Structural Damage Identification Using A Global Optimization Technique

This paper presents a two-stage, vibration-based structural damage detection, localization, and severity estimation using an adaptive simulated annealing (ASA) global optimization technique. First, a "damage-sensitive" response parameter that has a strong physical relationship with structural physical properties is presented for detection and localization of structural damage using a nonmodel-based damage identification approach. Secondly, an ASA optimization algorithm is employed to estimate structural damage severity via minimization of a cost-function expressed in terms of the scalar distance between the "damage-sensitive" response parameter determined from a potentially damaged structure and that computed from a finite element model of the undamaged structure. The significance of this study is that the method can be used for damage detection, localization, and estimation of damage severity in two stages and can be applied to input-output as well as output-only damage identification problems. Finally, the proposed technique is demonstrated on simulated data obtained from a simply supported reinforced concrete beam and experimental modal data obtained from the I-40 Bridge.