COMPARING ALGORITHMS FOR SOLVING SPARSE NONLINEAR-SYSTEMS OF EQUATIONS
COMPARING ALGORITHMS FOR SOLVING SPARSE NONLINEAR-SYSTEMS OF EQUATIONS
复制标题
DOI：
10.1137/0913025
复制
发表时间：
1992-03-01
期刊：
影响因子：
--
通讯作者：
MORETTI, AC
中科院分区：
文献类型：
--
作者：
GOMESRUGGIERO, MA;MARTINEZ, JM;MORETTI, AC
本文描述了求解大型非线性方程稀疏系统的八种牛顿型和准牛顿型算法的实现。对于线性代数计算，使用符号操作，以及George和Ng最近引入的静态数据结构，它允许求解线性系统的部分枢轴策略。给出了实现方法的数值比较。
This paper describes implementations of eight algorithms of Newton and quasi-Newton type for solving large sparse systems of nonlinear equations. For linear algebra calculations, a symbolic manipulation is used, as well as a static data structure introduced recently by George and Ng, which allows a partial pivoting strategy for solving linear systems. A numerical comparison of the implemented methods is presented.