The usually employed macroscopic models for magnetic hysteresis are the Jiles and the Preisach models. When incorporated in computer-aided design tools (CAD) for electromagnetic analysis, these models require numerical approximation methods that impose a certain computational burden in the hysteresis calculations. Many analysis applications are very sensitive to this burden, particularly the combination of magnetic hysteresis with finite element method. Further, to analyze real devices using these CAD tools, it is necessary to first adjust the model to experimental data. For these models, the available parameter determination procedures require many steps to correctly fit the model to experimental data. This paper introduces a simple algebraic model to describe magnetic hysteresis. With only four parameters, it has low computational burden and reduced mathematical complexity, permitting thus a fast numerical implementation and simple parameter estimation procedure. The proposed model is also presented in its differential form, and a comparison of its mathematical structure with those of the Jiles and the Preisach models is described. Simulation results are presented and the model performance is discussed in terms of its capacity to represent common nonlinearities associated with the magnetic hysteresis phenomenon. A MATLAB script for numerical implementation of the model, describing the magnetic hysteresis in a MnZn power ferrite is presented.
magnetic hysteresis; computational magnetics; finite elements; magnetic losses; minor loop; major loop; accommodation; Madelung rules; magnetization; rate-independent model
