Damage identification in civil engineering structures utilizing PCA-compressed residual frequency response functions and neural network ensembles

摘要

This paper presents a non-destructive, global, vibration-based damage identification method that utilizesndamage pattern changes in frequency response functions (FRFs) and artificial neural networks (ANNs) tonidentify defects. To extract damage features and to obtain suitable input parameters for ANNs, principalncomponent analysis (PCA) techniques are applied. Residual FRFs, which are the differences in the FRFndata from the intact and the damaged structure, are compressed to a few principal components and fed tonANNs to estimate the locations and severities of structural damage. A hierarchy of neural networknensembles is created to take advantage of individual information from sensor signals. To simulate field-ntesting conditions, white Gaussian noise is added to the numerical data and a noise sensitivity study isnconducted to investigate the robustness of the developed damage detection technique to noise. Bothnnumerical and experimental results of simply supported steel beam structures have been used to demonstrateneffectiveness and reliability of the proposed method. Copyright r 2009 John Wiley & Sons, Ltd.