Fractional observer to estimate periodical forces

In the present work we propose a fractional state observer with constant gain to estimate the periodical force exerted on a mechanical system by measuring only its displacement. The state observer is designed from both the Fourier series that approximates the periodical force and the equations of the damped harmonic oscillator that represents the behavior of the system. Specifically, the reconstruction of the force is carried out from the estimates of the series coefficients, which in fact are part of the dynamical system that composes the observer. Adams-Bashforth-Moulton method is used to compute the fractional derivatives of the observer in the Liouville-Caputo sense. Experiments based on real data are presented to show the advantages of using a fractional observer in the reconstruction of forces.     

气门电镦热锻成型机器人系统PC与PLC串行通信的实现

提出一种气门电镦热锻成型机器人系统PC机与PLC串行通信的实现方法,即采用定时扫描通讯状态的方式等待PLC发送回握信号或数据,基于Visual C++6.0 MSComm控件编写了应用程序。实验结果表明:该应用程序实现了PC与PLC间的串行通讯,且在连续读写PLC时,应用程序占用系统CPU资源少,实时性好,提高了发动机气门电镦热锻成型机器人系统控制的可靠性。     

An eigenelement method of periodical composite structures

Eigenelement method is an eigenvector expansion based finite element method, which was proposed by the authors to solve the macro behaviors of composites with less computational cost. To improve the macroscopic accuracy of the classical eigenelement method (CEEM), a serendipity eigenelement method (SEEM) is proposed, which takes the geometry and elastic properties of different phases of composites into account to some extent. Moreover, the shape function and its construction method of a multiscale eigenelement method (MEM) are presented, and the results of SEEM and MEM are compared with that of CEEM and the mathematical homogenization method (MHM) whose physical interpretation is revealed for the first time. It is shown that MEM is the most accurate eigenelement, SEEM is more accurate than CEEM, and MEM satisfies the two essential homogenization conditions: the strain energy equivalence and the deformation similarity. The extensive numerical comparison is given for stresses, displacements and frequencies.     

Shish-kebab-like cylindrulite structures resulted from periodical shear-induced crystallization of isotactic polypropylene

In this study, isotactic polypropylene (iPP) samples were prepared by conventional injection molding (CIM) and pressure vibration injection molding (PVIM), in which a periodical shear field was imposed on the iPP melt during the cooling solidification. The distribution of supermolecular structures of samples was investigated by Polarized Light Microscopy (PLM) and Scanning Electron Microscopy (SEM). Results show that the through-the thickness-morphology of sample prepared by CIM features a typical skin-core structure, as a result of general shear-induced crystallization. This structure can be divided into three layers, including a skin layer in which the shish-kebab structure was found, a transition region with deformed spherulite structure and a core layer with spherulitic structure. However, the morphology of the sample prepared by PVIM, as a result of periodical shear-induced crystallization, features a richer and fascinating supermolecular structure and can not be roughly divided into three layers. A region full of shish-kebab-like cylindrulite structures was found between the transition region and the core layer, which is rare to be seen in conventional injection molding. Based on their various core structures, two kinds of shish-kebab-like cylindrulites were defined: one is multi-fibril-core cylindrulite of which core is an assembly of multiple fibrils, and the other is single-fibril-core cylindrulite of which the core just contains a single fibril. Based on the investigated results, a schematic illustration is proposed to depict the through-the thickness-distribution of supermolecular structure of iPP sample prepared by PVIM. The mechanism of the formation of the two kinds of shish-kebab-like cylindrulite structures is also depicted by a schematic illustration, and it was discussed in terms of periodical shear-induced crystallization.     

一类非线性周期时间序列模型

为了描述周期时间序列中的偏倚和多峰等非线性特征,结合有限混合模型方法,提出混合周期自回归滑动平均时间序列模型（MPARMA）,给出了MPARMA模型的平稳性条件,讨论了期望最大化（EM）算法的应用,通过PM10浓度序列分析,评估了MPARMA模型的表现。     

与三次国家期刊奖的缘--从《电池》荣获第三届国家期刊奖百种重点期刊奖谈起

<电池>杂志2005年荣获了第三届国家期刊奖百种重点期刊奖,这是<电池>杂志继2002年荣获第二届国家期刊奖百种重点期刊奖、1999年荣获首届国家期刊奖后,再获此奖.评选结果出人意料之外,却又在情理之中!     

周期性圆环结构的拓扑优化设计

基于拓扑优化方法开展了周期性圆环结构的构型优化设计,分析了不同载荷工况下不同子结构数、内外半径比和内外蒙皮厚度对优化结果的影响规律。结果表明:周期性圆环结构具有明显的尺寸效应;相同体分比下,子结构数较少时优化结构刚度较高,但优化构型变动大;随着子结构数增多,结构刚度减弱并趋于极限值,优化构型稳定不变且受内外半径比和蒙皮厚度变化影响很小;而不同工况下,可忽略尺寸效应的临界子结构数存在差异,蒙皮厚度变化对结构效率的影响不同。     

应用小波分析进行短期负荷预测

小波分析是一种新兴的数学工具，它能任意地提取短期负荷序列的细节。通过使用小波分析，可以在任何水平上分析短期负荷序列，它对信息成分采取逐渐精细的时域与频域处理，尤其对突发与短时的信息分析具有明显的优势。本文将小波分析引入了短期负荷预测，针对电力系统本身具有的负荷以天，周，年为周期发生波动的特点，使用周期自回归模型有选择的对分解序列进行预测，并对直接使用周期自回归(PAR)模型的预测结果及先使用小波分析处理的预测结果进行了比较，实例显示这种方法提高了预测的精度。     

试析学报类期刊在科技期刊中的地位

运用文献理论及文献计量方法,对学报类期刊进行了数理统计与分析及比较研究,初步揭示了学报类期刊在科技期刊中的突出地位,讨论了其地位形成的原因与背景,提出了发展与繁荣学报类期刊和科技期刊的若干建议。     

基于J2EE的期刊管理系统的架构设计与实现

提出了一种基于J2EE的期刊网络化管理框架，阐明该框架的设计原理和运行过程，详细分析了期刊管理系统的设计需求和功能划分，并对实施实例的特点进行了说明。