用3个向量对构造梁振动系统的刚度矩阵

针对梁的离散化模型的刚度矩阵是五对角矩阵,梁振动反问题的实质是实对称五对角矩阵的特征值反问题.该文利用向量对、Moore-Penrose广义逆给出了实对称五对角矩阵向量对反问题存在唯一解的条件,并结合矩阵分块讨论了双对称五对角矩阵向量对反问题解存在唯一的条件,进而计算了次对角线位置元素为负,其它位置元素均为正的实对称五对角矩阵特征值反问题.由于构造梁的离散模型需要的数据可由测试得到,故而其结果适合于模态分析、系统结构的分析与设计等方面应用.最后给出了数值算例,通过数值讨论说明方法的有效性.     

实对称五对角矩阵逆特征值问题

1 引 言 对于n阶实对称矩阵A=(aij),r是一个正整数,且1≤r≤n-1,当|i-j|>r时,aij=0(i,j=1,2,…,n),至少有一个i使得ai,i+r≠0,则称矩阵A是带宽为2r+1的实对称带状矩阵.特别地,当r=1时,称A为实对称三对角矩阵;当r=2时,称A为实对称五对角矩阵. 实对称带状矩阵逆特征值问题应用十分广泛,这类问题不仅来自微分方程逆特征值问     

实对称五对角矩阵的两类广义特征值反问题

讨论了如下两类广义特征值反问题:(i)由给定的三个互异的特征对和给定的实对称正定五对角矩阵构造一个实对称五对角矩阵;(ii)由给定的三个互异特征对和给定的全对称正定五对角矩阵构造一个全对称五对角矩阵.利用线性方程组理论、对称向量和反对称向量的性质,分别得到了两类反问题存在唯一解的充要条件,并给出了解的表达式和数值算法;最后通过数值例子说明了算法的有效性.     

一类对称三对角矩阵的合同对角化算法的实现

从一个对称三对角矩阵的合同变换出发 ,阐述了对称三对角矩阵对应的二次型标准化的一种方法 .     

实对称区间矩阵特征值确界的交错定理及其应用

本文将实对称矩阵特征值的交错定理推广到实对称区间矩阵,给出了实对称区间矩阵特征值确界的交错定理,并应用该定理构造了估计实对称三对角区间矩阵特征值界的算法.文中数值例子表明,本文所给算法与一些现有算法相比在使用范围、计算精度和计算量等方面都具有一定的优越性.     

三对角与五对角Toeplitz矩阵求逆的算法

提出了一种求三对角与五对角Toeplitz矩阵逆的快速算法,其思想为先将Toeplitz矩阵扩展为循环矩阵,再快速求循环矩阵的逆,进而运用恰当矩阵分块求原Toeplitz矩阵的逆的算法.算法稳定性较好且复杂度较低.数值例子显示了算法的有效性和稳定性,并指出了算法的适用范围.     

实对称带状矩阵逆特征值问题

研究了一类实对称带状矩阵逆特征值问题：给定三个互异实数λ,μ和v及三个非零实向量x,y和z,分别构造实对称五对角矩阵T和实对称九对角矩阵A,使其都具有特征对(λ,x),(μ,y)和(v,z)．给出了此类问题的两种提法,研究了问题的可解性以及存在惟一解的充分必要条件,最后给出了数值算法和数值例子．     

特殊五对角与七对角对称正定矩阵的一类反问题

针对线性代数方程组Ax=b,利用矩阵分解的思想,构造一类特殊五对角与七对角对称正定阵的矩阵分解,获得这类矩阵反问题解存在的充要条件和通解表达式.最后,给出了具体算法与数值算例.     

实对称三对角矩阵正定性的两个简单判别方法

本文给出实对称三对角矩阵正定性的两个简单判别法     

求分块三对角矩阵和分块周期三对角矩阵逆矩阵的快速算法

给出了分块三对角矩阵逆矩阵的快速算法,并利用所给算法得到了求分块周期三对角矩阵逆矩阵的快速算法.最后通过算例表示算法的有效性.     

一类特殊五对角和七对角行列式的计算

采用递推法研究了对称和非对称五对角行列式的通项公式.对于对称情形,给出显式表达式;对于非对称情形,通项公式为六个指数函数的线性组合.还得到对称七对角行列式的通项公式.     

A vanishing theorem for modular symbols on locally symmetric spaces

A modular symbol is the fundamental class of a totally geodesic submanifold embedded in a locally Riemannian symmetric space , which is defined by a subsymmetric space . In this paper, we consider the modular symbol defined by a semisimple symmetric pair (G,G'), and prove a vanishing theorem with respect to the -component in the Matsushima-Murakami formula based on the discretely decomposable theorem of the restriction . In particular, we determine explicitly the middle Hodge components of certain totally real modular symbols on the locally Hermitian symmetric spaces of type IV. Received: December 8, 1996     

A discrete problem on d-dimensional symmetric weighted median filters

Using a tool-energy of d-dimensional signal sequences taking values 1 or -1, we have completely solved an interesting discrete mathematical problem on d-dimensional symmetric weighted median filters, i.e., when a d-dimensional symmetric weighted median filter is applied iteratively to a real signal sequence, then the limit of the iterates with even index as well as the limit of the iterates with odd index both converge.     

Numerical Schottky Uniformizations

A real algebraic curve of algebraic genus g ≥ 2 is a pair (S, τ), where S is a closed Riemann surface of genus g and τ is a reflection on S (anticonformal involution with fixed points). In this note, we discuss a numerical (Burnside) program which permits to obtain a Riemann period matrix of the surface S in terms of an uniformizing real Schottky group. If we denote by Aut₊(S, τ) the group of conformal automorphisms of S commuting with the real structure τ, then it is a well known fact that |Aut₊(S,τ)| ≥ 12(g−1). We say that (S,τ) is maximally symmetric if |Aut₊(S,τ)|=12(g−1). We work explicitly such a numerical program in the case of maximally symmetric real curves of genus two. We construct a real Schottky uniformization for each such real curve and we use the numerical program to obtain a real algebraic curve, a Riemann period matrix and the accessory parameters in terms of the corresponding Schottky uniformization. In particular, we are able to give for Bolza’s curve a Schottky uniformization (at least numerically), providing an example for which the inverse uniformization theorem is numerically solved.Partially supported by Projects Fondecyt 1030252 1030373 and UTFSM 12.03.21     

The Inertia Set of Nonnegative Symmetric Sign Pattern with Zero Diagonal

The inertia set of a symmetric sign pattern A is the set i(A) = {i(B) | B = B ^T ∈ Q(A)}, where i(B) denotes the inertia of real symmetric matrix B, and Q(A) denotes the sign pattern class of A. In this paper, a complete characterization on the inertia set of the nonnegative symmetric sign pattern A in which each diagonal entry is zero and all off-diagonal entries are positive is obtained. Further, we also consider the bound for the numbers of nonzero entries in the nonnegative symmetric sign patterns A with zero diagonal that require unique inertia.     

A new symmetric five-diagonal finite difference method for computing eigenvalues of fourth-order two-point boundary value problems

We present a new symmetric five-diagonal finite difference method for computing eigenvalues of two-point boundary value problems involving a fourth-order differential equation.     

Discriminant of symmetric matrices as a sum of squares and the orthogonal group

It is proved that the discriminant of n × n real symmetric matrices can be written as a sum of squares, where the number of summands equals the dimension of the space of n‐variable spherical harmonics of degree n. The representation theory of the orthogonal group is applied to express the discriminant of 3 × 3 real symmetric matrices as a sum of five squares and to show that it cannot be written as the sum of less than five squares. It is proved that the discriminant of 4 × 4 real symmetric matrices can be written as a sum of seven squares. These improve results of Kummer from 1843 and Borchardt from 1846. © 2010 Wiley Periodicals, Inc.     

On the infinity norm of the inverse of a class of toeplitz matrices of band width five

A useful theorem for precisely bounding the infinity norm of the inverse of a five-diagonal symmetric positive definite Toeplitz matrix is obtained.