电机控制中电压空间矢量脉宽调制算法的探究

分析了电压空间矢量脉宽调制（SVPWM）算法中变换矩阵系数k与输出电压Uout的关系，指出在正弦波条件下，k为2／3时，Uout为相电压，k为（2／3）^1／2时，Uout为线电压．建立了绝缘栅双极晶体管开通和关断的动态过程模型，推导出开关损耗与开关频率的关系式，指出开关损耗与开关频率成正比而与占空比无关．研究结果表明，以000矢量开始和结束的开关模式在扇区切换过程中没有开关状态的变化，因而在扇区切换过程中没有产生开关损耗，5段式SVPWM的开关损耗比7段式的少了1／3，在其他开关模式中，由于存在扇区切换过程的开关损耗，5段式SVPWM的开关损耗比7段式的少将近1／3．上述结论为有关技术人员在设计系统时选择最佳算法提供了依据．     

A new discovery of macroscopic fossils from the Ediacaran Doushantuo Formation in the Yangtze Gorges area

The Ediacaran section in the Yangtze Gorges area, South China, is one of the most important in the worldand contains abundant fossils[1-4]. The earliest discov- ery of macrofossils in 1984, including macroscopic algal fossils from the uppermost Doushantuo…     

Rational design of glycerol dehydratase： Swapping the genes encoding the subunits of glycerol dehydratase to improve enzymatic properties

1,3-propanediol (1,3-PD) is an important material for chemical industry,and there has been always much interest in the production of 1,3-PD using all possible routes. The genes encoding glyc-erol dehydratase (GDHt) from Citrobacter freundii,Klebsiella pneumoniae and metagenome were cloned and expressed in E. coli. All glycerol dehy-dratases but the one from metagenome could be detected to show enzyme activities. In order to im-prove the enzymatic properties of GDHts,the genes encoding α and β-γ subunits were cloned,and the enzyme characteristics were evolved by rational de-sign based on their 3D structures which were con-structed by homology modeling. Six heteroenzymes were obtained by swapping the α subunit genes of these three different-source-derived GDHts. The pH,thermal stability and Vmax of some heteroenzymes were dramatically improved by 2―5 times compared with the wild one (GDHtKP). The GDHt cloned from metagenome,originally proved to be with no enzyme activity,was converted into active enzyme by swap-ping its subunits with other different GDHts. In addi-tion,the effect of subtle 3D structural changes on the properties of the enzyme was also observed.     

New Public Key Cryptosystems from Combinatorial Group Theory

0 IntroductionThe security of most commonly used cryptographicschemesis essentially based onthree families of compu-tational problems :the integer factoring problem(RSAprob-lem) ,the Diffie-Hell man problemand the discrete logarithmproblem,andthe elliptic curve variants thereof .Inthe middleof 1990s , Shor[1], Bonehet al[2]presented some remarkablequantumalgorithms which can solveinteger factoring problemand discrete logarithmproblemover any group including Ga-lois fields and elliptic curve…     

Merkle Tree Digital Signature and Trusted Computing Platform

0 IntroductionDigital signatureis a key technique to provide datainteg-rity andsource authenticationcapabilities to enhance da-ta trustworthiness in trusted computing platform. Most cur-rent digital signature schemes are based on complex mathe-matical problems , such as integer factoring, discrete log-arithm,and discrete logarithm on elliptic curves . The hard-ness of these underlying mathematical problems is really un-certain at present . Merkle treeis a method used for authenti-cation and di…     

动态电压恢复器的控制策略

通过把负载电压看作动态电压恢复器(DVR)的控制对象,开关器件的占空比看作控制量,电网电压和负载电流看作干扰量,可将DVR的数学模型简化为单输入单输出系统.结合控制量的计算方法,推导出了控制量到被控对象和干扰量到被控对象的传递函数.分析传递函数可知,负载电流特别是其中的高频分量会使DVR的补偿性能变坏.提出了误差校正控制方法,在传统电网电压前馈的基础上通过误差校正实现了负载电流前馈,基本消除了负载电流的影响.仿真结果验证了理论分析的正确性.     

本真生命与诗意人生——《九月寓言》文本阐释

"人之栖居"是海德格尔后期的主要思想。他把栖居之所具化为天地神人所构成的"四重整体,"通过建立一个本真的自然世界而恢复神与人,存在与大地的本源联系,使人与物的存在居于澄明。无独有偶,张炜在其小说《九月寓言》里也做了同样的努力:对"地瓜"、"黑夜"、"九月"等独特的意象的歌吟,极富诗意地还原了自然生命的本真形态;同时,通过对小村人流浪与安居两种生存境遇的诗意描绘,昭示了诗意栖居的必要性。     

基于代理的网格作业全局自适应迁移

作业迁移是实现网格作业服务质量保证和系统高效能的重要方法,在分析了传统进程迁移技术的基础上,根据网格系统的特点,提出了一种基于代理的网格作业全局自适应迁移模型,通过设计网格作业全局自适应迁移对象选定原则、迁移时机确定机制和作业迁移信息监控管理,解决了网格环境下作业迁移的关键问题.试验结果以及在西北工业大学校园计算网格中的实际应用验证了模型的有效性.     

基于图的数据挖掘算法研究

对图数据频繁模式的挖掘是近年的研究热点,而从图数据库中挖掘频繁模式的关键是子图测试和候选子图生成操作。与传统广泛研究的频繁项集、频繁序列、频繁子树挖掘相比较,频繁子图的挖掘更复杂、更有难度,因为图同构问题是一个NP-完全问题。因此,必须有个好的规范化编码和有效的算法来避免子图同构这个难题。     

一种有效抑制睫毛干扰的虹膜定位算法

为了提高在图像质量较差时的虹膜定位性能,改善虹膜定位的成功率和速度,在对Daug-man算法改进的基础上,提出了一种基于形态学算子的虹膜定位算法(ILBM).该算法将睫毛干扰看作虹膜图像中出现的缝隙,通过应用形态学中的膨胀算子对缝隙进行填补进而消除干扰,虹膜边缘参数则采用Hough变换来求取.与Daugman算法相比,ILBM算法可以有效抑制虹膜定位过程中存在的睫毛干扰,提高虹膜定位的速度.对比实验表明,ILBM算法具有比Daugman算法更好的抑制干扰能力和鲁棒性,具有更高的定位成功率和更快的定位速度,定位时间仅为Daugman算法的17%,定位成功率提高了3.7%.