ADI preconditioned Krylov methods for large Lyapunov matrix equations (Q962079)

The author presents new preconditioned Krylov methods for solving large Lyapunov matrix equations \(AX+XA^T+BB^T=0\), based on Arnoldi's global alternating direction implicit iteration method LR-ADI(\(l\)). The original equation is transformed into a new symmetric Stein matrix equation with the same solution. Low rank approximate solutions are extracted by the global Arnoldi method. An upper bound for the norm of the error is presented. The numerical tests show the effectiveness of the method as compared with known methods. As the number \(l\) increases in LR-ADI(\(l\)), convergence is generally faster in terms of the number of iterations but requires higher CPU-time. It seems that the best results are obtained with \(l=1\) or \(l=2\).