Title :
New solving method of eigenvalue under parallel process
Author :
Osano, M. ; Gotze, C.
Author_Institution :
Dept. of Electr. Eng., Tokyo Univ., Japan
Abstract :
Presents a new method for finding real and complex eigenvalues and eigenvectors. We call this method the pole Gaussian method, since it was inspired by the Gauss-Seidel method. The idea is to use the approximate distance between the eigenvalue poles on the complex plane to iteratively calculate the eigenvalues and eigenvectors. A proof of convergence is presented in this paper, and it is shown that, in some cases, accurate solutions can be obtained by a small number of iterations. The dependence of the convergence speed on the initial values is also investigated.<>
Keywords :
convergence of numerical methods; eigenvalues and eigenfunctions; iterative methods; poles and zeros; stability; Gauss-Seidel method; accurate solutions; complex plane; convergence proof; convergence speed; eigenvalue poles; eigenvectors; initial values; iterative calculation; parallel processing; pole Gaussian method; Cities and towns; Concurrent computing; Eigenvalues and eigenfunctions; Equations; Gaussian processes; Iterative algorithms; Linear systems; Newton method; Software; Stability analysis;
Conference_Titel :
TENCON '93. Proceedings. Computer, Communication, Control and Power Engineering.1993 IEEE Region 10 Conference on
Conference_Location :
Beijing, China
Print_ISBN :
0-7803-1233-3
DOI :
10.1109/TENCON.1993.320016
