Parallelization of Finite Element Analysis of Nonlinear Magnetic Fields Using GPU

Author

Okimura, T. ; Sasayama, Teruyoshi ; Takahashi, Naoyuki ; Ikuno, Soichiro

Author_Institution

Dept. of Electr. & Electron. Eng., Okayama Univ., Okayama, Japan

Volume

49

Issue

5

fYear

2013

fDate

May-13

Firstpage

1557

Lastpage

1560

Abstract

The acceleration of a nonlinear magnetic field analysis by parallelizing the finite element method (FEM) is examined using the graphics processing unit (GPU). It is shown that the speedup of the magnetic field analysis is realized by parallelizing the variable preconditioned conjugate gradient (VPCG) method. The Jacobi over-relaxation (JOR) method, conjugate residual (CR) method and conjugate gradient (CG) method are also applied in the variable preconditioning. The results of computations demonstrate that VPCG using the GPU significantly improve the performance. Especially, CG applied by variable preconditioned on GPU is 39 times faster than ICCG on a CPU.

Keywords

Jacobian matrices; conjugate gradient methods; finite element analysis; graphics processing units; magnetic fields; parallel programming; FEM; GPU; JOR; Jacobi over-relaxation method; VPCG; conjugate residual method; finite element analysis; finite element method; graphics processing unit; nonlinear magnetic field analysis; parallelization; variable preconditioned conjugate gradient method; Finite element method; graphics processing unit (GPU); parallel computing;

fLanguage

English

Journal_Title

Magnetics, IEEE Transactions on

Publisher

ieee

ISSN

0018-9464

Type

jour

DOI

10.1109/TMAG.2013.2244062

Filename

6514787