Some Closed-Form Evaluations of Multiple Hypergeometric and <Emphasis Type="Italic">q</Emphasis>-Hypergeometric Series

The main object of the present paper is to show how some fairly general analytical tools and techniques can be applied with a view to deriving summation, transformation and reduction formulas for multiple hypergeometric and multiple basic (or q-) hypergeometric series. By making use of some reduction formulas for multivariable hypergeometric functions, the authors investigate several closed-form evaluations of various families of multiple hypergeometric and q-hypergeometric series. Relevant connections of the results presented in this paper with those obtained in earlier works are also considered. A number of multiple q-series identities, which are developed in this paper, are observed to be potentially useful in the related problems involving closed-form evaluations of multivariable q-hypergeometric functions. Dedicated to the Memory of Leonard Carlitz (1907–1999)Mathematics Subject Classifications (2000) Primary 33C65, 33C70, 33D70; secondary 33C20, 33D15.     

密集模态挠性结构模型多变量频域辨识和控制

密集模态挠性结构的模态不仅阻尼小而且耦合程度高,给模型辨识带来了很大困难.频域辨识是获取空间挠性结构模型的一个有效方法,但是目前频域辨识要么模型结构定义有缺陷,要么是集中在单变量情形,不适合作大型空间挠性结构多变量模型的辨识.本文提出了多变量极大似然频域辨识,给出了其模型结构定义、算法推导及实现,并将其应用于H型密集模态挠性板的模型辨识,根据辨识结果,设计主动控制律,实现了对密集模态挠性板振动的有效抑制,表明了辨识算法及主动控制的可行性.     

Pseudo-division algorithm for matrix multivariable polynomial and its application

ＩｎｔｒｏｄｕｃｔｉｏｎＴｈｅｐｒｏｂｌｅｍｏｆｇｉｖｅｎｄｉｆｆｅｒｅｎｔｉａｌｅｑｕａｔｉｏｎａｎｄｓｅｅｋｉｎｇｉｔｓｅｑｕｉｖａｌｅｎｔＨａｍｉｌｔｏｎｉａｎｉａｎｓｙｓｔｅｍｂｅｌｏｎｇｓｔｏｉｎｖｅｒｓｅｐｒｏｂｌｅｍｏｆａｎａｌｙｔｉｃａｌｍｅｃｈａｎｉｃｓ,ａｎｄｔｈｅｌａｔｔｅｒｉｓｏｎｅｏｆｐｒｉｍｅｐｒｏｂｌｅｍｓｉｎｃｌａｓｓｉｃａｌｍｅｃｈａｎｉｃｓ.ＴｈｅｆａｍｏｕｓＰｏｉｎｃａｒｅ＿ＣａｒｔｏｎｐｒｏｂｌｅｍｉｓｔｏｇｅｔｔｈｅＨａｍｉｌｔｏｎｉａｎｉａｎｒｅｇｕｌａ…     

智能信息处理的多指标面板数据聚类方法及其应用

为提高具有先验知识样本的学习效率,本文在吸引子传播聚类模型基础上,引入半监督学习策略,并综合考虑样本动态信息变化,融合多指标面板数据,提出智能信息处理的多指标面板数据聚类模型。选取30家房地产业上市公司2009-2013年相关财务数据,利用此模型进行聚类和绩效评价分析。结果表明,智能信息处理的多指标面板数据聚类模型能更加有效地区分样本类别特征,可为上市公司绩效评价、金融管理与决策提供一个更加有效的方法和手段。     

基于最大Lyapunov指数的多变量混沌时间序列预测

参考基于最大Lyapunov指数的单变量混沌时间序列预测方法,提出一种通过选取多个邻近重构向量,预测多变量混沌时间序列的局域法.采用新方法对两个完全不同的Rssler方程的耦合系统,Rssler方程和Hyper Rssler方程的耦合系统的多变量混沌时序进行一步和多步预测,结果表明了该方法的有效性,且算法具有较强的抗噪能力.讨论了参考邻近点数和预测结果的关系. 关键词： Lyapunov指数 混沌时间序列预测 多变量时间序列 最小二乘法     

Critical perspectives of pedagogical approaches to reversing the order of integration in double integrals

This paper presents some critical perspectives regarding pedagogical approaches to the method of reversing the order of integration in double integrals from prevailing educational literature on multivariable calculus. First, we question the message found in popular textbooks that the traditional process of reversing the order of integration is necessary when solving well-known problems. Second, we illustrate that the method of integration by parts can be directly applied to many of the classic pedagogical problems in the literature concerning double integrals, without taking the well-worn steps associated with reversing the order of integration. Third, we examine the benefits and limitations of such a method. In our conclusion, we advocate for integration by parts to be a part of the pedagogical conversation in the learning and teaching of double integral methods; and call for more debate around its use in the learning and teaching of other areas of mathematics. Finally, we emphasize the need for critical approaches in the pedagogy of mathematics more broadly.     

病人满意度影响因素的统计分析

本文在文献研究的基础上,选择了可能影响病人满意的27个指标,通过因子分析,把这27个指标综合成了5个因子,即医护人员技术与能力、医护人员的态度、医疗效果、花费、医疗环境和就医(辅助)过程5个方面;论文进一步对这5个方面构成的量表的可靠性作了验证;最后,应用多元回归技术分析了5个因子对病人满意度的影响程度,为医院提高病人满意度提供了科学的依据。     

Double convolution integral equations involving a general class of multivariable polynomials and the multivariableH-functions

In this paper we have solved a double convolution integral equation whose kernel involves the product of theH-functions of several variables and a general class of multivariable polynomials. Due to general nature of the kernel, we can obtain from it, solutions of a large number of double and single convolution integral equations involving products of several classical orthogonal polynomials and simpler functions. We have also obtained here solutions of two double convolution integral equations as special cases of our main result. Exact reference of three known results, which are obtainable as particular cases of one of these special cases, have also been included.     

High-performance numerical algorithms and software for subspace-based linear multivariable system identification

Basic algorithmic and numerical issues involved in subspace-based linear multivariable discrete-time system identification are described. A new identification toolbox—SLIDENT—has been developed and incorporated in the freely available Subroutine Library in Control Theory (SLICOT). Reliability, efficiency, and ability to solve industrial identification problems received a special consideration. Two algorithmic subspace-based approaches (MOESP and N4SID) and their combination, and both standard and fast techniques for data compression are provided. Structure exploiting algorithms and dedicated linear algebra tools enhance the computational efficiency and reliability. Extensive comparisons with the available computational tools based on subspace techniques show the better efficiency of the SLIDENT toolbox, at comparable numerical accuracy, and its capabilities to solve identification problems with many thousands of samples and hundreds of parameters.