The extended nonsymmetric block Lanczos methods for solving large-scale differential Lyapunov equations

In this paper, we present a new approach for solving large-scale differential Lyapunov equations.  The proposed approach is based on projection of the initial problem onto an extended block Krylov subspace by using  extended nonsymmetric block Lanczos algorithm then, we get a low-dimensional differential Lyapunov matrix equation.  The latter differential matrix equation is solved by the Backward Differentiation Formula method (BDF) or Rosenbrock method (ROS), the obtained solution allows to build a low-rank approximate solution of the original problem.  Moreover, we also give some theoretical results.  The numerical results demonstrate the performance of our approach.

extended block Krylov subspace
extended nonsymmetric block Lanczos algorithm
low-rank approximation
differential Lyapunov equations
  1. Abou-Kandil H., Freiling G., Ionescu V., Jank G.  Matrix Riccati equations in control and systems theory. Birkhäuser (2012).
  2. Datta B.  Numerical methods for linear control systems (Vol. 1). Academic Press (2004).
  3. Sadek L., Talibi Alaoui H.  The extended block Arnoldi method for solving generalized differential Sylvester equations.  Journal of Mathematical Modeling. 8 (2), 189–206 (2020).
  4. Sadek E. M., Bentbib A. H., Sadek L., Talibi Alaoui  H.  Global extended Krylov subspace methods for large-scale differential Sylvester matrix equations.  J. Appl. Math. Comput. 62, 157–177 (2020).
  5. Agoujil S., Bentbib A. H., Jbilou K., Sadek E. M.  A minimal residual norm method for large-scale Sylvester matrix equations.  Electronic Transactions on Numerical Analysis. 43, 45–59 (2014).
  6. Hached M., Jbilou K.  Numerical solutions to large-scale differential Lyapunov matrix equations.  Numer Algor. 79, 741–757 (2018).
  7. Sadek L., Talibi Alaoui H.  Numerical methods for solving large-scale systems of differential equations.  Ricerche di Matematica.  (2021).
  8. Bentbib A., Jbilou K., Sadek E. M.  On some Krylov subspace based methods for large-scale nonsymmetric algebraic Riccati problems.  Computers & Mathematics with Applications. 70 (10), 2555–2565 (2015).
  9. Mena H., Ostermann A., Pfurtscheller L. M., Piazzola C.  Numerical low-rank approximation of matrix differential equations.  Journal of Computational and Applied Mathematics. 340, 602–614 (2018).
  10. Bentbib A. H., Jbilou K., Sadek E. L.  On some extended block Krylov based methods for large scale nonsymmetric Stein matrix equations.  Mathematics. 5 (2), 21  (2017).
  11. Behr M., Benner P., Heiland J.  Solution formulas for differential Sylvester and Lyapunov equations.  Calcolo. 56, 51 (2019).
  12. Lang N., Mena H., Saak J.  On the benefits of the LDLT factorization for large-scale differential matrix equation solvers.  Linear Algebra and its Applications. 480, 44–71 (2015).
  13. Druskin V., Knizhnerman L.  Extended Krylov subspaces: approximation of the matrix square root and related functions.  SIAM Journal on Matrix Analysis and Applications. 19 (3), 755–771 (1998).
  14. Barkouki H., Bentbib A. H., Heyouni M., Jbilou K.  An extended nonsymmetric block Lanczos method for model reduction in large scale dynamical systems.  Calcolo. 55, 13 (2018).
  15. Parlett B. N., Taylor D. R., Li Z. A.  A look-ahead Lanczos algorithm for unsymmetric matrices.  Math. Comput. 44 (169), 105–124 (1985).
  16. Penzl T. L  YAPACK A MATLAB Toolbox for Large Lyapunov and Riccati Equations,  Model Reduction Problems, and Linear-Quadratic Optimal Control Problems.  Available online: http://www.tuchemintz.de/sfb393/lyapack.