New preconditioners based on symmetric-triangular decomposition for saddle point problems

详细信息    查看全文

• 作者：Xiao-Fei Peng (1) pxf6628@163.com
  Wen Li (2) liwen@scnu.edu.cn
  Shuhuang Xiang (3) xiangsh@mail.csu.edu.cn
• 关键词：Symmetric and triangular decomposition – ; Triangular preconditioners – ; Saddle point problems – ; Condition number – ; Conjugate gradient method
• 刊名：Computing
• 出版年：2011
• 出版时间：September 2011
• 年：2011
• 卷：93
• 期：1
• 页码：27-46
• 全文大小：216.9 KB
• 参考文献：1. Bai Z-Z, Li G-Q (2003) Restrictively preconditioned conjugate gradient methods for systems of linear equations. IMA J Numer Anal 23: 561–580
  2. Bai Z-Z, Parlett BN, Wang Z-Q (2005) On generalized successive overrelaxation methods for augmented linear systems. Numer Math 102: 1–38
  3. Bai Z-Z, Wang Z-Q (2008) On parameterized inexact Uzawa methods for generalized saddle point problems. Linear Algebra Appl 428: 2900–2932
  4. Benzi M, Golub GH, Liesen J (2005) Numerical solution of saddle point problems. Acta Numer 14: 1–37
  5. Bramble JH, Pasciak JE (1988) A preconditioning technique for indefinite systems resulting from mixed approximations of elliptic problems. Math Comp 50: 1–17
  6. Dollar HS, Gould NIM, Stoll M, Wathen A (2010) Preconditioning saddle-point systems with applications in optimization. SIAM J Sci Comput 32: 249–270
  7. Brezzi F, Fortin M (1991) Mixed and hybrid finite element methods. Springer-Verlag, New York
  8. Chen XJ, Hashimoto K (2003) Numerical validation of solutions of saddle point matrix equations. Numer Linear Algebra Appl 10: 661–672
  9. Cordeiro C, Yuan JY (2006) Row-wise algorithms of ST decomposition. Appl Math Comput 175: 748–761
  10. Elman HC (1996) Multigrid and krylov subspace methods for the discrete stokes equations. Int J Numer Meths Fluids 22: 755–770
  11. Elman HC, Ramage A, Silvester DJ (2007) Algorithm 866: IFISS, a Matlab toolbox for modelling incompressible flow. ACM Trans Math Softw 33: 251–268
  12. Golub GH, Yuan JY (2002) Symmetric-triangular decomposition and its applications—Part I: Theorems and algorithms. BIT 42: 814–822
  13. Hestenes MR, Stiefel E (1952) Methods of conjugate gradients for solving linear systems. J Res Nat Bur Stand 49: 409–436
  14. Kimura T, Chen XJ (2008) Validated solutions of saddle point linear systems. SIAM J Matrix Anal Appl 30: 1697–1708
  15. Liesen J, Strakos Z (2010) On the computational cost of Krylov subspace methods for solving linear algebraic systems. Foundations of Computational Mathematics (submitted)
  16. Santiago C, Yuan JY (2003) Modified ST algorithms and numerical experiments. Appl Numer Math 47: 237–253
  17. Stoll M, Wathen A (2008) Combination preconditioning and the Bramble-Pasciak† preconditioner. SIAM J Matrix Anal Appl 30: 582–608
  18. Rees T, Stoll M, Wathen A (2010) All-at-once preconditioners for PDE-constrained optimization. Kybernetika 46: 341–360
  19. Wu XN, Golub GH, Cuminato JA, Yuan JY (2008) Symmetric-triangular decomposition and its applications—Part Π: Preconditioners for indefinite systems. BIT Numer Math 48: 139–162
  20. Wright S (1997) Stability of augmented system factorizations in interior-point methods. SIAM J Matrix Anal Appl 18: 191–222
  21. Yuan JY (1996) Numerical methods for generalized least squares problems. J Comput Appl Math 66: 571–584
• 作者单位：1. Nanhai College, South China Normal University, Foshan, 528225 People鈥檚 Republic of China2. School of Mathematical Sciences, South China Normal University, Guangzhou, 510631 People鈥檚 Republic of China3. Department of Applied Mathematics and Software, Central South University, Changsha, 410083 People鈥檚 Republic of China
• 刊物类别：Mathematics and Statistics
• 刊物主题：Mathematics
  Computational Mathematics and Numerical Analysis
  
• 出版者：Springer Wien
• ISSN：1436-5057

文摘

In this paper, we construct some new triangular preconditioners for saddle point problems based on the symmetric and triangular (ST) decomposition. Furthermore, we obtain some estimations on the condition number for the preconditioned systems and give the quasi-optimal parameters. Numerical experiments on the Stokes problems are given to illustrate fast convergence of the associated conjugate gradient method.