Dynamic Analysis-Based Damage Identification in Laminate Composites via Wavelet Transform and Group Method of Data Handling Network

Understanding the extent of structural damage is critical for effective decision-making in structural health monitoring. While wavelet transforms are powerful tools for detecting damage, they lack the ability to assess damage severity. To address this limitation, this study integrates the Group Method of Data Handling (GMDH) to enhance the accuracy of damage identification in laminated composite plates. A finite element model is developed to simulate damaged laminated composite plates and generate 2D signals, which are then processed using wavelet transforms. The GMDH algorithm further quantifies the damage severity at the locations identified by the wavelet transform. To validate the effectiveness of the proposed Wavelet-based GMDH approach (WT-GMDH), multiple damage scenarios are analyzed. The novelty lies in the integration of GMDH with wavelet transforms for damage quantification in laminated composite plates. The results demonstrate that, while wavelet transforms alone struggle to detect low-severity damage in identifying such damage (the WT-GMDH method achieves 98.66% and 98.53% accuracy for train and test stages, respectively). These findings confirm that the integration of the GMDH algorithm significantly enhances the capabilities of wavelet transforms, providing a more robust and efficient solution for structural damage assessment. Also, according to our results, the weakness of wavelet transform in damage detection in boundaries remains controversial.