二阶系统特征值反问题的数值求解研究-计算数学专业论文.docxVIP

摘 要 二阶系统特征值反问题的研究具有重要的理论和实践意义。二阶系统特征 值反问题是研究在一定的条件限制下，如何根据特征值、特征向量等信息确定 质量矩阵、刚度矩阵及阻尼矩阵。二阶系统特征值反问题来自于处理诸如固体 力学、自动控制等数学物理反问题离散化的过程中。因此，该问题研究不仅充 实和丰富了矩阵理论，而且在许多生产实践领域都具有重要指导作用。 本文从二阶系统特征值反问题的本质出发，研究了二阶系统特征值反问题 的求解方法，即研究如何利用已知的特征信息得到二阶系统的具体参数表示（质 量矩阵、阻尼矩阵和刚度矩阵）。首先，基于 Lancaster 结构理论，针对给定的 二阶系统特征对信息，将求解二阶系统具体参 数表示的问题转化为齐次 Sylvester 方程的非奇异对称解的求解问题。其次，为了简化计算，利用矩阵的 Kronecker 积理论将齐次 Sylvester 方程的求解问题进一步转化为齐次线性方程 组的求解问题。再次，从理论上分析证明了二阶系统特征值反问题中涉及到的 齐次 Sylvester 方程必存在非零解，并且对解空间进行了相应的描述与刻画。最 后，利用 Matlab 数学程序语言，对算法进行了计算机实现，并给出一具体实例， 证明了算法的可行性。 关键词：二阶系统；特征值反问题；Sylvester 方程；Kronecker 积 Abstract The study of quadratic system inverse eigenvalue problem has important significance theoretically and practically. It aims to find out how to determine parameters of mass matrix, stiffness matrix and damping matrix according to given eigenvalues, eigenvectors and other information. The problem arises from discretization of inverse problems in mathematics and physics areas such as solid mechanics and automatic control system. So the study of this problem not only enriches the matrix theory, but also offers guidelines to practical production. This paper, from the bases of the nature of quadratic system inverse eigenvalue problem, has discussed how to use the given eigenvalues and eigenvectors to work out parameters of mass matrix, stiffness matrix and damping matrix. Firstly, by applying Lancaster structure theory to the work, the problem of finding relevant parameters is transformed into solving non-singular symmetric solutions of homogeneous Sylvester equation. Next, this work is further simplified into solving homogeneous linear equations by the use of Kronecker product matrix theory. Then, it is proved that the non-zero solution must exist for the homogeneous Sylvester equation involved in the quadratic system inverse eigenvalue problem, and the corresponding description and characterization of the solution space are offered. Finally, implementation of such algorithm is achieved making use