SAW带通滤波器的优化设计

本文将线性相移利FIR数字滤波器优化设计法——等波动切比雪夫逼近用于设计优化的SAW滤波器,设计出了任意通带形状SAW带通滤波器,并将这种优化设计方法同SAWF的传统设计方法进行了比较。实验与理论结果基本一致。     

Model of Next Generation Energy-Efficient Design Software for Buildings

Energy-efficient design for buildings (EEDB) is a vital step towards building energy-saving. In order to greatly improve the EEDB, the next generation EEDB software that makes use of latest technologies needs to be developed. This paper mainly focuses on establishing the model of the next generation EEDB software. Based on the investigation of literatures and the interviews to the designers, the requirements on the next generation EEDB software were identified, where the lifecycle assessment on both energy consumption and environmental impacts, 3D graphics support, and building information modeling (BIM) support were stressed. Then the workflow for using the next generation EEDB software was established. Finally, based on the workflow, the framework model for the software was proposed, and the partial models and the corresponding functions were systematically analyzed. The model lays a solid foundation for developing the next generation EEDB software.     

Study on Chart Based on Short Production Run

High intellectual, high flexible automatic manufacturing system based on low volume production is the trend in the future, but the traditional control charts whose control limits are based on at least 20～25 subgroups of five observations each are not suitable in this environment. The paper presents variance control chart used in small batch production in which there are not sufficient observations. The control limits are set reliably without restriction of the number of samples.     

Operation principle of a novel curvature plastic fiber optic sensor

The operation principle of a new type of intensity modulate macrobend curvature optical fiber senor was presented based on surface light scattering theory. Sensor‘s static and dynamic performance was investigated. This type of sensor can distinguish between positive and negative bending directions. When curvature radius is larger than 50mm, the sensor will keep good linearity. Two-dimensional shape measurement experiments using curvature sensors have been implemented.     

实时定量PCR研究大鼠肌肉组织基因表达

运用实时定量聚合酶链式反应(RQ-PCR)研究了正常以及患有糖尿病大鼠的肌肉组织中胰岛素信号转导通路中重要的信号传导分子基因表达,结果表明:四氧嘧啶诱导的糖尿病大鼠肌肉组织的胰岛素受体、胰岛素受体底物1和葡萄糖转运体4的mRNA表达下调,胰岛素磷脂酰肌醇3激酶的mRNA表达上调;糖尿病大鼠骨骼肌葡萄糖转运体4的mRNA含量下降可能会引起糖摄取能力的下降.从而导致四氧嘧啶诱导大鼠肌肉组织胰岛素抵抗.     

Effects of alloxan-induced diabetes on the expression of insulin signal transmission molecules

The mRNA of insulin receptor (IR), the insulin receptor substrates (IRSs), glucose transporter 2 (GLUT2) and glucokinase (GK) in alloxan-induced diabetic rat livers were amplified by RT-PCR. The protein of the insulin receptor in rat livers was determined by western-blotting. The results show that IR expression level decreased at the level of mRNA and protein. The gene expressions of IRSs, GLUT2 and GK changed significantly. The hepatic glycogen content in alloxan-diabetic rats treated with insulin ((13.2 ± 0.4) mg · g⁻¹) did not restore to normal level ((17.0 ± 0.4) mg · g⁻¹) by means of anthrone-H₂SO₄ methods. The results imply that insulin resistance is developed during inchoate phase of alloxan-induced diabetic rats.     

利用遗传算法改进的路由和波长分配算法

给定拓扑结构和流量请求的波分复用静态波长路由光网络中 ,利用路由和波长分配算法求得的网络最小所需波长数与网络节点编号有关。为了得到网络最小所需波长数所对应的网络节点编号 ,首次利用遗传算法对节点编号进行优化。和已有的统计方法相比较 ,采用遗传算法可以有效地消除原有方法的随机性和盲目性。模拟计算表明 ,利用遗传算法可以更快地求得网络最小所需波长数所对应的网络节点编号。以此求得的网络所需波长数 ,更接近理论最小值     

火焰原子吸收光谱法研究中药中镁的溶解状态

按照传统煎煮法提取中药枸杞、红花和菊花中金属镁,采取正辛醇和不同的酸度模拟枸杞、红花和菊花三味中药中的镁在人体胃肠中的溶解状态,用火焰原子吸收光谱法测定枸杞、红花和菊花中微量镁的总含量、水煎液中镁的溶出率,以及水煎液中人体的胃肠液酸度条件下水和辛醇中镁的溶解性,实验结果表明胃肠液酸度变化及中药配伍对三味中药中镁溶解状态...     

Raman激光频率扩展器参数分析

本文对所研制的高压氢Ｒａｍａｎ频率扩展器的主要参数指标：阈值、输出能量、转换效率、输出模式（包括纵模及空间分布）等进行了研究。并给出了它们与泵浦激光（波长、强度、模式等）及Ｒａｍａｎ介质（长度、气压等）的关系与实验结果。频率扩展器用０．５３ｕ脉冲激光（６０ｍＪ）泵浦可得到～１７ｍＪ（０．６８ｕ）的一阶Ｓｔｏｋｅｓ输出，转换效率达 ２８％，总转换效率（包括三阶 Ｓｔｏｋｅｓ及四阶以上的 Ａｎｔｉ－Ｓｔｏｋｅｓ）超过 ４０％。     

基于机器学习的材料疲劳和腐蚀寿命预测研究进展

机器学习在多变量拟合、复杂数据相关性分析、隐含信息挖掘利用等方面具有强大功能,在机械结构故障预测和材料寿命预测方向的应用研究成为热点.本文介绍机器学习在材料疲劳和腐蚀寿命预测中的应用,深入探讨BP神经网络、支持向量回归、聚类分析算法和集成算法综合应用的研究现状,总结模型参数的选取方法和模型性能的评估指标,并对机器学习在材料疲劳和腐蚀寿命预测研究领域存在的主要问题和发展趋势进行剖析.