Damage identification in composite laminates using a hybrid method with wavelet transform and finite element model updating

Abstract

Delamination is a common defect in composite plates that may cause significant losses in commercial applications. In this study, a hybrid damage detection method was introduced, which raised both the damage identification sensitivity level and detected quantitative damage parameters. The primary damage location was pinpointed by the wavelet transform, and the damage parameters, including location, depth, and intensity, were then isolated by the model updating process. The wavelet transform was obtained according to the signal’s nature, which leads to improvement in the wavelet transform’s operation. Then, the lifting scheme algorithm process was performed to increase the wavelet efficiency in damage detection. In addition, a proper signal, based on strain energy, was used for damage detection by the wavelet transform. Finally, the genetic algorithm method was employed in the proposed model updating method to identify the damage parameters through employing a novel error function. The selected error function was based on strain energy, having the best operation among previously identified modal criteria. Consequently, the accuracy of identifying the damage parameters was improved upon utilizing the proposed method, particularly in the presence of noise. In addition, the solution performed faster than the previously available updating methods which utilized only the genetic algorithm-based on mode shapes and natural frequencies for detecting the damage.