Damage assessment of earthquake dampers based on Preisach-Mayergoyz model

Abstract

Preisach-Mayergoyz (PM) model of hysteresis can be used to evaluate the mechanical or ferromagnetic properties of different categories of hysteretic materials. The statistical PM space can be expressed as weighted combination of hysterons. A computer programme was designed for automatic numerical identification of the PM space probability density of hysterons within a class of distribution mixtures using optimization stochastic algorithms and $\phi $-divergence measures. It was found that the FSA optimization technique (Fast Simulated Annealing), in combination with the Hellinger divergence measure between hysteretic loops, reliably founds the precise PM spaces of hysterons distributed along a mixture of two Guyer and parametrized Koen distributions. As an example, experimentally measured data of earthquake dampers are processed, the six identified PM spaces of six testing cycles for the dampers are presented and, based on them, new damage index $ID_{PM}$ is proposed and evaluated.