神经网络在CB油田储层预测和储层厚度计算中的应用

根据ＣＢ油田的特点提出了利用神经网络进行储层预测和计算地层厚度的方法。将传统的储层预测方法与人工神经网络相结合，能较好地解决ＣＢ油田的储层预测问题。利用神经网络的非线性映射特点实现了地震特征与地层厚度之间的映射，从而可以准确地求取地层厚度。用该方法设计了２口探井，实际钻井结果表明，这种方法的预测结果准确、可靠。     

Study on the Application of Standing Collar Pattern

The relations between the lining collar and the face collar in pattern design of standing collar by experiments and theoretical calculation is elucidated. The collar's shape and the fabric thickness, although with the same construction, influence the lining collar and the face collar, especially the back center of the lining and the two collar angles. With a proportional relation, the thicker the fabric is, the more the total variation is. It is supposed to have some practical value to the collar pattern design, as well as to the fashion CAD.     

A Reputation Evaluating Model for Non-Government-Run Scientific Research and Technical Service Organizations

This paper presents a reputation evaluating model for non-government-run scientific research and technical service organizations of China. The model is composed of quantitative and qualitative factors, fuzzy comment set and weight vector. The fuzzy judgement technique is applied in the model. The application of the model is also demonstrated in the paper.     

关于陆-海相互作用的若干问题

陆-海相互作用是一个宽泛的研究领域, 但其同海洋科学的基础研究与我们经济、社会的可持续发展均密切相关. 基于个人对本研究领域的理解, 在本文中提出了一些需要在陆-海相互作用研究中予以关注的问题, 包括陆架边缘的物质交换、流域盆地中的物质流失、近海的富营养化与有害水华、底层水缺氧与“死亡区”、沉积物的“源”与“汇”、对地球其他层圈的反馈、海平面上升及其对社会与经济可持续性的作用、沿海地区的人类活动及其影响, 等等. 本文中未能涉及的命题并非不重要, 仅仅系我个人的知识和能力所限.     

冀中坳陷前第三系顶面不整合结构特征及油气藏类型

前第三系顶面不整合在冀中坳陷油气成藏和潜山形成过程中起重要作用.利用地震资料将冀中坳陷前第三系顶面不整合划分为断层-超覆不整合、削截-超覆不整合、断褶不整合、削截不整合和平行不整合等5种类型,其中断层-超覆不整合分布于陡坡带,削截-超覆不整合分布于缓坡带,断褶不整合多分布于中央隆起带,削截不整合多分布在凸起上,平行不整合则分布在生烃区.通过野外踏勘、岩心观察、录井与测井综合解释以及钻井泥浆漏失统计,将冀中坳陷前第三系顶面不整合纵向上划分为底砾岩、风化黏土层及风化淋滤带3层结构,其中风化淋滤带溶蚀孔、洞、缝十分发育,是油气运移和聚集的良好场所.目前已发现削截-超覆型油气藏、侵蚀潜山型油气藏和断褶不整合型油气藏等3类与不整合相关的油气藏.     

Adaptive Neural Control for a Class of Low-triangular-structured Nonlinear Systems with H-∞ Performance Analysis

In this paper, a neural-network-based variable structure control scheme is presented for a class of nonlinear systems with a general low triangular structure. The proposed variable structure controller is proved to be Cl, thus can be applied for backstepping design, which has extended the scope of previous nonlinear systems in the form of strict-feedback and pure-feedback. With the help of neural network approximator, H-∞ performance analysis of stability is given. The effectiveness of proposed control law is verified via simulation.     

粗粒料“空间准滑面”模型及其在高面板坝三维分析中的应用

本文在松冈元“空间准滑面”模型的基础上对粗粒料三维本构关系模型做了合理的发展。该模型考虑了高压下颗粒破碎引起模型参数和破坏准则随平均正应力或固结压力而变化的特性、剪缩剪胀特性、以及中主应力对变形的影响,从而有可能推广应用于混凝土面板堆石坝三维应力应变分析。参数可由大型常规三轴压缩试验测定,也可由邓肯-张E～B或E～u模型参数换算求得,便于工程单位使用。通过天生桥面板坝三维计算,所得结果比较合理,基本上符合实际,说明该模型是适用的,有较好的发展前途。