A New Acceleration of the Projection Method in Nonsymmetric Eigenvalue Problems
نویسنده
چکیده
This study proposes a new method for the acceleration of the projection method to compute a few eigenvalues with the largest real parts of a large nonsymmetric matrix. In the eld of the solution of the linear system, an acceleration using the least squares polynomial which minimizes its norm on the boundary of the convex hull formed with the unwanted eigenvalues are proposed. We simplify this method for the eigenvalue problem using the same property of the orthogonal polynomial. This study applies the Tchebychev polynomial to the iterative Arnoldi method and proves that the new method computes necessary eigenvalues with far less complexity than the QR method. Its high accuracy enables us to compute the close eigenvalues that can not be obtained by the simple Arnoldi method.
منابع مشابه
A Polynomial Acceleration of the Projection Method for Large Nonsymmetric Eigenvalue Problems
This study proposes a method for the acceleration of the projection method to compute a few eigenvalues with the largest real parts of a large nonsymmetric matrix. In the eld of the solution of the linear system, an acceleration using the least squares polynomial which minimizes its norm on the boundary of the convex hull formed with the unwanted eigenvalues are proposed. We simplify this metho...
متن کاملA New Hybrid Conjugate Gradient Method Based on Eigenvalue Analysis for Unconstrained Optimization Problems
In this paper, two extended three-term conjugate gradient methods based on the Liu-Storey ({tt LS}) conjugate gradient method are presented to solve unconstrained optimization problems. A remarkable property of the proposed methods is that the search direction always satisfies the sufficient descent condition independent of line search method, based on eigenvalue analysis. The globa...
متن کاملSome new restart vectors for explicitly restarted Arnoldi method
The explicitly restarted Arnoldi method (ERAM) can be used to find some eigenvalues of large and sparse matrices. However, it has been shown that even this method may fail to converge. In this paper, we present two new methods to accelerate the convergence of ERAM algorithm. In these methods, we apply two strategies for the updated initial vector in each restart cycles. The implementation of th...
متن کاملImplicitly Restarted Arnoldi/lanczos Methods for Large Scale Eigenvalue Calculations
This report provides an introductory overview of the numerical solution of large scale algebraic eigenvalue problems. The main focus is on a class of methods called Krylov subspace projection methods. The Lanczos method is the premier member of this class and the Arnoldi method is a generalization to the nonsymmetric case. A recently developed and very promising variant of the Arnoldi/Lanczos s...
متن کاملA New Inexact Inverse Subspace Iteration for Generalized Eigenvalue Problems
In this paper, we represent an inexact inverse subspace iteration method for computing a few eigenpairs of the generalized eigenvalue problem Ax = Bx [Q. Ye and P. Zhang, Inexact inverse subspace iteration for generalized eigenvalue problems, Linear Algebra and its Application, 434 (2011) 1697-1715 ]. In particular, the linear convergence property of the inverse subspace iteration is preserved.
متن کامل