A wavelet-based damage diagnosis algorithm using principal component analysis

The applicability of the Haar and Morlet wavelet transforms of the vibration signals for structural damage diagnosis was demonstrated in previous papers by the authors. Two significant developments that followed the initial research are presented in the current paper. In the first part of the paper, a data preprocessing algorithm is developed to enable optimal signal selection from a database of baseline signals for damage diagnosis purposes. The second part of the paper describes the extraction of the damage sensitive feature vector as a function of the energies at the fifth, sixth, and seventh dyadic scales of the vibration signal that serve for damage detection. Both data preprocessing and feature extraction models involve the use of principal components analysis. The process of damage detection is automated using the k-means algorithm and the gap statistic. A simple damage extent measure is also discussed. Finally, the migration of damage sensitive feature vectors with increased damage is illustrated for vibration signals obtained from the American Society of Civil Engineers' benchmark structure numerical simulations. The results indicate that the proposed algorithm is able to consistently detect and quantify damage for the damage patterns specified by the American Society of Civil Engineers' benchmark experiment. Copyright (c) 2011 John Wiley & Sons, Ltd.