地理空间信息网格高性能调度技术中应用程序调度模型的研究

地理空间信息网格调度技术,要比传统的高性能计算中的调度技术复杂,原因是如果将全部网格资源作为一个应用程序的调度和执行目标,必将导致通信延迟、成本昂贵、执行低效等。为此,综合考虑应用程序特性、机器特性等,研究设计了地理空间信息网格高性能调度技术中的应用程序调度模型,包括地理空间信息网格应用程序分析;资源特性分析;应用程序分解;性能预测;资源调度;机器选择;任务映射和任务调度;任务调度器和调度器管理等九个模块。以实现为不同的应用程序匹配不同的计算资源,提高计算资源的利用率和应用程序的执行效率。     

高精度遥感三维可视化在岩溶地区工程初勘调查中的应用——以云南小哨机场为例

本文对高精度卫星遥感图像(QucikBird)进行了图像预处理,图像融合,正射校正,并基于12∶000地形图上构建了DEM。在Erdas软件支持下,实现了遥感影像三维可视化。同时,针对岩溶地区工程初勘需要,应用ArcGIS实现了工程量计算、机场净空分析、岩溶漏斗信息自动提取、漏斗形态显示、地下溶洞网分析等,这为机场的进一步勘查提供了准确可靠的地质灾害信息,可为指导工程设计与施工提供参考。     

Remote sensing and GIS based determination of groundwater dependent ecosystems in the Western Cape, South Africa

Finding the location of groundwater dependent ecosystems (GDEs) is important in determining the extent of restrictions that need to be placed upon the abstraction of groundwater. Remote sensing was combined with geographical information system (GIS) modelling to produce a GDE probability rating map for the Sandveld region, South Africa. Landsat TM imagery identified the areas indicating the probable presence of GDEs and GIS assisted in their delineation. Three GIS models were generated: a GIS model predicting landscape wetness potential (LWP model) based on terrain morphological features; the LWP model was modified to highlight groundwater generated landscape wetness potential (the resulting GglWP model); and a groundwater elevation model was interpolated, combining groundwater level measurements in boreholes in the region with digital elevation model data. Biomass indicators generated from Landsat were classified and combined with the GIS models, followed by field verification of riverine and wetland GDEs. The LWP model provided the most accurate results of the three models tested for GDEs in this region.

基于自适应时序剖分与KNN的短时交通流量预测

在智能交通系统中,准确和高效的短时交通流量预测是交通诱导、管理和控制的前提。由于交通流量动态变化中表现出的时变性和非平稳性特征,其预测难度较大,是交通领域中亟待解决的难题。为提高短时交通流量的预测精度,本文设计与实现了基于自适应时序剖分与KNN（A-TS-KNN）的短时交通流量预测算法。① 基于动态时间规整（Dynamic Time Warping,DTW）动态剖分单日时序为不同的交通模式;② 在不同交通模式,采用互信息法求解每个预测时刻时间延迟的最大阈值,构造不同时间延迟的状态向量,生成交通流量历史数据库;③ 采用十次十折交叉验证的方法求解每个时刻不同时间延迟与不同K值的正交误差结果分布,提取误差最小的正交结果,得到自适应时间延迟与K值的参数组合;④ 采用K个最相似的近邻的距离倒数加权值作为预测结果。对比K近邻（K-nearest neighbors, KNN）、支持向量回归（Support vector regression,SVR）、长短期记忆神经网络（Long-short term memory neural network,LSTM）以及门控递归单元神经网络（Gate recurrent unit neural network,GRU）共4种主流预测模型,A-TS-KNN算法预测精度显著提升;将A-TS-KNN算法用于福州市城市路网中其他交叉路口的短时交通流量预测,结果表现出良好的泛化能力。     

A tentative discussion on the method of comprehensive integration of geographic information

lintroductionGeographicinformahonSystem(GIS)hasbeendevelopedtosuchadegreethatitseemstobeakindofpanaceawhengeographicproblemsarediscussedwhereverinPapers,onclasses,inrePOrtSorinresearchplans.AtpresentitisasthoughthatgeographywasnotgeographywithoutmentioningGIS.Itremindsusofquanhtahvegeography1flatoncebroughtanewlookforgeogr'aphicresearchfromthe1960stothe1980s.DuringthatPeriod,quanhtahvegeographyhadthesimilargoodfortUnetoGIStodayasitwasfrequenhyappliedindiscussingconcernedgeognphicproblems…     

滇东北山区坡耕地水土流失与可持续利用图的基本内容与编制方法

坡耕地水土流失与可持续利用图是综合表示滇东北山区各个坡耕地单元的水土流失状况及其可持续利用的专题地图。该图表示的基本内容是坡耕地类型、土壤类型、坡度级别、侵蚀强度、可持续利用程度、合理利用方向与措施。以土壤流失方程为基础,应用地理信息系统编制了该图。该图的编制为治理本区坡耕地水土流失、合理利用坡耕地资源提供了科学依据。     

Equity vs. Efficiency: Public Capital Investment in Ohio, 1988–1992

Recent research on the role of public capital in the economy focuses primarily on assessing its economic, and sometimes spatial economic, impacts. Access to more detailed and disaggregate data than typically used in these analyses allows us to take a fresh perspective on the often conflicting goals of interregional equity and aggregate efficiency. Using the state of Ohio as a case study, and classic definitions of equity and efficiency, we assess the correspondence between distributions of infrastructure investment and the social/economic distress they are intended to alleviate. Traditional map and statistical analysis combined with a graphical device we call the variegated distribution plot reveals that, in both rates and levels, investment is highest in areas of greatest distress. Both patterns are consistent with equity-driven investment distributions.     

基于GIS与ANN耦合技术的非线性矿产预测系统设计

利用地理信息系统(GIS)和人工神经网络(ANN)相结合,提出的基于GIS的人工神经网络矿产预测系统设计方案,实现了从地质变量优选到人工神经网络成矿预测结果图形显示的计算机自动化处理。其中的主要模块为:地质变量优选模块、数据库管理模块、神经网络成矿预测模块和图形处理功能模块。它们之间通过数据库进行数据传输与共享达到互相连接。     

海洋划界法理与领海基点信息系统建立

以研究《联合国海洋法公约》及相关法律文书为出发点,探讨海洋划界理论与技术,分析我国领海现状,在此基础上提出领海基点信息管理与应用,建立领海基点信息数据库,进而实际制作领海基点信息系统。