Fast Preisach-based magnetization model and fast inverse hysteresismodel

Fast computational methods for Preisach-based models and their inverses are presented. The methods are based on a differential equation approach to computing a sequence of magnetization values due to a sequence of applied fields. The method used to speed up calculations can be applied to any Preisach model. The Della Torre, Oti, Kadar model was used here as an illustration. Sequential computations for the magnetization model are substantially faster than for standard Preisach models with comparable output error. Computations for the inverse hysteresis model are even faster than for the model. Using the inverse, open loop control of magnetic hysteresis is simulated, showing hysteretic material tracking a desired magnetization in a linear manner for both major loops and minor loops, with less than 2% error in inversion for the waveforms used. The effect of misidentification of material parameters on operation of the inverse is also investigated through simulations