Truncated low-rank methods for solving general linear matrix equations

被引:25
作者
Kressner, Daniel
Sirkovic, Petar [1 ,2 ]
机构
[1] EPF Lausanne, Chair Numer Algorithms, CH-1015 Lausanne, Switzerland
[2] EPF Lausanne, MATHICSE, HPC, CH-1015 Lausanne, Switzerland
关键词
general linear matrix equation; Lyapunov equation; greedy low-rank; generalized Lyapunov equation; Galerkin projection; KRYLOV SUBSPACE METHODS; LYAPUNOV EQUATIONS; REDUCTION; SYSTEMS;
D O I
10.1002/nla.1973
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
This work is concerned with the numerical solution of large-scale linear matrix equations A1XB1T++AKXBKT=C. The most straightforward approach computes XRmxn from the solution of an mn x mn linear system, typically limiting the feasible values of m,n to a few hundreds at most. Our new approach exploits the fact that X can often be well approximated by a low-rank matrix. It combines greedy low-rank techniques with Galerkin projection and preconditioned gradients. In turn, only linear systems of size m x m and n x n need to be solved. Moreover, these linear systems inherit the sparsity of the coefficient matrices, which allows to address linear matrix equations as large as m = n = O(10(5)). Numerical experiments demonstrate that the proposed methods perform well for generalized Lyapunov equations. Even for the case of standard Lyapunov equations, our methods can be advantageous, as we do not need to assume that C has low rank. Copyright (c) 2015 John Wiley & Sons, Ltd.
引用
收藏
页码:564 / 583
页数:20
相关论文
共 29 条
[1]   A new family of solvers for some, classes of multidimensional partial differential equations encountered in kinetic theory modeling of complex fluids [J].
Ammar, A. ;
Mokdad, B. ;
Chinesta, F. ;
Keunings, R. .
JOURNAL OF NON-NEWTONIAN FLUID MECHANICS, 2006, 139 (03) :153-176
[2]   On the decay rate of Hankel singular values and related issues [J].
Antoulas, AC ;
Sorensen, DC ;
Zhou, Y .
SYSTEMS & CONTROL LETTERS, 2002, 46 (05) :323-342
[3]   A projection method for model reduction of bilinear dynamical systems [J].
Bai, Zhaojun ;
Skoogh, Daniel .
LINEAR ALGEBRA AND ITS APPLICATIONS, 2006, 415 (2-3) :406-425
[4]   Adaptive low-rank approximation of collocation matrices [J].
Bebendorf, M ;
Rjasanow, S .
COMPUTING, 2003, 70 (01) :1-24
[5]  
Benner P., 2013, GAMM-Mitteilungen, V36, P32, DOI [10.1002/gamm.201310003, DOI 10.1002/GAMM.201310003]
[6]   Low rank methods for a class of generalized Lyapunov equations and related issues [J].
Benner, Peter ;
Breiten, Tobias .
NUMERISCHE MATHEMATIK, 2013, 124 (03) :441-470
[7]   INTERPOLATION-BASED H2-MODEL REDUCTION OF BILINEAR CONTROL SYSTEMS [J].
Benner, Peter ;
Breiten, Tobias .
SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 2012, 33 (03) :859-885
[8]   LYAPUNOV EQUATIONS, ENERGY FUNCTIONALS, AND MODEL ORDER REDUCTION OF BILINEAR AND STOCHASTIC SYSTEMS [J].
Benner, Peter ;
Damm, Tobias .
SIAM JOURNAL ON CONTROL AND OPTIMIZATION, 2011, 49 (02) :686-711
[9]   On the ADI method for Sylvester equations [J].
Benner, Peter ;
Li, Ren-Cang ;
Truhar, Ninoslav .
JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS, 2009, 233 (04) :1035-1045
[10]   CONVERGENCE OF A GREEDY ALGORITHM FOR HIGH-DIMENSIONAL CONVEX NONLINEAR PROBLEMS [J].
Cances, Eric ;
Ehrlacher, Virginie ;
Lelievre, Tony .
MATHEMATICAL MODELS & METHODS IN APPLIED SCIENCES, 2011, 21 (12) :2433-2467