Genetic algorithm-based approach for Bayesian damage identification using spectral density analysis in beam-like structures

This paper, which is a part of an ongoing research study, describes a Genetic Algorithm-based approach towards a global damage identification framework for the continuous/periodic monitoring of civil structures. In order to localise and estimate the severity of damage regions, a one stage model-based Bayesian probabilistic damage detection approach is proposed. The method is based on response power spectral density of the structure which enjoys the advantage of broadband frequency information and can be used for input-output as well as output-only damage identification studies. The suitability of the proposed method is investigated for an error-free numerical model with noise-free response data sets in a beam-like structure with different severity of damage. The results obtained indicate that the Genetic Algorithm-based damage identification approach is suitable for damage detection and can be considered for further implementation using more realistic noisy response data that would be associated with the monitoring of real civil structures.