Load Dependent Hysteresis Model for GMM and its Parameter Identification

Abstract

This paper focuses on the development of load-dependent hysteresis model for Giant magnetostrictive materials (GMM). GMM are a class of smart materials and which are used mostly as actuators for active vibration control. Magnetostrictive actuators can deliver high ouput forces and relatively high displacements. Here, Terfenol-D, a magnetostrictive material is studied. Unlike the hysteresis seen in magnetic materials, The shape of Terfenol-D hysteresis curve changes significantly if the load is changed. To meet performance requirements for active vibration control, an accurate hysteresis model is needed. By modeling the Gibbs energy for each dipole and the equilibrium states, load-dependent hysteresis of GMM is modeled. Then a new PSO-LSM algorithm is brought forward by combing the Particle Swarm Optimization (PSO) with the least square method (LSM).Throughout this algorithm the model parameters were identified. The model results and experimental data were compared at different loads. The simulation results show that the load-dependent hysteresis model optimized by PSO-LSM yields outstanding performance and perfect accuracy.