Selection of Optimal Parameters for the Jiles–Atherton Magnetic Hysteresis Model

Abstract

A method is proposed for implementation of the most popular hysteresis model, the Jiles–Atherton model, which has a number of advantages over other models. A technique for optimization of the parameters of the hysteresis model based on a real coded genetic algorithm is presented. The method is implemented in two stages. The first stage involves preliminary estimation of the model parameters and the range of their variation. The second stage is the direct implementation of the genetic algorithm. The criterion of convergence is based on the achievement of a preset value of the standard deviation and the maximum permissible number of generations. The genetic algorithm was implemented with 50 individuals. Each individual is associated with four variables that correspond to the hysteresis model parameters. The maximum number of generations was set to 50 and 100. The initial probabilities of the crossover and mutations were set to 90 and 5%, respectively. A specific feature of the proposed implementation of the genetic algorithm consists in internal optimization of the fifth parameter for each individual of the population. The computer code was developed using the Delphi environment. Comparison of the experimental and simulated curves showed good agreement. A method that involves preliminary estimation of the parameters and further application of the genetic algorithm yields rather accurate results, is easy to implement, and provides a high data-processing speed.