毕业论文--矩阵方程AX+XB=D的极小范数最小二乘解.doc

PAGE PAGE 14 矩阵方程AX+XB=D的极小范数最小二乘解 摘 要 矩阵理论既是学习经典数学的基础，又是一门最有实用价值的数学理论。它不仅是数学的一个重要的分支，而且也已经成为现代各科技领域处理大量有限维空间形式与数量关系的强有力的工具。特别是计算机的广泛应用，为矩阵论的应用开辟了广阔的前景。例如，系统工程、优化方法以及稳定性理论等，都与矩阵论有着密切的联系。 当前，在矩阵理论领域，对矩阵方程解的研究一直是最热点的问题之一，矩阵方程及其解的问题在生物学、电学、光子光谱学、振动理论、线性最优控制等很多领域都有重要的应用。矩阵方程解的研究与探索更是没有间断过，可见该方程的求解问题确实是一个非常重要的课题。 本文将围绕该命题展开讨论，利用矩阵的直积（Kronecker积）、按行拉直和矩阵的满秩分解等算法解出矩阵方程的极小范数最小二乘解，并且它可由Moore-Penrose逆表出。 关键词：矩阵方程 极小范数最小二乘解 矩阵的直积（Kroneck积） 满秩分解算 法 Moore-Penrose逆 ABSTRACT Matrix theory is the foundation of learning classical mathematics, as well as one of the most meaningful mathematical theories. It is not only an important branch of mathematics, but also has become a powerful tool of handling many relationships between the finite dimension space structures and quantities in varity fields of modern technology. The extensive use of computer has brought a bright prospect to the application of matrix theory. Many problem have close relation with matrix theory, such as system engineering, optimization method, stability theory, and so on. Nowadays, the research on matrix equations has been turning into one of the hottest topics in matrix theory . It is widely used in different areas, for example, biology, electrics, spectroscopy, vibration theory, linear optimal control, etc. And people have researched and explored the solution of matrix equation all the time, which means the subject study is absolutely crucial. This paper discusses the available solutions of matrix equation, using Kronecker product, full rank decomposition of matrix to obtain the minimal norm least squares solution of matrix equation . And it can be expressed by Moore-Penrose inverse. Key words: matrix equation; the minimum norm least squares solution; Kronecker product; full rank factorization algorithm; Moore-Penrose inverse 目 录 TOC \o 1-3 \h \z \u HYPERLINK \l _Toc326618053 摘 要 PAGEREF _Toc326618053 \h 1 HYPERLINK \l _Toc326618054 ABSTRACT PAGEREF _Toc326618054 \h 2 HYPERLINK \l _Toc3266