Damage identification of structures is an important task of a health monitoring system. The ability to detect damages on-line or almost on-line will ensure the reliability and safety of structures. Analysis methodologies for structural damage identification based on measured vibration data have received considerable attention, including the least-square estimation (LSE), extended Kalman filter (EKF), etc. Recently, new analysis methods, referred to as the sequential non-linear least-square estimation (SNLSE) and quadratic sum-squares error (QSSE), have been proposed for the damage tracking of structures. In this paper, these newly proposed analysis methods will be compared with the LSE and EKF approaches, in terms of accuracy, convergence and efficiency, for damage identification of structures based on experimental data. A series of experimental tests using a small-scale 3-story building model have been conducted. In these experimental tests, white noise excitations were applied to the model, and different damage scenarios were simulated and tested. Here, the capability of the adaptive LSE, EKF, SNLSE and QSSE approaches in tracking the structural damage are demonstrated using experimental data. The tracking results for the stiffness of all stories, based on each approach, are compared with the stiffness predicted by the finite-element method. The advantages and drawbacks for each damage tracking approach will be evaluated in terms of the accuracy, efficiency and practicality.

Date Published: 8 April 2008
PDF: 11 pages
Proc. SPIE 6932, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2008, 693225 (8 April 2008); doi: 10.1117/12.774621

Published in SPIE Proceedings Vol. 6932:
Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2008
Masayoshi Tomizuka, Editor(s)