Multiobjective lightining search applied to Jiles-Atherton hysteresis model parameter estimation

Hysteresis is a fundamental property commonly encountered in physical systems of a wide variety of engineering and science fields and the parameters determination and/or optimization of hysteresis dynamical models is an essential approach for adequate hysteretic material simulations. In magnetic vector hysteresis models as the Jiles-Atherton (J-A) the work increases in complexity since one must solve a nonlinear system with a relative large number of design variables. In this context, fitting methods based on efficient stochastic optimization metaheuristics is an attractive solution to solv problems related to the phenomena of nonlinear hysteresis applications. In this study, an improved multiobjective lightning search algorithm (IMLSA), a stochastic optimization metaheuristic algorithm, is introduced for solving J-A hysteresis model parameter estimation. The proposed IMLSA based on a mutation operator of the differential evolution is validated using measured hysteresis data from a rotational single sheet tester. In addition, thorough examination of comparison results of IMLSA with optimization results using a multiobjective lightning search showed that the performance of the IMLSA is promising in parameters estimation of nonlinear hysteretic J-A models.