14 February 2022 in IEEE Transactions on Applied Superconductivity
For the difficulty of accurate simulation of dynamic hysteresis loop at high frequency, a simple and effective method is proposed. Firstly, combined with accuracy and generality, the classical inverse Jile-Atherton (J-A) model is selected to simulate the hysteresis characteristics. The parameters of the J-A model are identified by the velocity controlled particle swarm optimization (VCPSO); Secondly, combined with the characteristics of high-frequency magnetic field distribution of silicon steel, an improvement of the parametric magneto dynamics is proposed to accurately calculate the magnetic field corresponding to eddy current loss. Finally, according to the theory of Bertotti model, the magnetic field corresponding to excess loss is introduced to ensure the accuracy of total dynamic magnetic field strength simulation. The effectiveness of the method is verified by a comparison with the measured hysteresis cycle. In addition, under the coupling of improved parametric magneto dynamics (IPMD) model and J-A model, the high-frequency power loss can be accurately predicted.
1 March 2022 in AIP Advances
To accurately predict the eddy current loss of silicon steel laminations over broadband frequency range, an improved parametric magneto dynamic model (IPMD) is proposed in this paper. According to the nonuniform distribution of magnetic field in the cross-section of the silicon steel lamination under high-frequency magnetization, the analysis area of the lamination is divided into several slices by nonlinear segmentation, wherein each slice the magnetic field distribution is regarded as uniform. Meanwhile, to accelerate the calculation speed of the IPMD model, the relationship between skin depth and slice number is deduced and verified. Furthermore, the Tellinen (TLN) hysteresis model is used when implementing the IPMD model to further increase the accuracy of eddy current loss calculation.