Observational study of the circulation in the southern South China Sea

CTD data from two oceanographic survey cruises, conducted by “Shiyan No. 3” R/V during two reversing monsoons, were employed to calculate and analyze the distributions of temperature, salinity and geopotential anomaly in the southern South China Sea (SSCS). Based on these distributions, the horizontal structures of the SSCS circulation are described systematically. Study results show that the SSCS circulation structure could be classified as a two-layer (the upper and the lower) pattern, and that the main circulation in the upper layer (0–400 m) is driven by the South China Sea monsoons and is opposite in direction to the reversal of the monsoons. The distributions of principal current systems and their features are presented. Two important local phenomena, local eddy and oceanic front, were observed.     

Osmium isotope determination on mantle-derived peridotite xenoliths from Panshishan with N-TIMS

The¹⁸⁷Os/¹⁸⁸Os ratios of spinel lherzolite xenoliths from Panshishan determined with N-TIMS are lower than that of the primitive mantle, which shows depleted mantle characteristics. Their positive correlation with Al₂O₃/MgO suggests that the geochemical behavior of trace elements Re and Os is similar to that of Al₂O₃ and MgO respectively during the magmatic evolution and Re-Os isotopic system is largely immune to mantle metasomatism. A model age of 2.8–3.4 Ga obtained by¹⁸⁷Os/¹⁸⁸Os-Al₂O₃/ MgO correlation might represent the homogeneous age of the mantle lithosphere beneath the area.     

用^1H—NMR测定纤维素的摩尔取代度

用＾１Ｈ－ＮＭＲ测定羟丙基纤维素的摩尔取代度,根据纤维素在ＤＣｌ中水解后的＾１Ｈ－ＮＭＲ谱峰面积来计算ＭＳ值。同时还应用铬酸氧化蒸馏法测定了羟丙基纤维素的ＭＳ值,以及利用ＭＳ与羟丙基含量的关系得到ＭＳ值,可以发现ＮＭＲ的方法是可行的。     

成组技术中面向对象的参数化模型

针对传统成组技术中的编码系统难以使用、编码不够精确和重用困难等问题,提出在成组技术中构造面向对象的参数化模型．该模型用类来描述零件簇,构成零件类的结构层次,生成零件描述的框架;用参数化方法从零件类实例化生成对象,以对象来表示具体的零件;查询时,以零件的特征和零件的具体参数来驱动,加速搜索的过程,并使结果更加精确;找到相同或相近零件对象后,在某个类层次上重用以前的设计．该模型的提出为成组技术的应用提供了新的途径．     

基于实例的工艺知识获取模型

介绍了耦合神经网络的工艺设计实例推理系统的实现方法论,提出了一种基于神经网络的工艺设计实例推理索引模型．与现存大多数实例推理系统不同,该方法用神经网络实现实例的动态分类和索引．实例层次分类的三层结构和基于特征的聚类模板概念,为实现基于符号处理的实例推理求解模式向基于神经计算的模式识别求解模式映射提供了条件．提出了基于实例的工艺知识获取模型,采用新实例的入库操作实现工艺知识的隐式获取,从而使知识获取得以简化．神经网络的自适应、自学习能力将减少系统的日常维护工作．基于实例的系统可望解决知识获取的难题．     

基于改进混合模型的一类固定床反应器推断控制

对一类非绝热式固定床反应器提出了一种基于改进混合模型的非线性自适应推断控制策略：给出一种基于结合先验知识和人工神经网络的改进混合模型的推断估计器设计方法;提出一种改进程序变增益ＰＩＤ算法设计控制器;设计了该固定床反应器基于改进混合模型的推断控制系统。仿真结果表明：该推断估计器具有良好的静态和动态特性,较高的估计精度和较好的实时性,所需训练样本少,易于理解和研究及外推性能良好等特点,该推断控制系统还具有良好的跟踪和抗干扰性能。     

新老混凝土的粘结机理和测试方法研究综述

就影响新老砼粘结性能的主要因素,新老砼结合面处理方法,修补材料的选择和应用,粘结剂的种类,新老砼粘结性能试验方法,新老砼粘结机理五个方面的问题进行了论述。依实际工程砼结构修补加固统计结果,指出实际砼修补量最大的是道路和桥梁,提出了关于道路、桥梁、汽车库、飞机场等交通设施,砼修补需进一步研究的问题。     

Preparation of enantiomerically pure 1,1′-bi-2-naphthols from 85% cinchonine and 1,1′-bi-2-naphtholboric anhydride

A possibility of preparation of enantiomerically pure 1,1′-bi-2-naphthols using an impure cinchonine has been examined. The solid and the mother liquor formed from the reaction of 1,1′-bi-2-naphtholboric anhydride and 85% cinchonine in toluene could give optically pure (S)-(−)- and (R)-(+)-1,1′-bi-2-naphthol after acidification and kinetic crystallization, the overall yields were 40% and 28%, respectively. Supported by the National Natural Science Foundation of China (2937206) and the Natural Science Foundation of Hubei Province (93J27) Xiong Ying, born in 1971, Graduate student     

关于时滞Liénard型方程的周期解

本文研究具有时滞的Lienard型方程周期解的存在性,得到了一些新的结果。     

巴非蛤冷冻调理食品的加工工艺

本文介绍巴非蛤一种新颖的加工方法,工艺流程简单,只需冷冻、冷藏等设备就可生产,产品外观鲜美,味美可口,食用方便,该工艺对提高巴非蛤的食用价值和经济价值都具有一定的实际意义。