A Novel Algebraic Multigrid Preconditioning for Large-Scale Edge-Element Analyses

• Published in: IEEE transactions on magnetics Vol. 43; no. 4; pp. 1481 - 1484
• Main Authors: Mifune, T., Iwashita, T., Shimasaki, M.
• Format: Journal Article Conference Proceeding
• Language: English
• Published: New York, NY IEEE 01.04.2007; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algebra; Algebraic multigrid (AMG); Character generation; Coefficients; Cross-disciplinary physics: materials science; rheology; edge-elements; Equations; Exact sciences and technology; Finite element method; finite element method (FEM); Finite element methods; Iterative methods; Large-scale systems; Linear systems; Magnetic analysis; Magnetism; Magnetostatics; Materials science; Mathematical analysis; Mathematical models; Matrices; Matrix methods; Null space; Other topics in materials science; Physics; Preconditioning; Vectors; Finite element method; Algebraic multigrid (AMG); Calculus; Modelling; finite element method (FEM); Magnetic fields; edge-elements; Magnetic properties; preconditioning
• ISSN: 0018-9464; 1941-0069
• DOI: 10.1109/TMAG.2007.891420

The algebraic multigrid (AMG) method is an efficient preconditioner for large-scale finite element analyses. However, in edge-element analyses, the large null space of the coefficient matrices causes difficulties in applying AMG techniques. Here, we present a new AMG preconditioning for edge-element analyses. Numerical results demonstrate a promising performance for our proposed preconditioner