Hysteresis Identification Using Extended Preisach Neural Network

Hysteresis phenomena have been observed in different branches of physics and engineering sciences. Therefore several models have been proposed for hysteresis simulation in different fields; however, almost neither can be utilized universally. This paper introduces a universal adaptive model by inspiring Preisach Neural Network, called the Extended Preisach Neural Network Model (EPNN). It enjoys two hidden layers. The first hidden layer incorporates Deteriorating Stop (DS) neurons, which their activation function follows the (DS) operator. (DS) operator can generate noncongruent hysteresis loops. The second hidden layer includes Sigmoidal neurons. Adding the second hidden layer helps the neural network learn non-Masing and asymmetric hysteresis loops very smoothly. At the input layer, besides x(t), which is input data, x˙(t)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\dot{x}}(t)$$\end{document}, the rate at which x(t) changes, is included as well in order to give (EPNN) the capability of learning rate-dependent hysteresis loops. Hence, the proposed approach has the capability of simulation of both rate-independent and rate-dependent hysteresis with either congruent or noncongruent loops and symmetric and asymmetric loops. A new hybridized algorithm has been adopted for training the (EPNN), which is based on a combination of GA and the optimization method of sub-gradient with space dilatation. The generality of the proposed model has been evaluated by applying it to various hystereses from different areas of engineering with different characteristics. The results show that the model is successful in the identification of the considered hystereses. Furthermore, the proposed neural network shows excellent agreement with experimental data.