This paper aims to propose an effective modeling method of dynamic hysteresis properties for soft magnetic composite (SMC) core using an equivalent circuit representation. Because the eddy currents flowing inside iron powder particles should be considered, it is well known that an accurate magnetic field analysis of the SMC core in a wide range of excitation frequency is not easy. To overcome this difficulty, a dynamic hysteresis modeling based on the standard Cauer circuit is investigated.
In the proposed method, the first inductance represents the static magnetic property of the SMC, and the latter part represents the dynamic effect because of the eddy currents. The values of the circuit elements were determined by an optimization method based on symmetric loops measured at several frequencies. To verify the validity of the proposed modeling method, finite-element analyses of a ring core inductor and an alternating current reactor were performed.
By comparing the simulated and measured magnetic properties, the necessity to consider magnetic hysteresis in the equivalent circuit model is clarified. Furthermore, the frequency-dependent inductances of practical reactors can be obtained from the finite-element analysis combined with the proposed method.
This paper demonstrates the significance of determining the circuit parameters in the equivalent circuit for dynamic hysteresis modeling based on the measured magnetic properties. The effectiveness of the proposed method is verified by comparing frequency-dependent inductances of two kinds of reactors between the simulation and measurement.
Watanabe, N., Takahashi, Y. and Fujiwara, K. (2019), "Dynamic hysteresis modeling for soft magnetic composites based on isotropic vector play model and Cauer ladder network", COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, Vol. 38 No. 5, pp. 1413-1426. https://doi.org/10.1108/COMPEL-12-2018-0521Download as .RIS
Emerald Publishing Limited
Copyright © 2019, Emerald Publishing Limited