A neural network approach for determination of Preisach model parameters under a sinusoidal induction at various frequencies

In this paper, we present a neural network-based approach, which allows us to predict the hysteretic loop, whatever the value of the frequency and flux density. The approach makes use of the Preisach-hysteretic model (PM) which provides a mathematical model to the B(H)B(H) curve, while the neural enables us to identify and predict the behaviour of parameters that the PM needs.