地球化学场精细结构解析方案与应用

为满足勘查地球化学对化探数据处理越来越精细的要求，把人工神经网络和统计分析结合起来形成地球化学场精细结构解析方案。该方案包括对地球化学样品的分类和对各类样品性质的详细研究，目的在于提供便捷的数学和计算机软件工具，以提取化探数据中的深层信息。可以用于：①研究各组样品中元素含量水平和组合特征以解释可能的矿种；②研究元素组合特征与指示元素的关联和区别以揭示可能的成矿作用过程；③研究异常样品的空间组合属性以揭示地球化学场的精细结构；④研究异常样品之间的差异性以缩小找矿靶区；⑤综合分析以确定剥蚀深度和找矿前景。该方案也适用于环境地球化学及矿产资源预测研究。     

In-plane and anti-plane strong shaking of soil systems and structures

The concept of in-plane and anti-plane shaking is introduced with a rigid block on a plane surface with Coulomb friction. Using a hypoplastic constitutive relation to model the mechanical behaviour of the soil, numerical solutions for a rigid block on a thin dry or saturated soil layer are obtained. The coupled nature of dynamic problems involving granular materials is shown, i.e. the motion of the block changes the soil state—skeleton stresses and density—which in turn affects the block motion. Motions of the block as well as soil response can be more realistically calculated by the new model. The same constitutive equation is applied to the numerical simulation of the propagation of plane waves in homogeneous and layered level soil deposits induced by a wave coming from below. Experiments with a novel laminar shake box as well as real seismic records from well-documented sites during strong earthquakes are used to verify the adequacy of the hypoplasticity-based numerical model for the prediction of soil response during strong earthquakes. The response of a homogeneous earth dam subjected to in-plane and anti-plane shaking is investigated numerically. In-plane and anti-plane shaking is shown to cause nearly the same spreading of a sand dam under drained conditions, whereas under undrained conditions anti-plane shaking causes stronger spreading of the dam. The dynamic behaviour of a breakwater founded on rockfill and soft clay during the 1995 Kobe earthquake is back-calculated to show the good performance of the proposed numerical model also with a structure. Section 9 deals with buildings on mattresses of densified cohesionless soils or fine-grained soils with granular columns, slopes with ‘hidden’ dams and structures on piles traversing clayey slopes to show the suitability of hypoplasticity-based models for the earthquake-resistant design and safety assessment of geotechnical systems.     

人工神经网络方法预报长江上游流域面雨量的探讨

利用1998～2000年3～11月长江上游六大流域逐日降水实况和MAPS、T106(T213)、欧洲中心等数值产品资料，采用人工神经网络方法，建立了六大流域强降水面雨量等级预报模型。并于2002年6～9月进行了业务试验。     

人工神经网络在岩体初始应力场反演中的应用

岩土工程问题的有限元分析一般需要考虑初始地应力场，而可供工程应用的初始应力值，主要靠少数实测资料。根据少数测点值，采用人工神经网络与有限元相结合的方法求解岩体初始应力场。通过算例说明：文中所提方法是合理的。     

变形监测基准点的稳定性检验

基准点的稳定性评价，是变形观测数据处理时不可忽视的重要内容。该文对基准点构网和不构网2种情况时的数据处理方法进行了探讨。     

人工神经网络在岩体质量分级中的应用

结合四川省金沙江某水电站工程实例，应用BP人工神经网络方法建立3层BP网络模型，选取岩石单轴抗压强度等6个影响因素为输入变量，对坝基复杂岩体进行质量分级。通过机算机Visual C 语言编程实现神经网络模型，进行网络的学习和运算。以神经网络合理结构分析方法选取合理结构，确定合理隐层单元的数量，提高网络测试的精度。对测试结果的分析发现，经过优化的BP网络模型经多次学习后，测试精度提高，结果可靠，取得较好的实际应用效果。     

利用VB编程完成基点网联测中的各项计算

整套重力计算程序包括了除“地改”以外的所有项目的计算，在这篇文章中我们先仅介绍了“测区参数初始化”及“基点网联测”(采用三程小循环方式)的计算方法。“测区参数初始化”功能可生成gdw．csh初始化文件(该套程序的各项功能都要读取该文件中的相关参量)；“基点网联测”功能可计算基点网联测的各项精度，对联测进行平差，求出各分基点绝对重力值。     

湖北省土壤侵蚀景观空间异质性及驱动因子分析

地域分异是地球表层大小不等、内部具有一定相似性地段之间的相互分化以及由此产生的差异。为了研究不同区位土壤侵蚀问题，从土壤生态景观及系统论出发，运用地质学、地理学、景观生态学、环境学的理论和研究方法，研究湖北省土壤侵蚀景观空间格局及其驱动因子，使土壤侵蚀问题研究提高到一个新的水平。湖北省土壤侵蚀景观具有南北分带、东西分区，为一不对称的断块一环组合，土壤流呈现向长江、江汉盆地中心轴带辐聚、单流向特点。景观空间异质性形成的首要驱动因子是大地构造背景，以房县一襄樊一广济断裂带为界，南北两侧地壳物质组成和构造发展史存在较明显的差异，现代气候带、降雨量、温热程度及土地利用等差异，造成了湖北省区域土壤地理、土壤生态的分异，形成湖北省土壤生态带、区具有南北分带，东西分区的宏观格局；其次大兴安岭一武陵山深部构造陡变带两侧新构造运动强度差异、大别造山带构造强烈隆升，导致土壤侵蚀强度的西强东弱、南北强中间弱的态势；成土母岩差异性决定了土壤可蚀性的多变；空间上“土壤侵蚀内城区”分布在湖北省的周边地区，经济贫困、管理落后，这一地区的经济水平与水土流失间形成“自反馈作用”，这一现象在我国水土保持、生态建设工作中应该引起重视。     

增量位移反分析在水电地下洞室工程中的应用

反分析是确定计算模型参数的有效方法.通常多采用量测所得全量位移进行反演计算.但地下工程中许多量测数据为增量位移,且实际施工过程可以通过建立分步开挖的有限元模型来模拟.据此,结合某水电站地下洞室工程中地下厂房的开挖,建立了模拟动态施工的有限元模型,利用某一开挖步施工前后量测值之差,采用增量位移优化反分析方法对洞室附近初始地应力场及围岩弹性模量进行了反演.计算所得增量位移与实测值符合较好,表明了这种方法的可行性.同时,根据分析结果对该方法进行了评价.     

Determination and application of the weights for landslide susceptibility mapping using an artificial neural network

The purpose of this study is the development, application, and assessment of probability and artificial neural network methods for assessing landslide susceptibility in a chosen study area. As the basic analysis tool, a Geographic Information System (GIS) was used for spatial data management and manipulation. Landslide locations and landslide-related factors such as slope, curvature, soil texture, soil drainage, effective thickness, wood type, and wood diameter were used for analyzing landslide susceptibility. A probability method was used for calculating the rating of the relative importance of each factor class to landslide occurrence. For calculating the weight of the relative importance of each factor to landslide occurrence, an artificial neural network method was developed. Using these methods, the landslide susceptibility index (LSI) was calculated using the rating and weight, and a landslide susceptibility map was produced using the index. The results of the landslide susceptibility analysis, with and without weights, were confirmed from comparison with the landslide location data. The comparison result with weighting was better than the results without weighting. The calculated weight and rating can be used to landslide susceptibility mapping.