时间：12月5号10:00-11:30

地点：博识楼433

报告人简介：

陈果良，华东师范大学教授、博导。主要研究方向为数值代数和矩阵论。主持完成多项国家自然科学基金项目、上海市科委重点项目、上海市教委重点项目。在国内外重要学术期刊发表论文近160篇、已培养博士及博士后10多位。曾担任过两届上海市工业与应用数学会副理事长，获得过上海市园丁奖、先进工作者等荣誉。担任过教育部全国数学学科评估通讯评审专家和数学学科博士点通讯评审专家等。

报告题目：

Some iterative methods for saddle point problems

摘要：

In this talk, we give the semi-convergent analysis of the Uzawa-SOR method, the GPIU method, GPPIU method and RHSS method when they are applied to solve the singular saddle point problem. Because the eigenvalue distribution of the system matrix is very important to the convergence of Krylov subspace method, so we also propose two different kinds of preconditioners for generalized saddle point problems and analyze the spectral properties of corresponding preconditioned matrix. Illustrative numerical examples are presented.