arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1403.7013math.NAcs.NAmath.OC
收藏

非线性HSS-like迭代法用于绝对值方程

The nonlinear HSS-like iteration method for absolute value equations

School of Mathematics and Statistics, Lanzhou University(兰州大学数学与统计学院);School of Mathematics and Statistics, Hexi University(河西学院数学与统计学院);School of Chemistry and Chemical Engineering, Hexi University(河西学院化学化工学院)

Mu-Zheng Zhu, Ya-E Qi

AI总结:

本文提出非线性HSS-like迭代法以改进Picard-HSS方法,通过单层迭代结构提升效率,数值实验验证其有效性。

摘要:

Salkuyeh提出了求解绝对值方程(AVE)的Picard-HSS迭代法,该问题属于非光滑NP难问题。为提升性能,本文提出非线性HSS-like迭代法。与Picard-HSS的内外双层结构不同,HSS-like仅采用单层结构,可及时更新迭代向量。数值实验表明该方法可行、稳健且有效。

英文摘要:

Salkuyeh proposed the Picard-HSS iteration method to solve the absolute value equation (AVE), which is a class of non-differentiable NP-hard problem. To further improve its performance, a nonlinear HSS-like iteration method is proposed. Compared to that the Picard-HSS method is an inner-outer double-layer iteration scheme, the HSS-like iteration is only a monolayer and the iteration vector could be updated timely. Some numerical experiments are used to demonstrate that the nonlinear HSS-like method is feasible, robust and effective.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑