A Computationally Compact Algorithm for Real-time Detection of Abrupt Structural Stiffness Degradations

Abstract

Various algorithms for structural system identification and damage detection have been developed. Many of these algorithms are off-line approaches while real-time structural identification is absolutely necessary for on-line structural damage detection and optimal structural vibration control during severe loading such as earthquakes. Frequently, structural damage can be reflected by the degradation of structural element stiffness. In this paper, a time domain algorithm with computational compact is proposed for real-time detection of the onset, locations and extents of abrupt degradations of structural element stiffness using partial measurements of structural acceleration responses. No prior knowledge of structural dynamic parameters before damage is required as these parameters are recursively estimated. Then, the time instant and possible locations of degraded structural elements are detected by tracking the errors between the measured data and the estimated values by Kalman filter. Finally, the exact locations and extents of structural element stiffness degradations are determined by solving simple constrained optimization problems. Both a numerical example and an experimental test are used to validate the proposed algorithm for real-time detection of abrupt degradations of structural element stiffness using only measurements of structural acceleration responses polluted by noises. doi: 10.12783/SHM2015/141