Visualization of Volcanic Rock Geochemical Data and Classification with Artificial Neural Networks

An unsupervised neural network technique, Growing Cell Structures (GCS) was used to visualize geochemical differences between four different island arc volcanic rock types: basalts, andesites, dacites and rhyolites. The output of the method shows the cluster structure of the dataset clearly, and the relevant geochemical patterns and relationships between its variables. The data can be separated into four clusters, each associated with a specific volcanic rock type (basalt, andesite, dacite and rhyolite), according to a unique combination of major element concentrations. Following clustering, performance of the trained GCS network as a classifier of volcanic rock type was tested using two test datasets with major element concentration data for 312 and 496 island arc volcanic rock samples of known volcanic type. Preliminary classification results are promising. In the first test dataset 94% of basalts, 76% of andesites, 83% of dacites and 100% of the rhyolites were classified correctly. Successful classification rates in the second dataset were 100%, 80%, 77%, and 98% respectively. The success of the analysis suggests that neural networks analysis constitutes a useful analytical tool for identification of natural clusters and examination of the relationships between numeric variables in large datasets, and that can be used for automatic classification of new data.     

Concept for a Web-based Information System for Flood Management

Recent emergency flood situations at European rivers have revealed the demand for better and in-time information for citizens in flood prone areas about flood development, as well as better coordination of resources and actions during pre-flood phases and its critical stage. Information and Communication Technology (ICT) has a large potential to improve the situation. Decisions may be supported by information about resources available at the region and national level, by information about means and access to critical locations at the prevention as well as the evacuation phases, and by including citizens as well as managers into one common information and communication process. The paper outlines the potential of ICT for these aspects.     

嵌入式导航终端互补分级路网拓扑模型的研究及实现

路径分析是嵌入式导航终端的基本功能。在层次空间数据模型的基础上,本文提出了互补分级数据模型,将整个路网拓扑分为概括拓扑和辅助拓扑两级,并给出两级路网拓扑数据快速合并机制形成详细路网拓扑,使得嵌入式硬件平台上远距离路径分析效率更高。     

Optimization strategies for sediment reduction practices on roads in steep,forested terrain

Many forested steeplands in the western United States display a legacy of disturbances due to timber harvest, mining or wildfires, for example. Such disturbances have caused accelerated hillslope erosion, leading to increased sedimentation in fish‐bearing streams. Several restoration techniques have been implemented to address these problems in mountain catchments, many of which involve the removal of abandoned roads and re‐establishing drainage networks across road prisms. With limited restoration funds to be applied across large catchments, land managers are faced with deciding which areas and problems should be treated first, and by which technique, in order to design the most effective and cost‐effective sediment reduction strategy. Currently most restoration is conducted on a site‐specific scale according to uniform treatment policies. To create catchment‐scale policies for restoration, we developed two optimization models – dynamic programming and genetic algorithms – to determine the most cost‐effective treatment level for roads and stream crossings in a pilot study basin with approximately 700 road segments and crossings. These models considered the trade‐offs between the cost and effectiveness of different restoration strategies to minimize the predicted erosion from all forest roads within a catchment, while meeting a specified budget constraint. The optimal sediment reduction strategies developed by these models performed much better than two strategies of uniform erosion control which are commonly applied to road erosion problems by land managers, with sediment savings increased by an additional 48 to 80 per cent. These optimization models can be used to formulate the most cost‐effective restoration policy for sediment reduction on a catchment scale. Thus, cost savings can be applied to further restoration work within the catchment. Nevertheless, the models are based on erosion rates measured on past restoration sites, and need to be updated as additional monitoring studies evaluate long‐term basin response to erosion control treatments. Copyright © 2006 John Wiley & Sons, Ltd.     

基于遗传算法的道路网综合模型

提出了一种基于遗传算法的道路网综合模型。实验表明，该模型在保持道路网的空间分布性和实用性方面具有庭好的效果。     

最短路径算法用于城市公交出行路径最优化

介绍了经典Dijk st ra算法的主要思想快速实现方法，指出了公交网络不同于一般的道路网络的特点，给出了选择公交出行路径的算法思想。     

基于模态分析和神经网络的裂缝损伤识别

提出了裂缝损伤诊断的神经网络方法,探讨了用模态技术和神经网络对混凝土结构裂缝损伤进行识别与定位的方法。文中以一简支矩形截面梁为研究对象,通过完好结构和损伤结构的有限元分析,获取两者的损伤标识量,输入BP神经网络训练。以损伤位置和裂缝高度作为输出参数,对其进行单处损伤定位的研究。数值仿真结果表明,采用神经网络方法可以对裂缝做出较好的诊断。     

自适应抗差Kalman滤波在公路GPS测量中的应用

本文介绍了用载波相位平滑伪距观测数据和自适应抗差Kalman滤波算法进行差分动态定位的方法，用试验数据说明了车载GPS道路测量系统所能达到的精度水平。     

Interface Roughness by Optical Coherence MicroscopyZHU Yong-kai~(1),LI Ji-nian~(1)

IntroductionIf let roadway entities be denoted by arcs and in-tersections be denoted by nodes, then a road net-work can be represented with node-arc model. Be-cause the designation of nodes can be determinedby arcs according to the definition of node-arc …     

用人工神经网络模型获取采油指数的方法

介绍了油藏工程中一种新的神经计算技术的实现方法——用人工神经网络模型获取开发初期分采时的采油指数;通过实例对该方法中的数据处理、模型确定及精度分析、预测效果进行了详尽地描述,实践表明该方法可对从事油田开发的现场研究人员提供借鉴。