A Loss Separation-Based Dynamic Jiles–Atherton–Bingham Model for Magnetorheological Dampers

TL;DR

This paper introduces a new model for magnetorheological dampers that better captures their complex behavior under dynamic conditions.

Contribution

The novel LS-DJAM model integrates multiple loss mechanisms and uses a hybrid PSO–GA optimization for improved accuracy.

Findings

The LS-DJAM model reduces maximum modeling error by 87.5% compared to conventional models.

Hybrid PSO–GA optimization improves parameter estimation accuracy by over 60%.

Abstract

Magnetorheological (MR) dampers exhibit significant hysteretic nonlinearities, particularly under dynamic operating conditions, where accurately modeling the complex coupling between magnetic flux density and excitation current remains challenging. To overcome the limitations of the conventional static Jiles–Atherton (JA) model in capturing dynamic hysteresis responses, a dynamic JA model incorporating multiple loss mechanisms (LS-DJAM) is proposed for MR dampers. Building on loss separation theory, the model integrates eddy current and excess loss mechanisms to more accurately represent the dynamic hysteresis behavior of MR dampers. By coupling the Bingham mechanical model, a magneto-mechanical constitutive relation for MR dampers is established. Furthermore, to enhance the accuracy of LS-DJAM parameter identification, a hybrid particle swarm optimization–genetic algorithm (PSO–GA) is developed. Genetic operators are embedded within the PSO framework to strengthen the global search capability and mitigate premature convergence, thereby enabling efficient LS-DJAM parameter identification. The proposed LS-DJAM, identified via the PSO–GA, significantly enhances the modeling of MR damper output forces. PSO–GA parameter estimation improves accuracy by over 60%, and the LS-DJAM reduces the maximum modeling error by 87.5% compared with the conventional JA model. It accurately captures the dynamic hysteresis characteristics of MR dampers, providing a robust theoretical basis and practical framework for high-performance control and engineering optimization.

Figures