当前位置: X-MOL 学术Appl. Math. Comput. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A new constraint preconditioner based on the PGSS iteration method for non-Hermitian generalized saddle point problems
Applied Mathematics and Computation ( IF 4 ) Pub Date : 2020-12-25 , DOI: 10.1016/j.amc.2020.125864
Hongyu Wu , Shuhuang Xiang

For the non-Hermitian generalized saddle point problems, we propose a new constraint preconditioner. The new constraint preconditioner is constructed based on the preconditioned generalized shift-splitting (PGSS) iteration method. We also analyze the invertibility condition of the new preconditioner in detail. Moreover, the convergence properties of the new constraint preconditioning iteration method are derived. Finally, the effectiveness of the proposed preconditioner is illustrated by numerical experiments.



中文翻译:

基于PGSS迭代方法的非Hermitian广义鞍点问题新约束预条件子

对于非Hermitian广义鞍点问题,我们提出了一种新的约束预处理器。基于预处理的广义移位分裂(PGSS)迭代方法构造了新的约束预处理器。我们还将详细分析新预处理器的可逆性条件。此外,推导了新的约束预处理迭代方法的收敛性。最后,通过数值实验说明了所提出的预处理器的有效性。

更新日期:2020-12-25
down
wechat
bug