摘要早在2000多年前,中国的古人就对线性方程组有所研究和记载。最早出现在公元初《九章算术》中,比欧洲的记载整整早了1500年。
线性方程组在线性代数中扮演着不可或缺的角色,随着数学和科技的不断发展进步,使其在生活中的各个领域有着广泛的应用,电子工程、软件开发、人员管理、交通运输等。当然在一些科学研究中,线性方程组也起着重要的辅助作用,在很多大型实验和调查后期的数据处理上,线性方程组的应用节约了很大的精力也大大提高了工作效率。因此,如何使用合理的计算量求解一个线性方程组就成了数值计算方法的一个重要课题。根据不同的方程组特点,我们要找到相对快捷和精准的解法。线性方程组的迭代方法是一种极限方法,尤其是对于解大型稀疏矩阵方程组而言快捷、有效。他的基本思路是:用某种极限过程去逐渐逼近线性方程组的精确解,是一种逐步逼近法求解的方法。本文主要介绍古典迭代法(Jacobi迭代法、Gauss-Seidel迭代法以及SOR迭代法)、简单了解 H—矩阵松弛型矩阵多分裂迭代法、 子空间迭代法,并对古典迭代法每一种方法的收敛性进行分析,比较Jacobi迭代法Gauss-Seidel迭代法以及SOR迭代法的收敛速度,探索,这几种迭代法的适用性。将Jacobi迭代法、Gauss-Seidel迭代法以及SOR迭代法在具体的题目中进行对比分析,并在MATLAB中运行。48381 毕业论文关键字:古典迭代法、收敛性、MATLAB、 H—矩阵松弛型矩阵多分裂迭代法、 子空间迭代法
ABSTRACT As early as 2000 years ago, the Chinese people to record and study on the linear system of equations. First appeared in the early party Cheng Zhang "nine chapter arithmetic", than the record 1500 years before the whole of Europe.
Linear equations plays an indispensable role in linear algebra, with the continuous development of mathematics and science and technology progress, make it has been widely used in various areas of life, electronic engineering, software development, personnel management, transportation, etc. In some scientific research system of linear equations, of course also plays an important supporting role, in many large later data processing experiment and study, a system of linear equations of the application to save a great deal of energy is also greatly improve the working efficiency. Therefore, how to use the reasonable calculation of solving a linear completed ?
Key words: classical iterative method ; efficient iterative method,;convergence ;MATLAB
目录