This conference paper was submitted for presentation at the NAFEMS World Congress 2025, held in Salzburg, Austria from May 19–22, 2025.

Abstract

The numerical simulation of electromagnetic field problems often requires to incorporate the unbounded volume that surrounds an object of interest. This exterior region cannot be fully captured by a domain-based discretization scheme like the Finite Element Method (FEM) and requires a truncation of the domain in combination with complicated boundary conditions. However, the surface-based Boundary Element Method (BEM) exactly represents such exterior regions without the need of a volume mesh, but can only handle a linear and homogeneous material behavior. The coupling of FEM and BEM thus leverages the advantages of FEM for electromagnetic simulations and BEM for modelling exterior domains, resulting in a comprehensive solution framework that is perfectly suited for the simulation of multi-physics phenomena. This coupling approach is especially useful when analyzing situations in which conducting domains and/or magnets move with respect to each other and, possibly, deform due their mutual electromagnetic forces. Now, the mechanical behavior of such objects is calculated with a FEM model whose mesh is then used within the FEM-BEM coupling to compute the forces. The change of constellation due to the mechanical deformation does not require a re-meshing of the computational domain. Prominent examples are electromagnetic metal-forming, sensors and actuators consisting of electro-magnetostrictive materials, and magnetohydrodynamics. We are currently developing a flexible, and yet easy-to-use framework to address such kinds of multi-physical problems. This is achieved by using an open-source coupling library for multi-physics simulations which offers a well-defined API for exchanging data between a wide range of different numerical solvers. In this work we present an adapter for our FEM-BEM coupling code. Through numerical experiments and validation against measurements and analytical solutions, the efficiency and accuracy of the proposed approach is demonstrated across a range of scenarios like simple eddy current brakes or the TEAM benchmark problem 28. This work contributes to advancing the understanding and simulation capabilities of multi-physics phenomena involving electromagnetic interactions, paving the way for enhanced design and optimization of electromechanical systems in diverse engineering applications.