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。     

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.     

概率神经网络与BP网络模型在遥感图像分类中的对比研究

通过分析概率神经网络(以下称PNN)的基本结构及其训练算法,建立了卫星图像分类的概率神经网络模型,并通过实例对比分析了概率神经网络与BP网络分类模型的分类效果。实验表明,PNN图像分类方法在分类精度上优于误差反向传播神经网络模型,且分类时间相当,是一种有效的图像分类方法。     

GPS高程转换的神经元网络方法分析

GPS高程转换是GPS高程应用的关键问题之一，试图应用神经元网络方法来进行GPS高程的转换，并在网络的拓扑结构与网络结构的改进等方面作了分析，通过GPS高程转换的二次曲面法和神经元网络法进行了比较试验，结果表明了神经元网络转换GPS高程的可行性和可靠性，对于应用神经网络转换GPS高程具有实际的指导意义。     

基于图论的树状河系结构化绘制模型研究

结合树状河系自身的结构特点和图论的思想，提出了基于图论的河系结构化绘制模型的建立，利用图论的思想描述了河系的结构并建立其自动符号化模型，阐述了主流、流向自动确定和结构化绘制实现的详细算法，并对通用图形数据格式DXF文件记录的河系数据进行了测试。     

IGGOS as a potential partner in IGOS

This short note reviews our thinking on how IGGOS can best achieve a high status within the set of global monitoring programmes. If such a high status can be obtained, then the importance of geodetic networks and services will be recognized more widely, and their activities will consequently be better resourced in the long term. One particular aspect concerns how IGGOS can complement the roles of the various IGOS partners within global monitoring. The different ways in which IGGOS can contribute to IGOS are outlined.     

The International GPS Service tracking network: Enabling diverse studies and projects through international cooperation

The International GPS Service (IGS), formulated beginning in 1989 and formalized in 1994, was founded on the collaborative operation of approximately 30 permanent GPS stations to benefit global geodynamics. The same cooperative principles, today applied to a network of over 300 stations, still serve to maximize global benefit without unnecessary duplication of investment in global infrastructure. The scope of applications of the dataset has grown to include atmospheric, oceanographic, subdaily, and low-earth orbiter activities through working groups and pilot projects fostered within the IGS in the now traditional IGS spirit of collaboration. These activities and the IGS infrastructure are viewed as critical elements to the Global Geodetic Observing System. This presentation will review the present nature of the IGS tracking network and its ability to support new applications.     

Assessing the Comparability of Ammonium,Nitrate and Sulfate Concentrations Measured by Three Air Quality Monitoring Networks

Airborne fine particulate matter across the United States is monitored by different networks, the three prevalent ones presently being the Clean Air Status and Trend Network (CASTNet), the Interagency Monitoring of PROtected Visual Environment Network (IMPROVE) and the Speciation and Trend Network (STN). If combined, these three networks provide speciated fine particulate data at several hundred locations throughout the United States. Yet, differences in sampling protocols and samples handling may not allow their joint use. With these concerns in mind, the objective of this study is to assess the spatial and temporal comparability of the sulfate, nitrate and ammonium concentrations reported by each of these networks. One of the major differences between networks is the sampling frequency they adopted. While CASTNet measures pollution levels on seven-day integrated samples, STN and IMPROVE data pertain to 24-hour samples collected every three days. STN and IMPROVE data therefore exhibit considerably more short-term variability than their CASTNet counterpart. We show that, despite their apparent incongruity, averaging the data with a window size of four to six weeks is sufficient to remove the effects of differences in sampling frequency and duration and allow meaningful comparison of the signals reported by the three networks of concern. After averaging, all the sulfate and, to a lesser degree, ammonium concentrations reported are fairly similar. Nitrate concentrations, on the other hand, are still divergent. We speculate that this divergence originates from the different types of filters used to collect particulate nitrate. Finally, using a rotated principal component technique (RPCA), we determined the number and the geographical organization of the significant temporal modes of variation (clusters) detected by each network for the three pollutants of interest. For sulfate and ammonium, the clusters’ geographical boundaries established for each network and the modes of variations within each cluster seem to correspond. RPCA erformed on nitrate concentrations revealed that, for the CASTNet and IMPROVE networks, the modes of variation do not correspond to unified geographical regions but are found more sporadically. For STN, the clustered areas are unified and easily delineable. We conclude that the possibility of jointly using the data collected by CASTNet, IMPROVE and STN has to be weighed pollutant by pollutant. While sulfate and ammonium data show some potential for joint use, at this point, combining the nitrate data from these monitoring networks may not be a judicious choice.     

Interlocking directorships and trans-national linkages within the British Empire, 1900–1930

Recent research has highlighted the value of employing the idea of networks to describe the interrelatedness of today's globalizing economy. Networks facilitate flows of knowledge, ideas, managerial techniques and capital between firms both within and across political borders. This paper argues that the reconstruction of social connections through which information is created, given value and exchanged is fundamental to an understanding of not only contemporary but also historical patterns of economic globalization. We focus on the networks of the capitalist elite running 12 major multinational enterprises, active across British imperial territories, between c .1900 and c .1930. An examination is made of the economic and spatial interlocks between firms created by board members who were multiple directors. Social underpinnings of multiple directorates are examined by exploring the common, overlapping social spheres within which individuals engaged. A clearer grasp of the ways in which corporate activity operated in the early 1900s can provide a better understanding of the social context of global economic operations.     

Prediction of creep characteristic of rock under varying environment

The strain developed due to creep is mainly proportional to the logarithm of the time under load, and is mostly proportional to the stress and temperature. At higher temperature the creep rate falls slowly with respect to time, and the creep strain is proportional to a fractional power of time, with the exponent increasing as the temperature increases and reaching a value approximately one-third at temperatures of about 0.5°C. At these temperatures, the creep increases with stress according to a power greater than unity and possibly exponentially. It increases with temperature as (−U/kT), where U is an activation energy and k is Boltzman’s constant. There are different methods to determine the creep strain and the energy of Jog (B) including experimental methods, multivariate regression analysis, and by numerical simulation. These methods are less cumbersome and time consuming. In the present investigation, artificial neural network technique has been used for prediction of the creep strain and energy of Jog (B). Two different networks have been tested and validated. Both the networks have four input neurons and one hidden layer with five neurons, and one output neuron. The data for different rocks at temperatures up to 750°C under conditions of compressive or tortional stress are taken from the literatures. The training and testing data sets used were 163 and 14, respectively. To deal with the problem of overfitting of data, Bayesian regulation has been used and network is trained with suitable training epochs. The coefficients of correlation among the predicted and observed values are found high and they improve the confidence of the users. The mean absolute percentage error obtained are also very low.