Effective Material Transformation for Ferromagnetic Sheets

01.11.2022–31.10.2025
FWF Stand-Alone project
Principal investigator: Dr. Karl HOLLAUS (E101-03)

The demands on the energy efficiency of electrical devices are increasing, not least due to increasing electromobility. When designing electrical devices, efforts are therefore made to keep losses as low as possible. A large proportion of the losses are caused by eddy currents in the iron cores of the devices. To keep these losses as low as possible, the iron cores are composed of very thin ferromagnetic sheets. A precise and efficient calculation of the eddy currents in the laminated iron cores for an optimal design of electrical devices is therefore of enormous practical importance.

For the numerical calculation of problems with complex geometry and non-linear materials (hysteresis), the finite element method is preferably used. The overall dimensions of an iron core are in the range of meters, while the thickness of the sheets is less than one millimeter. A detailed finite element model of large devices therefore leads to extremely large systems of equations that cannot be solved with reasonable computational effort.

The aim of this project is to drastically reduce the computing time and memory requirements in the finite element simulations to support the efficient design of the devices in the best possible way. Therefore, homogenization methods with effective materials are developed, which require only a very coarse finite element mesh and very simple material models and thus lead to much smaller and efficiently solvable nonlinear equation systems. The results, opens an external URL in a new window already achieved in the FWF-funded research project Effective Material Transformation for Ferromagnetic Sheets, opens an external URL in a new window are very promising. Depending on the considered problem, finite element simulations using homogenization with effective materials can reduce memory requirements to up to a hundredth and computing time to up to a thousandth compared to detailed simulations.

Project members: Valentin HANSER and Markus SCHÖBINGER