Purpose
Electrical machines commonly consist of moving and stationary parts. The field simulation of such devices can be demanding if the underlying numerical scheme is solely based on a domain discretization, such as in the case of the finite element method (FEM). This paper aims to present a coupling scheme based on FEM together with boundary element methods (BEMs) that neither hinges on re-meshing techniques nor deals with a special treatment of sliding interfaces. While the numerics are certainly more involved, the reward is obvious: the modeling costs decrease and the application engineer is provided with an easy-to-use, versatile and accurate simulation tool.
Design/methodology/approach
The authors present the implementation of a FEM-BEM coupling scheme in which the unbounded air region is handled by the BEM, while only the solid parts are discretized by the FEM. The BEM is a convenient tool to tackle unbounded exterior domains, as it is based on the discretization of boundary integral equations (BIEs) that are defined only on the surface of the computational domain. Hence, no meshing is required for the air region. Further, the BIEs fulfill the decay and radiation conditions of the electromagnetic fields such that no additional modeling errors occur.
Findings
This work presents an implementation of a FEM-BEM coupling scheme for electromagnetic field simulations. The coupling eliminates problems that are inherent to a pure FEM approach. In detail, the benefits of the FEM-BEM scheme are: the decay conditions are fulfilled exactly, no meshing of parts of the exterior air region is necessary and, most importantly, the handling of moving parts is incorporated in an intriguingly simple manner. The FEM-BEM formulation in conjunction with a state-of-the-art preconditioner demonstrates its potency. The numerical tests not only reveal an accurate convergence behavior but also prove the algorithm to be suitable for industrial applications.
Originality/value
The presented FEM-BEM scheme is a mathematically sound and robust implementation of a theoretical work presented a decade ago. For the application within an industrial context, the original work has been extended by higher-order schemes, periodic boundary conditions and an efficient treatment of moving parts. While not intended to be used under all circumstances, it represents a powerful tool in case that high accuracies together with simple mesh-handling facilities are required.
[1]
P. Raviart,et al.
A mixed finite element method for 2-nd order elliptic problems
,
1977
.
[2]
R. Kress,et al.
Inverse Acoustic and Electromagnetic Scattering Theory
,
1992
.
[3]
L. Greengard,et al.
A new version of the Fast Multipole Method for the Laplace equation in three dimensions
,
1997,
Acta Numerica.
[4]
J. Nédélec.
Mixed finite elements in ℝ3
,
1980
.
[5]
D. A. Dunnett.
Classical Electrodynamics
,
2020,
Nature.
[6]
Stefan Kurz,et al.
Application of the adaptive cross approximation technique for the coupled BE-FE solution of symmetric electromagnetic problems
,
2003
.
[7]
M. Saunders,et al.
Solution of Sparse Indefinite Systems of Linear Equations
,
1975
.
[8]
Wolfgang L. Wendland,et al.
Some applications of a galerkin‐collocation method for boundary integral equations of the first kind
,
1984
.
[9]
R. Fox,et al.
Classical Electrodynamics, 3rd ed.
,
1999
.
[10]
Marc Duruflé,et al.
High-order optimal edge elements for pyramids, prisms and hexahedra
,
2013,
J. Comput. Phys..
[11]
Ralf Hiptmair,et al.
Symmetric Coupling for Eddy Current Problems
,
2002,
SIAM J. Numer. Anal..
[12]
Jinchao Xu,et al.
Nodal Auxiliary Space Preconditioning in H(curl) and H(div) Spaces
,
2007,
SIAM J. Numer. Anal..
[13]
Johannes Tausch,et al.
Numerical Exploitation of Equivariance
,
1998
.
[14]
R. Hiptmair.
Finite elements in computational electromagnetism
,
2002,
Acta Numerica.
[15]
Ralf Hiptmair,et al.
Generators of $H_1(\Gamma_{h}, \mathbbZ)$ for Triangulated Surfaces: Construction and Classification
,
2002,
SIAM J. Comput..
[16]
Ronald H. W. Hoppe,et al.
Finite element methods for Maxwell's equations
,
2005,
Math. Comput..
[17]
Jinchao Xu,et al.
Iterative Methods by Space Decomposition and Subspace Correction
,
1992,
SIAM Rev..