After a disaster such as an earthquake has struck, the damage assessment of the affected buildings, bridges and other forms of structures is often urgently required for follow-up action. Research in using system identification for damage assessment in a quantifiable and non-destructive way has rapidly increased in recent years, due to advances in computing power and sensing technology. Though considerable progress has been made, many challenges still remain in achieving robust and effective identification of large structural systems using incomplete and noisy measurement signals. In this paper a novel strategy to tackle this problem is presented. A modified genetic algorithm (GA) strategy incorporating a search space reduction method, progressively and adaptively reduces the search space for each unknown parameter. By concurrent evolution of multiple species, it provides an excellent balance between exploration of the search space and exploitation of good solutions. The modified GA is incorporated into a damage detection strategy that works by comparing identified parameters for the undamaged and damaged structures and quantifies damage as a relative change in the stiffness of a member or a group of members. The additional information obtained from the analysis of the undamaged structure is used to greatly improve speed and accuracy in the identification of the damaged structure. Numerical studies on 10 and 20 degree-of-freedom (DOF) systems and an experimental study of a 7-storey small-scale steel frame are presented to illustrate the applicability of the method in accurately identifying even small amounts of damage.
[1]
Chan Ghee Koh,et al.
Modified genetic algorithm strategy for structural identification
,
2006
.
[2]
R. Ghanem,et al.
Structural-System Identification. I: Theory
,
1995
.
[3]
Chan Ghee Koh,et al.
A hybrid computational strategy for identification of structural parameters
,
2003
.
[4]
Ricardo Perera,et al.
Structural Damage Detection via Modal Data with Genetic Algorithms
,
2006
.
[5]
Surya B. Yadav,et al.
The Development and Evaluation of an Improved Genetic Algorithm Based on Migration and Artificial Selection
,
1994,
IEEE Trans. Syst. Man Cybern. Syst..
[6]
G-C Luh,et al.
Non-linear system identification using genetic algorithms
,
1999
.
[7]
Hong Hao,et al.
Vibration-based Damage Detection of Structures by Genetic Algorithm
,
2002
.
[8]
John H. Holland,et al.
Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence
,
1992
.
[9]
David E. Goldberg,et al.
Genetic Algorithms in Search Optimization and Machine Learning
,
1988
.