Enhanced electromagnetic induction thermography detection of internal damage in CFRP-steel adhesively bonded structures

Carbon fiber reinforced polymer (CFRP) are widely used in steel structure reinforcement through adhesively bonding, making it crucial to inspect CFRP-steel adhesively bonded structures (ABCSS) to ensure their structural integrity and safety. However, the distinct physical properties of CFRP, epoxy resin, and steel pose challenges in accurately detecting internal damages in such specialized hybrid structures. To address this issue, this study proposes an enhanced electromagnetic induction thermography detection method to enhance the detection of internal damages in ABCSS. This method initially utilizes a conventional electromagnetic induction thermography system to obtain surface temperature data of the object under inspection, followed by preprocessing of the surface temperature data. Subsequently, a designed convolutional autoencoder (CAE) model is employed to extract pixel-level deep thermal features from the preprocessed surface temperature data. Finally, the extracted deep thermal features are utilized to generate enhanced detection results, thereby improving the visibility of damages. Experimental results on ABCSS specimens containing delamination, debonding, and cracks demonstrate that enhanced electromagnetic in-duction thermography effectively enhances the visibility of internal damages. This enhancement contributes to accurately assessing the quality of ABCSS, thereby improving the safety of such structures.