集渲染对象动态建立拓扑网络模型;然后基于拓扑网络模型建立视频与渲染对象的拓扑关系,完成融合对象的快速确定及融合渲染;最后利用滕州市三维场景进行了验证。试验结果表明,本文提出的方法在视频与三维GIS场景融合时,整体耗时少、效率高,极大地降低了融合的计算代价,具有良好的可行性及稳健性。     

基于结点逼近提取的平面点集Voronoi图构建算法

提出一种基于结点空间逼近、精确提取以及面向拓扑关系生成的2维平面点集的构建方法。主要给出了搜索矩形域及其剖分概念、Voronoi图的基本性质、矩形域与Voronoi图结点关系的定理及其证明、基于链队的矩形域剖分和结点逼近机制及结点提取策略、基于条带有序表的最近邻近发生元快速检索算法、矢量Voronoi图的拓扑关系建立算法等。经过算法分析和程序试验验证本文算法的时间复杂度为0（n log2 n）,本方法可以扩展到平面任意发生元Voronoi图的构建,具有简洁、高精度、鲁棒性、高效、适合于海量数据等特点,并且具有较好的实用价值和应用前景。     

一种改进的自动联结三角网算法

针对自动联结三角网算法的缺点,对该算法进行了改进。改进的算法提出了先建立分区索引后构建三角网的思想,首先将大规模的离散高程点进行分区索引,然后根据实时构建三角形与各分区的关系对候选高程点进行判断。最后用改进的自动联结三角网算法进行了实验,取得了明显的改进效果。     

Tunnel reinforcement via topology optimization

Anchoring is a fundamental method for supporting tunnels. It reinforces the original rock mass and reduces the deformation along the tunnel surface. The topological complexity of its layouts renders a design methodology difficult. A numerical approach to reinforce the host ground becomes desirable. The present paper proposes a topology optimization method based on a two‐phase cell model and finite‐element discretization of the host ground. The element consists of the original rock and the reinforcing material. The design issue involves the distribution of the reinforcing materials. The relative ratios of the two phases in various elements will be optimized to reduce the compliance of the tunnel. The method enables the computer‐aided design for the support of underground structures. The capabilities of the method are demonstrated by the designs to support a deep tunnel under various in situ stresses. The results indicate that oriented reinforcement is needed along the direction of the largest absolute value of the principal stress. Copyright © 2000 John Wiley & Sons, Ltd.     

Topology optimization to prevent tunnel heaves under different stress biaxialities

Floor or sidewall heave imposes a severe issue in tunnel construction. We propose a two‐phase mixture model consisting of the original rock and the fully anchored rock mass. The host ground is discretized by finite elements. The relative ratios of two phases in various elements are optimized by optimality criteria method so that the floor or sidewall heave is minimized. The topology of reinforcement is designed to prevent floor heave or sidewall heave of a tunnel surrounded by homogeneous rock. The stress biaxiality k dictates whether the floor heave or the sidewall heave control the tunnel distortion. The results indicate the importance to set reinforcement near the tunnel corners to prevent sidewall and floor heaves. Part of the results is confirmed in the practices. Copyright © 2000 John Wiley & Sons, Ltd.     

人工神经网络在区域地质调查中的应用--反向传播模型(BP网)建立与应用

类比与识别是１∶５ 万区调工作中应用的基本方法之一。把人工神经网络技术———反向传播模型（ＢＰ网） 应用于这些基本方法中,不但有助于区调技术人员使脑科学朝向定量、精确和理论化方向发展,而且有助于我们对神经网络及其成为一种崭新的信息处理方法的理解,从而提高１∶５ 万区调理论研究水平。