Avoiding complex arithmetic in the low-rank ADI method efficiently

We present a new reformulation of the low-rank ADI method for solving large-scale Lyapunov equations which uses only real arithmetic operation and storage in the presence of complex shift parameters. This makes the method applicable on computing environments where complex computations and storage are not supported or not efficiently available. For generalized Lyapunov equations it is significantly more efficient than the older completely real formulation of the low-rank ADI as confirmed by numerical examples. (© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

A Reformulated Low-Rank ADI Iteration with Explicit Residual Factors

Based on our recent findings [1, 2] regarding the low-rank ADI iteration for large-scale Lyapunov equations, we propose a mathematically equivalent formulation of the LR-ADI whose iteration is directly connected to low rank factors of the Lyapunov residual. If complex shift parameters occur and are handled efficiently [1], this reformulation is slightly more efficient from a computational point of view, since it guarantees that the right hand sides of all linear systems that need to be solved are real. (© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

On the numerical solution of large-scale sparse discrete-time Riccati equations

We discuss the numerical solution of large-scale discrete-time algebraic Riccati equations (DAREs) as they arise, e.g., in fully discretized linear-quadratic optimal control problems for parabolic partial differential equations (PDEs). We employ variants of Newton??s method that allow to compute an approximate low-rank factor of the solution of the DARE. The principal computation in the Newton iteration is the numerical solution of a Stein (aka discrete Lyapunov) equation in each step. For this purpose, we present a low-rank Smith method as well as a low-rank alternating-direction-implicit (ADI) iteration to compute low-rank approximations to solutions of Stein equations arising in this context. Numerical results are given to verify the efficiency and accuracy of the proposed algorithms.     

A Modified Low-Rank Smith Method for Large-Scale Lyapunov Equations

In this note we present a modified cyclic low-rank Smith method to compute low-rank approximations to solutions of Lyapunov equations arising from large-scale dynamical systems. Unlike the original cyclic low-rank Smith method introduced by Penzl in [20], the number of columns required by the modified method in the approximate solution does not necessarily increase at each step and is usually much lower than in the original cyclic low-rank Smith method. The modified method never requires more columns than the original one. Upper bounds are established for the errors of the low-rank approximate solutions and also for the errors in the resulting approximate Hankel singular values. Numerical results are given to verify the efficiency and accuracy of the new algorithm.     

An LDLT factorization based ADI algorithm for solving large-scale differential matrix equations

Large-scale differential matrix equations appear in many applications like optimal control of partial differential equations, balanced truncation model order reduction of linear time varying systems etc. Here, we will focus on matrix Riccati differential equations (RDE). Solving such matrix valued ordinary differential equations (ODE) is a highly storage and time consuming process. Therefore, it is necessary to develop efficient solution strategies minimizing both. We present an LDL^T factorization based ADI method for solving algebraic Lyapunov equations (ALE) arising in the innermost iteration during the application of Rosenbrock ODE solvers to RDEs. We show that the LDL^T-type decomposition avoids complex arithmetic, as well as cancellation effects arising from indefinite right hand sides of the ALEs appearing in the classic ZZ^T based approach. Additionally, a certain number of linear system solves can be saved within the ADI algorithm by reducing the number of column blocks in the right hand sides while the full accuracy of the standard low-rank ADI is preserved. (© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Parallel algorithms for solving tridiagonal and near-circulant systems

Many problems in mathematics and applied science lead to the solution of linear systems having circulant coefficient matrices. This paper presents a new stable method for the exact solution of non-symmetric tridiagonal circulant linear systems of equations. The method presented in this paper is quite competitive with Gaussian elimination both in terms of arithmetic operations and storage requirements. It is also competitive with the modified double sweep method. This method can be applied to solve the near-circulant tridiagonal system. In addition, the method is modified to allow for parallel processing.     

Balanced truncation model order reduction in limited time intervals for large systems

In this article we investigate model order reduction of large-scale systems using time-limited balanced truncation, which restricts the well known balanced truncation framework to prescribed finite time intervals. The main emphasis is on the efficient numerical realization of this model reduction approach in case of large system dimensions. We discuss numerical methods to deal with the resulting matrix exponential functions and Lyapunov equations which are solved for low-rank approximations. Our main tool for this purpose are rational Krylov subspace methods. We also discuss the eigenvalue decay and numerical rank of the solutions of the Lyapunov equations. These results, and also numerical experiments, will show that depending on the final time horizon, the numerical rank of the Lyapunov solutions in time-limited balanced truncation can be smaller compared to standard balanced truncation. In numerical experiments we test the approaches for computing low-rank factors of the involved Lyapunov solutions and illustrate that time-limited balanced truncation can generate reduced order models having a higher accuracy in the considered time region.     

An improved numerical method for balanced truncation for symmetric second-order systems

We consider balanced truncation model order reduction for symmetric second-order systems. The occurring large-scale generalized and structured Lyapunov equations are solved with a specially adapted low-rank alternating directions implicit (ADI) type method. Stopping criteria for this iteration are investigated, and a new result concerning the Lyapunov residual within the low-rank ADI method is established. We also propose a goal-oriented stopping criterion which tries to incorporate the balanced truncation approach already during the ADI iteration. The model reduction approach using the ADI method with different stopping criteria is evaluated on several test systems.     

A low-rank Krylov squared Smith method for large-scale discrete-time Lyapunov equations

The squared Smith method is adapted to solve large-scale discrete-time Lyapunov matrix equations. The adaptation uses a Krylov subspace to generate the squared Smith iteration in a low-rank form. A restarting mechanism is employed to cope with the increase of memory storage of the Krylov basis. Theoretical aspects of the algorithm are presented. Several numerical illustrations are reported.     

On the ADI method for Sylvester equations

This paper is concerned with the numerical solution of large scale Sylvester equations AX−XB=C, Lyapunov equations as a special case in particular included, with C having very small rank. For stable Lyapunov equations, Penzl (2000) [22] and Li and White (2002) [20] demonstrated that the so-called Cholesky factor ADI method with decent shift parameters can be very effective. In this paper we present a generalization of the Cholesky factor ADI method for Sylvester equations. An easily implementable extension of Penz’s shift strategy for the Lyapunov equation is presented for the current case. It is demonstrated that Galerkin projection via ADI subspaces often produces much more accurate solutions than ADI solutions.     

A comparison of iterative methods to solve complex valued linear algebraic systems

Complex valued linear algebraic systems arise in many important applications. We present analytical and extensive numerical comparisons of some available numerical solution methods. It is advocated, in particular for large scale ill-conditioned problems, to rewrite the complex-valued system in real valued form leading to a two-by-two block system of particular form, for which it is shown that a very efficient and robust preconditioned iterative solution method can be constructed. Alternatively, in many cases it turns out that a simple preconditioner in the form of the sum of the real and the imaginary part of the matrix also works well but involves complex arithmetic.     

Finding roots of a real polynomial simultaneously by means of Bairstow's method

Aberth's method for finding the roots of a polynomial was shown to be robust. However, complex arithmetic is needed in this method even if the polynomial is real, because it starts with complex initial approximations. A novel method is proposed for real polynomials that does not require any complex arithmetic within iterations. It is based on the observation that Aberth's method is a systematic use of Newton's method. The analogous technique is then applied to Bairstow's procedure in the proposed method. As a result, the method needs half the computations per iteration than Aberth's method. Numerical experiments showed that the new method exhibited a competitive overall performance for the test polynomials.     

Solution of complex matrix equations with changing phase

A method is presented for solving a succession of complex matrix equations in which the phase of the real and imaginary components changes. The method is more efficient than the technique obtained by using complex Gaussian elimination on each of the matrix equations separately. In addition, some interesting theoretical relationships are presented for the solution of complex matrix equations in general, using only real-valued arithmetic operations.This work was performed while the author was with the Electrical Engineering Department, McGill University, Montreal, Canada. Financial support was provided by the National Research Council of Canada.     

The parameterized upper and lower triangular splitting methods for saddle point problems

Complex moment-based eigensolvers for solving interior eigenvalue problems have been studied because of their high parallel efficiency. Recently, we proposed the block Arnoldi-type complex moment-based eigensolver without a low-rank approximation. A low-rank approximation plays a very important role in reducing computational cost and stabilizing accuracy in complex moment-based eigensolvers. In this paper, we develop the method and propose block Krylov-type complex moment-based eigensolvers with a low-rank approximation. Numerical experiments indicate that the proposed methods have higher performance than the block SS–RR method, which is one of the most typical complex moment-based eigensolvers.     

Efficient fourth orderP-stable formulae

In this paper, a family of fourth orderP-stable methods for solving second order initial value problems is considered. When applied to a nonlinear differential system, all the methods in the family give rise to a nonlinear system which may be solved using a modified Newton method. The classical methods of this type involve at least three (new) function evaluations per iteration (that is, they are 3-stage methods) and most involve using complex arithmetic in factorising their iteration matrix. We derive methods which require only two (new) function evaluations per iteration and for which the iteration matrix is a true real perfect square. This implies that real arithmetic will be used and that at most one real matrix must be factorised at each step. Also we consider various computational aspects such as local error estimation and a strategy for changing the step size.     

Factorized solution of the Lyapunov equation by using the hierarchical matrix arithmetic

We investigate the solution of the Lyapunov equation with the matrix sign function method. In order to obtain the factorized solution we use a partitioned Newton iteration, where one part of the iteration uses formatted arithmetic for the hierarchical matrix format while the other part converges to an approximate full‐rank factor of the solution. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Error Bounds for Low-Rank Approximations of the First Exponential Integral Kernel

A hierarchical matrix is an efficient data-sparse representation of a matrix, especially useful for large dimensional problems. It consists of low-rank subblocks leading to low memory requirements as well as inexpensive computational costs. In this work, we discuss the use of the hierarchical matrix technique in the numerical solution of a large scale eigenvalue problem arising from a finite rank discretization of an integral operator. The operator is of convolution type, it is defined through the first exponential-integral function and, hence, it is weakly singular. We develop analytical expressions for the approximate degenerate kernels and deduce error upper bounds for these approximations. Some computational results illustrating the efficiency and robustness of the approach are presented.     

A hybrid iterative method for symmetric indefinite linear systems

Hybrid iterative methods that combine a conjugate direction method with a simpler iteration scheme, such as Chebyshev or Richardson iteration, were first proposed in the 1950s. The ease with which Chebyshev and Richardson iteration can be implemented efficiently on a large variety of computer architectures has in recent years lead to renewed interest in iterative methods that use Chebyshev or Richardson iteration. This paper presents a new hybrid iterative method for the solution of linear systems of equations with a symmetric indefinite matrix. Our method combines the conjugate residual method with Richardson iteration. Special attention is paid to the determination of two real intervals, one on each side of the origin, that contain most of the eigenvalues of the matrix. These intervals are used to compute suitable iteration parameters for Richardson iteration. We also discuss when to switch between the methods. The hybrid scheme typically uses the Richardson method for most iterations, and this reduces the number of arithmetic vector operations significantly compared with the number of arithmetic vector operations required when only the conjugate residual method is used. Computed examples illustrate the competitiveness of the hybrid scheme.