Hysteresis nonlinearity identification by using RBF neural network approach

Abstract

In systems with hysteresis behavior like magnetic cores, Piezo actuators, Shape Memory Alloy(SMA), we essentially need an accurate modeling of hysteresis either for design or performance evaluation; also in some control applications accurate system identification is needed. One of the famous methods of Hysteresis modeling is Preisach model. In this numerical method hysteresis is modeled by linear combination of smaller hysteresis loops as an elemental operator and local memory. In this paper we discuss those Radial Base artificial neural networks (RBF) which provides natural settings in accordance with the Preisach model. It is shown that the proposed approach can represent hysteresis modeling accurately in compare with classical Preisach model and can be used for many applications such as hysteresis nonlinearity control, hysteresis identification and realization for performance evaluation and system design. For evaluation we use measured experimental data from hysteresis SMA wire as an actuator.

Dynamic Ferromagnetic Hysteresis Modelling Using a Preisach-Recurrent Neural Network Model

“…The function can then be applied in software using a look-up table of values. Other 'black-box' identification techniques used alongside the Preisach model include genetic algorithms [10], fuzzy models [11] and artificial neural networks (ANN) [12][13][14]. In [15], Saliah et al…”