Gath-Geva approach for modeling ferromagnetic behavior

Abstract

On the basis of fuzzy logic identification and approximation capability of any kind of nonlinear, continuous functions represented by its discrete set of measured data, a new modeling technique for dynamic ferromagnetic hysteresis is presented using a Gath and Geva fuzzy clustering approach. The construction of dynamic behavior model is done in two steps. In the first step, the fuzzy membership functions in the rule antecedents are determined. In the second step, the parameters of the consequent functions are estimated by means of standard linear least squares methods. The partition coefficient and the classification entropy indices are used for validation of clusters number. Very accurate prediction of dynamic hysteresis loops is observed, proving that the clustering technique is suitable for hysteresis modeling.