On Eddy Current Brakes with Anisotropic Material Structures: Electromagnetic Model and Its Detailed Validation

Electrically excited eddy current brakes (ECBs) with solid active material suffer from low power density due to non-uniform heat dissipation caused by the skin effect. This paper presents an anisotropic material structure consisting of flux-conducting steel pins surrounded by electrically conductive perforated sheets, which simultaneously reduces the skin effect and enables direct liquid cooling of the active material. Two structural variants are investigated: a PIN structure with cylindrical flux-conducting elements and a cluster structure with elements in a hexagonal cluster configuration. A computationally efficient 2D reluctance network model is developed for both variants, combining analytically derived reluctances with FEM pre-computed reluctances for air spaces and frequency-dependent reluctances of the flux-conducting elements. The model is validated against measured torque characteristics of two experimental demonstrators over the full range of excitation currents from 10 to 80 A. Good agreement between simulation and measurement is achieved after correcting for two systematic deviations identified during validation: a geometric effect at the inner and outer edges of the perforated sheets that reduces the effective eddy current path resistance, and the temperature dependence of the sheet resistance. The critical rotational speed is accurately reproduced by the corrected model. The uncertainty range due to undetermined thermal operating conditions is quantified, providing a reliable basis for the design and optimization of ECBs with anisotropic material structure. Quantitative error analysis shows that the corrections reduce the root mean square error for the cluster structure by up to 77%, yielding a maximum error below 10% across the 50–100% excitation-current range. For the PIN structure, in contrast, a residual deviation of up to 15% remains at maximum excitation current, attributed to uncertainty in the magnetization curve of the flux-conducting elements. The model requires fewer than 100 degrees of freedom and computes a full torque–speed characteristic in under 5 min, four to five orders of magnitude faster than a full 3D transient FEM simulation. The uncertainty range due to undetermined thermal operating conditions is quantified, providing a reliable basis for the design and optimization of ECBs with anisotropic material structure.