非线性理论在开采沉陷中的应用

简要介绍了近年来把非线性理论应用于矿山开采沉陷领域的一些研究成果,主要包括两大部分：（１）神经网络理论在开采沉陷中的应用;（２）离散单元法在开采沉陷中的应用     

大气环境质量综合评价的B－P网络模型

选取大气环境质量的三级标准作为３个学习样本，以ＳＯ＿２，ＮＯ＿ｘ和ＴＳＰ三种污染物的三级标准浓度限值作为样本的输入特征值，建立了三参数的Ｂ－Ｐ网络大气环境质量综合评价模型，用该模型对３１个测点的大气环境质量进行评价，并与模糊数学和灰色系统法的评价结果比较，表明Ｂ－Ｐ网络用于大气环境质量评价具有优越性。     

我校图书馆自动化建设的实践与思考

本文简述了我校图书馆自动化建设的历史进程，分析了图书馆自动化建设的现状、达到的效果和尚存在的问题，并针对这些问题，提出了相应的改进措施。     

北京市数字遥测地震台网技术系统

地球科学是一门观测科学,地震观测是研究与应用工作的基础。因此,布设科学合理、运行稳定可靠的观测台网,可更好地为地震监测预报、地震应急以及科学研究提供保障。本文较系统的介绍了北京市数字遥测地震台网,并就系统建设过程中的一些技术问题进行了论述。     

人工神经网络模型在石油资源预测中的应用

锦270井区大凌河油层是盆地大幅度沉降条件下的产物，以深陷湖环境为主，发育湖底扇浊积岩。准确预测大凌河油层的砂体分布和泥质含量，是勘探目标选择的关键。由于钻井资料少，不能反映储层横向变化。为了准确预测储层的横向变化，综合钻井地质资料和地震勘探资料，并为了反映地震资料的多个参数与储层横向变化间的非线性相关关系，采用人工神经网络模型进行预测，计算了大凌河油层砂层厚度和泥质含量的平面分布。依据计算结果，分析了有利油气区的分布，并给出了几点结论。     

Advanced nitrogen removal using pilot-scale SBR with intelligent control system built on three layer network

Since eutrophication has become increasingly severe in China, nitrogen and phosphorous have been the concern of wastewater treatment, especially nitrogen removal. The stabilization of the intelligent control system and nitrogen removal efficiency were investigated in a pilot-scale aerobic-anoxic sequencing batch reactor (SBR) with a treatment capacity of 60 m³/d. Characteristic points on the profiles of dissolved oxygen (DO), pH, and oxidation reduction potential (ORP) could exactly reflect the process of nitrification and denitrification. Using the intelligent control system not only could save energy, but also could achieve advanced nitrogen removal. Applying the control strategy water quality of the effluent could stably meet the national first discharge standard during experiment of 10 months. Even at low temperature (t = 13°C), chemical oxygen demand (COD) and total nitrogen (TN) in the effluent were under 50 and 5 mg/L, respectively. Translated from Acta Scientise Circumstantiae, 2006, 26(5): 745–750 [译自: 环境科学学报]     

内蒙古自治区生态环境监测网络站建设方案

本文通过对自治区生态环境状况和现有工作能力的充分分析 ,阐述了建立自治区生态环境监测网络站的必要性、意义和可行性。提出了网络站建设目标和建设内容 ,较详尽地阐述了网络站的监测任务、监测指标、工作职责、内部机构设置、人员素质要求以及仪器设备配置和运行需求等内容。     

Assessing vulnerabilities to the effects of global change: an eight step approach

In the recent years, global environmental change research has seen increased attention to the concept of vulnerability. There have been a growing number of vulnerability assessments, but relatively little discussion on appropriate and common methods. Here we propose a method to guide vulnerability assessments of coupled human–environment systems toward a common objective: informing the decision-making of specific stakeholders about options for adapting to the effects of global change. We suggest five criteria vulnerability assessments must at least possess to achieve this objective. They should have a knowledge base from various disciplines and stakeholder participation, be place based, consider multiple interacting stresses, examine differential adaptive capacity, and be prospective as well as historical. On the basis of these criteria, we present a general methodological guideline of eight steps. To examine whether these eight steps, if attentively coordinated, do in fact achieve the criteria, and in turn satisfy the objective of the assessment, we discuss two case studies. We expect most readers to identify some of the steps as part of their well-established disciplinary practices. However, they should also identify one or more steps as uncommon to their research traditions. Thus taken together the eight steps constitute a novel methodological framework. We hypothesize that if researchers employ this framework, then the products of the research will (1) achieve the objective of preparing stakeholders for the effects of global change on a site-specific basis, and (2) further the “public good” of additional insights through cross-study comparisons of research projects designed according to common principles.     

分流制雨水管网沉积物中典型抗生素抗性基因污染特性

针对城市雨水管网沉积物中抗生素抗性基因（ARGs）污染特性尚不清楚的问题，本研究选择南京江北新区3种不同功能区（文教区、交通区及商业区）分流制雨水管网沉积物作为研究对象，考察了各功能区雨水管网沉积物中细菌耐药性，分析了β-内酰胺类和四环素类两类典型ARGs的组成种类及其相对丰度差异，重点探讨了上述典型ARGs在不同粒径颗粒中的分布特征.结果表明，各功能区可培养细菌总数大小顺序为商业区 > 文教区 > 交通区，而携带典型ARGs的耐药菌（ARB）相对含量大小顺序却为文教区 > 商业区 > 交通区；交通区的典型ARGs相对丰度高于文教区和商业区1个数量级；随着干期长度的增加，各功能区雨水管网沉积物中典型ARGs种类及相对丰度均有不同程度的减少；随着颗粒粒径的减小，文教区雨水管道沉积物中典型ARGs的丰度变化不大，交通区的雨水管道沉积物中典型ARGs的丰度则呈增加趋势，而商业区的无明显规律；粒径较小的颗粒中可移动遗传因子（MGEs）对ARGs的分布影响更大，且以<75 μm粒径段颗粒中的最明显（相关系数为0.874）.     

基于人工神经网络方法的水质预测初探

常规的水质预测模型因存在许多简化与假定而限制了其精度与实用性的提高.文章通过引入神经网络技术来建立水质预测模型,分别采用LM算法和RBF算法来提高预测的精度.结果表明,在建立三门峡水库流量和水质的输入响应关系模型的实际应用中,RBF算法取得了较好的预测效果.