[Back]

Y. Kanto, G. Shilyashki, H. Pfützner, I. Matkovic:
"Numerical and Experimental Determination of Local Building Factors of a Three-Phase Transformer Core Package";
IEEE Transactions on Magnetics, Vol. 55 (2019), No. 2; 11.

The analyses of flux distributions and losses in transformer cores need highest expenditure if performed in experimental ways. This paper presents a novel combined methodology for the evaluation of local and global loss building factors BF. As a first step, the local patterns of the dynamic induction vector B(t) were calculated, by means of a novel numerical "magnetic anisotropic circuit calculation (MACC) methodology. The latter considers the core materials non-linear permeability functions in rolling direction and transverse direction, as well as non-linear effects of overlaps in corners and T-joints. In a second step, the calculated induction patterns were exactly simulated on the hexagonal Vienna rotational single sheet tester and the corresponding local losses p were measured electrodynamically. Finally, the values p were allocated to the corresponding core positions, taking into account the entire core geometry. BF of the core was estimated from the total of loss values weighted to the corresponding volume portions. The method was applied for a three-phase core of highly grain-oriented material, magnetized with BNOM = 1.7 T. The procedure shows the local distributions of losses, yielding also information on those of eddy currents and rotational magnetization (RM). As an
example, the T-joint, as the area of highest RM, yields a local loss increase of about 50%. Apart from RM, the local induction
variations-including waveform distortions, caused mainly by circulating magnetization-prove to have a high impact on losses,
locally yielding 50% as well. These highly convincing results indicate that the complicated sensor studies of cores can partly be
replaced by: 1) numerical modeling of induction distributions and 2) subsequent allocations of loss values as detected by RSST
simulations.

Building factors, losses, numerical modelling, rotational magnetization, transformer cores

http://dx.doi.org/10.1109/TMAG.2018.2882765