At the Board of Trustees meeting on 28 September 2026, the FWF decided to fund the individual project “Homogenisation methods for laminated open cores” (project number PAT3001326 https://www.fwf.ac.at/forschungsradar/10.55776/PAT3001326, opens an external URL in a new window) led by Dr Karl HOLLAUS with a grant of approximately €440,063.
The research project focuses on open-lapped iron cores, which pose a particular challenge for the numerical simulation of eddy currents. Open iron cores do not have a closed magnetic circuit and therefore exhibit very large magnetic stray fields and, consequently, eddy currents. Put simply, the pattern of these stray fields resembles that of the magnetic field lines visualised by iron filings outside a permanent magnet.
For optimum equipment design, it is therefore crucial in practice to be able to calculate these eddy currents accurately and efficiently. The finite element method (FEM) is the preferred approach for calculating eddy currents in complex geometries and non-linear materials. When designing large electrical equipment using FEM, extreme differences in scale present a significant challenge. Whilst the entire iron core has dimensions in the metre range, the sheet thicknesses and penetration depths are in the sub-millimetre range. A detailed FEM model of these fine structures would result in huge, non-linear systems of equations that cannot be solved with a reasonable computational effort. Efficient numerical calculation methods are therefore of enormous importance.
Instead of resolving the fine structure of the thin sheets in detail, the simulations are carried out using the new methods on a considerably coarser finite-element mesh. This results in significantly smaller systems of equations. To this end, advanced multi-scale finite element methods and hybrid methods based on effective materials are being developed. These new methods are capable of adequately accounting for magnetic stray fields and accurately determining eddy current losses with minimal computational effort compared to the standard FEM.
The new methods have a modular and hierarchical structure, meaning that the computational effort and the required accuracy of the simulations can be tailored to specific requirements. This provides optimal support for the design of devices in practice. Whilst the new hybrid methods enable very fast simulations, which are ideal for preliminary design, the advanced multi-scale finite element methods allow for highly accurate simulations of problems involving open iron cores.