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。     

南秦岭十里坪锑矿床成矿时代及成因的初步研究

十里坪锑矿床受赵川陆缘隆_滑构造的主滑脱拆离带的控制。矿体呈脉状赋存于韧_脆性主滑脱带上部的脆性次级断层_节理中,矿石类型主要为萤石石英辉锑矿型。围岩为太古宙_元古宙变质岩系,围岩蚀变弱。成矿流体属H2O_CO2_NaCl体系,流体包裹体盐度w(NaCleq)为3.6%～11.3%,均一温度为109～232℃,形成压力大致为800×105Pa。硫、铅同位素研究表明,矿质主要来源于变火山岩围岩;氢、氧同位素显示,成矿流体以大气降水为主,初步将该矿床定为变质岩源就地式大气降水热液矿床。矿石中萤石Sm_Nd等时线年龄为(392±24)Ma,与南秦岭北部晚古生代拗陷区热水喷流_沉积成矿时代相一致,它们都形成于秦岭微板块泥盆纪非造山裂解阶段。     

Tectonic and hydrodynamic control of landslides in the northern area of the Central Rif, Morocco

The northern edge of the Central Rif (Morocco) is subject to numerous landslides where mechanisms do not correspond to the classical models used by geomechanics specialists. It is necessary to adopt a multidisciplinary approach that combines geomorphology, geology, hydrogeology, and geotechnics in order to understand how such slope failures are generated, especially in a region with a heterogeneous structure characterised by significant lithological differences, severe fracturing, and thrust sheets where tectonic contacts play a major role in groundwater circulation. This report shows that these failures are essentially controlled by the tectonic contact separating the Tisirene and Chouamat thrust sheets and by subsurface hydrodynamic conditions. A model of spatial and temporal variations in the factor of safety is proposed.     

长江三角洲主要城市产业结构转换的综合评价

文章在对地区产业结构转换影响因素系统分析的基础上，采用定量分析方法，对长江三角洲15个主要城市的产业结构转换能力、产业转换速度、产业转换方向以及它们之间的关系进行探讨。研究表明：长江三角洲主要城市产业结构转换能力的区域差异明显，产业结构转换能力与区域经济发展所处阶段有很大的关系；产业结构转换能力、速度与经济发展水平呈正相关。     

The method of design for seismic data database system based on tree structure

Introduction In recent years, with the development of earthquake observation technology, increasing ofobservation stations and improvement of observation instruments which are digitalized and con-nected by network, a great deal of data is recorded, to some extend, which brings difficulty for usto manage and store them. It leads to traditional methods, which use the file system to managemass data, could not satisfy our requirement, and it is necessary to find a new method in place oftradit…