TIME-DOMAIN SYSTEM IDENTIFICATION FOR STRUCTURAL HEALTH MONITORING OF BUILDINGS

Abstract

This study aims to develop a system identification methodology for structural health monitoring and damage detection based on vibration signature analysis. The approach uses a time-domain least-squares based method _ to identify the reduced-order system matrices of an equivalent linear model whose order matches the number of available sensors. A practical constraint of limited number of sensors is also taken under consideration. Having identified the reduced-order model, a matrix update method is presented here for determination of perturbations in the property matrix. Updating of the perturbed matrix is carried out by sequential quadratic programming by minimizing an error function. Damage location can be easily find out by obser-vation of elements of perturbed matrix. The efficiency of the proposed technique is shown by simulation study for ten story shear building subjected to earthquake excitation. It is shown that the proposed methodology gives reasonably accurate identification in terms of location of structural damage.