由正交拉丁方导出的一族新的结合方案

<正> 部分平衡不完备区组(PBIB)设计(以下简称部分平衡设计),作为平衡不完备区组(BIB)设计的推广,最初由Bose-Nair引进.后来,Bose-Shimamoto为了阐明部分平衡设计的实验数据的统计分析,明确地引进了结合方案的概念.本文的目的是利用具有     

极大3无关组的函数特征

有限域Fq上r维向量空间Frq中的极大n元t-无关组是—个很重要的组合结构,可以用来构造强部分平衡设计,从而可构造最优认证码;也可用来构造正交阵列.本文揭示有限域F2上r维向量空间Fr2中极大3无关组的函数特征,即每一个极大3无关组恰好是某个线性函数的支撑.     

具有多级选择的离散网络平衡设计模型与算法

本文提出了一种新的离散网络平衡设计二层规划模型,模型同时考虑了新增路段及已有路段的扩容,而且允许不同等级的扩容选择.模型求解中,上层采用粒子群算法,而下层则采用本文作者提出的仿射尺度内点算法.数值计算结果显示,本文构建的算法能够快速有效地求解这类新的网络平衡设计二层规划模型.     

平衡设计多向分类多元重复测量模型的Wilks检验

对平衡设计多向分类多元重复测量模型,利用极大似然比方法,推导了对各单个固定效应分别进行检验的Wilks型检验规则.并推导了对多个固定效应进行同时检验的检验规则.推导了非中心分布的参数与原始参数和样本容量的关系.     

独立随机序列加权和的强收敛性

本文研究独立随机变量序列加权和的强收敛性,利用截尾法和Borel-Cantelli引理,证明了加权系数ank为列阵情形的强收敛性,在一般双下标加权系数的加权部分和的强收敛性,并对Jamison型加权部分和情形证明了其强收敛的充要条件,推广了Chow与Teicher(1971)[3]的相应结果.     

一类新的部分差集以及分圆数的新性质

本文研究了分圆理论与部分差集,强正则图的关系.利用分圆方法,构造了一类新的部分差集,并反过来得到了分圆数的一些新性质.     

一类新的部分差集以及分圆数的新性质（英文）

本文研究了分圆理论与部分差集,强正则图的关系.利用分圆方法,构造了一类新的部分差集,并反过来得到了分圆数的一些新性质.     

随机删失下半参数估计的强相合性

主要讨论了随机删失下的部分线性模型,利用基于分布函数的核估计和最小二乘法,给出了删失情况下参数和非参数部分的估计,并证明了它们的强相合性.     

若干族拉丁方型部分平衡设计

§1.引言 拉丁方型部分平衡设计是由Bose-Shimamoto根据结合方案的概念引进的一类带有两个结合类的部分平衡不完备区组设计。一个拉丁方型结合方案,记作L_i(s),是由v=s2个处理排成s×s方阵并在其上叠加i-2个s阶相互正交拉丁方(如果这种拉丁方存在的话),使得出现在方阵的同一行上的,同一列上的或对应于所加拉丁方之一的同一字母上的处理对成为第一结合,其余处理对成为第二结合,此处i叫做约束数。     

M值随机序列的一个普遍成立的强极限定理

利用区间部分法构造几乎处处收敛的鞅,得到了一个对任意M－值随机变量序遍成立的强极限定理。     

(γ,α)型广义强凸性

引入(γ,α)型广义强凸集与强凸函数,讨论了广义强凸性质,并在此基础上提出对强凸函数进行分类的标准和判定方法.然后引入标准强凸函数概念,推出最小标准强凸函数形式,并探讨了广义强凸集与强凸函数的关系.     

A modified complete ranking of DMUs using restrictions in DEA models

Alirezaee and Afsharian [1] have proposed a new index, namely, Balance Index, to rank DMUs. In this paper, we will use their examples to illustrate that the proposed index is not stable. As a result, the corresponding rankings are also unstable. Then we analyze where an error occurs in the new method for complete ranking of decision making units and amend it by introducing the Maximal Balance Index. The numeral example reports the reasonability of our methods.     

Parameter identification for nonlinear dynamical systems using Multi Harmonic Balance techniques

An usual approach to investigate nonlinear systems in the frequency domain is the application of the Harmonic Balance Method (HBM), assuming that a harmonic excitation of the system leads to a harmonic response. However, for systems where the steady state response is not just harmonic but periodic or for systems which are excited periodically, this assumption does not longer lead to satisfying results. Therefore, the Multi Harmonic Balance Method (MHBM) is utilized. (© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Comparison of Energy Balance Period with Exact Period for Arising Nonlinear Oscillator Equations

In this paper, He’s energy balance method is applied to nonlinear oscillators. The new algorithm offers a promising approach by constructing a Hamiltonian for the nonlinear oscillator. We proved that the energy balance is very effective, convenient and does not require any linearization or small perturbation. In contradicts of the conventional methods, He’s Energy Balance method (HEBM) using just one iteration, leads us to high accuracy of solutions. Energy Balance method is very effective, convenient and adequately accurate to both linear and nonlinear problems in physics and engineering.     

基于全局搜索算法的不平衡指派问题研究

目前求解不平衡指派问题的主要是将其转化为平衡的指派问题后再去处理.针对不平衡指派问题提出了全局搜索算法,算法不用将不平衡问题转化为平衡问题进行求解,而是基于全局最优策略对任务进行指派,方法理论更加简单,操作更加方便,使得不平衡指派问题得到了很好地解决,同时,这种算法对平衡指派问题、运输问题等依然有效.     

关于图的强乘积的可迁性的研究（英文）

Since many large graphs are composed from some existing smaller graphs by using graph operations, say, the Cartesian product, the Lexicographic product and the Strong product. Many properties of such large graphs are closely related to those of the corresponding smaller ones. In this short note, we give some properties of the Strong product of vertex-transitive graphs. In particular, we show that the Strong product of Cayley graphs is still a Cayley graph.     

求解线性方程组的顺序算法和并行算法的理论时耗估价

本文讨论了求解密集型线性方程组的两种并行算法。这两种算法都在下上单元(LU)分解。法的基础上使用了前向和后向置换进行的。这些算法在数值上是稳定的,并在顺序平衡机上用各种处理程序进行试验,都得到良好效果。     

Searle freed of every flaw

Strong Al presupposes (1) that Super-Searle (henceforth ‘Searle’) comes to know that the symbols he manipulates are meaningful, and (2) that there cannot be two or more semantical interpretations for the system of symbols that Searle manipulates such that the set of rules constitutes a language comprehension program for each interpretation. In this paper, I show that Strong Al is false and that presupposition #1 is false, on the assumption that presupposition #2 is true. The main argument of the paper constructs a second program, isomorphic to Searle’s, to show that if someone, say Dan, runs this isomorphic program, he cannot possibly come to know what its mentioned symbols mean because they do not mean anything to anybody. Since Dan and Searle do exactly the same thing, except that the symbols they manipulate are different, neither Dan nor Searle can possibly know whether the symbols they manipulate are meaningful (let alone what they mean, if they are meaningful). The remainder of the paper responds to an anticipated Strong Al rejoinder, which, I believe, is a necessary extension of Strong Al.     

Software Reviews

The Sky 2.0 Geometric preSupposer: Points and Lines Balance of Power Codes and Cyphers Success With Math: Ratio and Proportion Science Toolkit