用矩阵方程法求解循环卷积的改进算法

从循环卷积的定义出发,描述了用以m为变量的方法求解循环卷积的步骤.特别详述了其中的矩阵方程法,指出了该方法的不足——含有冗余项,提出了矩阵方程法的改进算法,消除了冗余项,简化了矩阵方程,给出了矩阵方程内部各列的构成规律.实例表明该方法简便、有效.     

一类四元数矩阵方程的循环解及其最佳逼近

本文研究了四元数体上矩阵方程XB=C的循环解及其最佳逼近问题.利用循环矩阵的结构表示式,以及四元数矩阵的复分解,得到了方程XB=C的循环解存在条件及其通解形式;在循环矩阵约束条件下,给出了该方程的最小二乘解集合;与此同时,在最小二乘解集合中,获得与给定四元数循环矩阵的最佳逼近解.推广了约束矩阵方程的数值求解范围.数值算例验证了本文算法的可行性.     

一类四元数矩阵方程的循环解及其最佳逼近

本文研究了四元数体上矩阵方程XB = C 的循环解及其最佳逼近问题. 利用循环矩阵的结构表示式, 以及四元数矩阵的复分解, 得到了方程XB = C 的循环解存在条件及其通解形式; 在循环矩阵约束条件下, 给出了该方程的最小二乘解集合; 与此同时, 在最小二乘解集合中, 获得与给定四元数循环矩阵的最佳逼近解. 推广了约束矩阵方程的数值求解范围. 数值算例验证了本文算法的可行性.     

四元数体上Sylvester方程的循环解及其最佳逼近

Sylvester方程AX-XB=C是一类具有广泛应用背景的矩阵方程,本文在四元数体上讨论它的循环解及其最佳逼近问题.主要利用四元数矩阵的实分解和循环矩阵的特定结构,借助Kronecker积把约束四元数矩阵方程转化为实域上的无约束方程,从而得到四元数体上Sylvester方程的循环解存在条件及其通解形式.同时,在循环解集合中,寻找到与预先给定的四元数循环矩阵有极小Frobenius范数的最佳逼近解.数值算例验证了本文方法的可行性.     

一种求解大型Lyapunov矩阵方程的预处理并行算法

研究了一种求解大型Lyapunov矩阵方程的并行预处理变形共轭梯度法.首先将处理小型矩阵方程的Smith预处理方法引入该问题的求解,将原矩阵方程转变为Stein方程,然后采用变形共轭梯度法并行求解预处理后的矩阵方程.其中遇到的难点是需要确定参数μ及求矩阵(A+μI)的逆.基于估计特征值的Gerschgorin圆定理给出了参数μ的估值,再采用变形共轭梯度法并行求得矩阵(A +μ l)的逆,从而形成预处理后的矩阵方程.通过数值试验,该算法与未预处理的变形共轭梯度法相比较,预处理算法明显优于未预处理的算法,而且其并行效率高达0.85.     

一种新的退卷积解谱技术

本文介绍一种新的退卷积方法,它是将退卷积问题化为求解逆矩阵问题,用奇异值分解(SVD)法处理病态矩阵体系,结合本文提出的最佳判据,求得退卷积结果。该方法已用 FORTRAN 5编成程序,对实验观测能谱及计算机模拟谱进行了退卷积处理,方法可靠,且退卷积解是稳定的。     

求Lyapunov矩阵方程的双对称解的迭代算法

本文研究了Lyapunov矩阵方程.利用共轭梯度法,建立了求该矩阵方程双对称解的迭代算法.同时,也能给出指定矩阵的最佳逼近双对称矩阵.     

矩形薄板瞬态响应的卷积型DQ半解析法

卷积型的Gurtin变分原理是目前在数学上唯一能和动力学初值问题完全等价的变分原理,它完全反映了有关初值问题的全部特征,通过卷积将矩形薄板原始控制方程构造成包含初始条件的新的具有完整初值问题特征的控制方程.对新的控制方程在时间域取解析函数,在空间域采用离散的DQ(differential quadrature)法,从而构造了卷积型DQ半解析法.该方法既可以达到和Gurtin变分原理相同的效果,又避开了Gurtin泛函的繁复,经对矩形薄板的动力响应问题的计算表明,该方法是一种精度好效率高的求解动力响应问题的计算方法.     

一类含有梯度项的非局部椭圆微分不等式解的Liouville型定理

本文主要研究一类具有卷积型非局部项和梯度项的拟线性椭圆微分不等式解的Liouville型定理.主要定理的证明基于非线性容度法,该方法可以处理卷积型非局部项,而且不需要使用比较原理或者极值原理.     

求解带Toeplitz矩阵的线性互补问题的一类预处理模系矩阵分裂迭代法

针对系数矩阵为对称正定Toeplitz矩阵的线性互补问题,本文提出了一类预处理模系矩阵分裂迭代方法.先通过变量替换将线性互补问题转化为一类非线性方程组,然后选取Strang或T.Chan循环矩阵作为预优矩阵,利用共轭梯度法进行求解.我们分析了该方法的收敛性.数值实验表明,该方法是高效可行的.     

Transformation of an axialsymmetric disk problem for the helmholtz equation into an ordinary differential equation

The problem of solving the three-dimensional Helmholtz equation in the exterior of a circular disk is considered where radially symmetric Dirichlet data on the disk are assumed to be prescribed. This problem for example arises in the scattering of plane (sound) waves at an infinite plane screen with a circular aperture if the direction of the incident wave is normal to the screen, as well as in the process of diffusion through a circular hole. By applying the factorization technique developed in [N. GORENFLO, M. WERNER,Solution of a finite convolution equation with a Hankel kernel by matrix factorization, SIAM J. Math. Anal., 28 (1997), pp. 434–451] to the disk problem an equivalent ordinary differential equation is derived, whose solution leads directly to the solution of the disk problem. This differential equation belongs to a class of ordinary differential equations which are of higher complexity than the standard ordinary differential equations of mathematical physics. The examination of this new class of differential equations therefore is motivated.     

A CQM-based BEM for transient heat conduction problems in homogeneous materials and FGMs

A method for numerical solution of time-domain boundary integral formulations of transient problems governed by the heat equation is presented. The heat conduction problem is analyzed considering homogeneous and non-homogeneous media. In the case of the non-homogeneous media, the conductor material is assumed to be a functionally graded material, i.e., the material properties vary spatially according to known smooth functions. For some specific spatial variations of the material properties, the fundamental solution and the boundary integral equation of the problem are obtained thanks to a change of variables that transforms the original problem to the standard heat conduction problem for homogeneous materials. For the treatment of time-dependent terms, the convolution quadrature method is adopted to approximate numerically the integral equation of the time-domain boundary element method. In the case that the responses are required at a large number of interior points, the convolution performed to calculate them is very time consuming. It is shown that the discrete convolution of the proposed formulation can be computed by means of the fast Fourier transform technique, which considerably reduces the computational complexity. Results for some transient heat conduction examples are presented to validate the numerical techniques studied.     

An efficient pseudo‐spectral Legendre–Galerkin method for solving a nonlinear partial integro‐differential equation arising in population dynamics

The pseudo‐spectral Legendre–Galerkin method (PS‐LGM) is applied to solve a nonlinear partial integro‐differential equation arising in population dynamics. This equation is a competition model in which similar individuals are competing for the same resources. It is a kind of reaction–diffusion equation with integral term corresponding to nonlocal consumption of resources. The proposed method is based on the Legendre–Galerkin formulation for the linear terms and interpolation operator at the Chebyshev–Gauss–Lobatto (CGL) points for the nonlinear terms. Also, the integral term, which is a kind of convolution, is directly computed by a fast and accurate method based on CGL interpolation operator, and thus, the use of any quadrature formula in its computation is avoided. The main difference of the PS‐LGM presented in the current paper with the classic LGM is in treating the nonlinear terms and imposing boundary conditions. Indeed, in the PS‐LGM, the nonlinear terms are efficiently handled using the CGL points, and also the boundary conditions are imposed strongly as collocation methods. Combination of the PS‐LGM with a semi‐implicit time integration method such as second‐order backward differentiation formula and Adams‐Bashforth method leads to reducing the complexity of computations and obtaining a linear algebraic system of equations with banded coefficient matrix. The desired equation is considered on one and two‐dimensional spatial domains. Efficiency, accuracy, and convergence of the proposed method are demonstrated numerically in both cases. Copyright © 2012 John Wiley & Sons, Ltd.     

非线性Volterra积分方程非平凡解的逼近方法及其应用

本文对一类具有卷积核的非线性Ｖｏｌｔｅｒｒａ型积分方程进行了讨论，给出了关于这类方程的非平凡解的存在性和解的逼近方法的一些结果。通过有关微分方程问题的实例，说明了所给结果的重要应用。     

Some integral relations of Hankel transform type and applications to elasticity theory

The dominant part of an integral equation arising in connection with boundary value problems for the circular disc is evaluated in terms of orthogonal polynomials. This relation leads to an efficient method for numerical solution of the complete integral equation even in the presence of a complicated bounded kernel. The static problem of a circular crack in an infinite elastic body under general loads is used to illustrate vector boundary conditions leading to two coupled integral equations, while the problem of a vibrating flexible circular plate in frictionless contact with an elastic half space is solved by use of the associated numerical method.     

Solving singular convolution equations using the inverse Fast Fourier Transform

The inverse Fast Fourier Transform is a common procedure to solve a convolution equation provided the transfer function has no zeros on the unit circle. In our paper we generalize this method to the case of a singular convolution equation and prove that if the transfer function is a trigonometric polynomial with simple zeros on the unit circle, then this method can be extended.     

Determination of an unknown function in a parabolic inverse problem by Sinc‐collocation method

In this article, an inverse problem of determining an unknown time‐dependent source term of a parabolic equation is considered. We change the inverse problem to a Volterra integral equation of convolution‐type. By using Sinc‐collocation method, the resulting integral equation is replaced by a system of linear algebraic equations. The convergence analysis is included, and it is shown that the error in the approximate solution is bounded in the infinity norm by the condition number and the norm of the inverse of the coefficient matrix multiplied by a factor that decays exponentially with the size of the system. Some examples are given to demonstrate the computational efficiency of the method. © 2010 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq 27: 1584–1598, 2010     

A fast numerical method for a natural boundary integral equation for the Helmholtz equation

A Neumann boundary value problem of the Helmholtz equation in the exterior circular domain is reduced into an equivalent natural boundary integral equation. Using our trigonometric wavelets and the Galerkin method, the obtained stiffness matrix is symmetrical and circulant, which lead us to a fast numerical method based on fast Fourier transform. Furthermore, we do not need to compute the entries of the stiffness matrix. Especially, our method is also efficient when the wave number k in the Helmholtz equation is very large.     

A Galerkin boundary element method based on interpolatory Hermite trigonometric wavelets

A Galerkin boundary element method based on interpolatory Hermite trigonometric wavelets is presented for solving 2-D potential problems defined inside or outside of a circular boundary in this paper. In this approach, an equivalent variational form of the corresponding boundary integral equation for the potential problem is used; the trigonometric wavelets are employed as trial and test functions of the variational formulation. The analytical formulae of the matrix entries indicate that most of the matrix entries are naturally zero without any truncation technique and the system matrix is a block diagonal matrix. Each block consists of four circular submatrices. Hence the memory spaces and computational complexity of the system matrix are linear scale. This approach could be easily coupled into domain decomposition method based on variational formulation. Finally, the error estimates of the approximation solutions are given and some test examples are presented.     

A novel matrix method for coupled vibration and damping effect analyses of liquid-filled circular cylindrical shells with partially constrained layer damping under harmonic excitation

A novel matrix method is further developed for a liquid-filled circular cylindrical shell with partially constrained layer damping (CLD), which consists of treating liquid domain with Bessel function approach, and shell domain with transfer matrix equation based on a new set of first order matrix differential equation. In order to indicate its advantage to the finite element method (FEM), free vibration analysis on such an empty shell under clamped–clamped boundary are carried out by using the present method together with FEM. Meanwhile, coincident result is yielded for the liquid-filled shell by adopting the present method with by other transfer matrix method. Finally, a series of valuable numerical results are obtained by this method.