The corrected Uzawa method for solving saddle point problems

Summary In this paper, we consider the solution of linear systems of saddle point type by correcting the Uzawa algorithm, which has been proposed in [K. Arrow, L. Hurwicz, H. Uzawa, Studies in nonlinear programming, Stanford University Press, Stanford, CA, 1958]. We call this method as corrected Uzawa (CU) method. The convergence of the CU method is analyzed for solving nonsingular saddle point problem as well as the semi-convergence for the singular case. First, the corrected model for the Uzawa algorithm is established, and the CU algorithm is presented. Then we study the geometric meaning of the CU model. Moreover, we introduce the overall reduction coefficient α to measure the effect of the CU process. It is shown that the CU method converges faster than the Uzawa method and several other methods if the overall reduction coefficient α satisfies certain conditions. Numerical experiments are presented to illustrate the theoretical results and examine the numerical effectiveness of the CU method. Copyright © 2015 John Wiley & Sons, Ltd.

[1]  X. Wu,et al.  Conjugate Gradient Method for Rank Deficient Saddle Point Problems , 2004, Numerical Algorithms.

[2]  Y. Saad,et al.  GMRES: a generalized minimal residual algorithm for solving nonsymmetric linear systems , 1986 .

[3]  A. Wathen,et al.  Minimum residual methods for augmented systems , 1998 .

[4]  Nicholas I. M. Gould,et al.  On the Solution of Equality Constrained Quadratic Programming Problems Arising in Optimization , 2001, SIAM J. Sci. Comput..

[5]  Alla Sheffer,et al.  Preconditioners for Indefinite Linear Systems Arising in Surface Parameterization , 2001, International Meshing Roundtable Conference.

[6]  Qingqing Zheng,et al.  A class of accelerated Uzawa algorithms for saddle point problems , 2014, Appl. Math. Comput..

[7]  Gene H. Golub,et al.  Preconditioned Hermitian and skew-Hermitian splitting methods for non-Hermitian positive semidefinite linear systems , 2004, Numerische Mathematik.

[8]  Beresford N. Parlett,et al.  On generalized successive overrelaxation methods for augmented linear systems , 2005, Numerische Mathematik.

[9]  Andrew J. Wathen,et al.  Performance and analysis of saddle point preconditioners for the discrete steady-state Navier-Stokes equations , 2002, Numerische Mathematik.

[10]  Jun Zou,et al.  An Iterative Method with Variable Relaxation Parameters for Saddle-Point Problems , 2001, SIAM J. Matrix Anal. Appl..

[11]  Qingqing Zheng,et al.  A new SOR-Like method for the saddle point problems , 2014, Appl. Math. Comput..

[12]  Yimin Wei,et al.  Semi-convergence analysis of Uzawa methods for singular saddle point problems , 2014, J. Comput. Appl. Math..

[13]  Howard C. Elman,et al.  Algorithm 866: IFISS, a Matlab toolbox for modelling incompressible flow , 2007, TOMS.

[14]  Fang Chen,et al.  A generalization of the inexact parameterized Uzawa methods for saddle point problems , 2008, Appl. Math. Comput..

[15]  W. Jason,et al.  アルゴリズム869:ODRPACK95:範囲制約のある重み付け直交距離回帰コード , 2007 .

[16]  G. Golub,et al.  Inexact and preconditioned Uzawa algorithms for saddle point problems , 1994 .

[17]  Constantin Bacuta,et al.  A Unified Approach for Uzawa Algorithms , 2006, SIAM J. Numer. Anal..

[18]  Zeng-Qi Wang,et al.  On parameterized inexact Uzawa methods for generalized saddle point problems , 2008 .

[19]  Zhong-Zhi Bai,et al.  On semi-convergence of Hermitian and skew-Hermitian splitting methods for singular linear systems , 2010, Computing.

[20]  Wen Li,et al.  The alternating-direction iterative method for saddle point problems , 2010, Appl. Math. Comput..

[21]  Junfeng Lu,et al.  A Modified Nonlinear Inexact Uzawa Algorithm with a Variable Relaxation Parameter for the Stabilized Saddle Point Problem , 2010, SIAM J. Matrix Anal. Appl..

[22]  Gene H. Golub,et al.  SOR-like Methods for Augmented Systems , 2001 .

[23]  Gene H. Golub,et al.  Accelerated Hermitian and skew-Hermitian splitting iteration methods for saddle-point problems , 2007 .

[24]  Gene H. Golub,et al.  Optimal Parameter in Hermitian and Skew-Hermitian Splitting Method for Certain Two-by-Two Block Matrices , 2006, SIAM J. Sci. Comput..

[25]  Zhi-Hao Cao.  Comparison of performance of iterative methods for singular and nonsingular saddle point linear systems arising from Navier-Stokes equations , 2006, Appl. Math. Comput..

[26]  Gene H. Golub,et al.  Numerical solution of saddle point problems , 2005, Acta Numerica.

[27]  Apostol T. Vassilev,et al.  Analysis of the Inexact Uzawa Algorithm for Saddle Point Problems , 1997 .

[28]  Mingrong Cui.  A sufficient condition for the convergence of the inexact Uzawa algorithm for saddle point problems , 2002 .

[29]  Yang Cao,et al.  On local Hermitian and skew-Hermitian splitting iteration methods for generalized saddle point problems , 2009, J. Comput. Appl. Math..

[30]  Gene H. Golub,et al.  Hermitian and Skew-Hermitian Splitting Methods for Non-Hermitian Positive Definite Linear Systems , 2002, SIAM J. Matrix Anal. Appl..

[31]  Nira Dyn,et al.  The numerical solution of equality constrained quadratic programming problems , 1983 .

[32]  Gene H. Golub,et al.  Convergence properties of preconditioned Hermitian and skew-Hermitian splitting methods for non-Hermitian positive semidefinite matrices , 2007, Math. Comput..

[33]  C. Bacuta.  Schur complements on Hilbert spaces and saddle point systems , 2009 .

[34]  Ahmed H. Sameh,et al.  An Efficient Iterative Method for the Generalized Stokes Problem , 1998, SIAM J. Sci. Comput..

[35]  E. Haber,et al.  On optimization techniques for solving nonlinear inverse problems , 2000 .

[36]  Bing Zheng,et al.  On semi-convergence of parameterized Uzawa methods for singular saddle point problems☆ , 2009 .

[37]  Jun Zou,et al.  Nonlinear Inexact Uzawa Algorithms for Linear and Nonlinear Saddle-point Problems , 2006, SIAM J. Optim..