A Deep Learning Framework for Damage Assessment of Composite Sandwich Structures

The vibrational behavior of composite structures has been demonstrated as a useful feature for identifying debonding damage. precision the damage localization can be greatly improved by addition more measuring points. Therefore, full-field vibration measurements, such those obtained using high-speed digital image correlation (DIC) techniques, are particularly useful. In this study, deep learning which have excellent performance in classification and segmentation, incorporated into novel approach assessing structures. This article presents damage-assessment algorithm sandwich that uses mode shapes learning. First, identified 3D DIC measurements. Then, Gaussian process regression is implemented to estimate shape curvatures, baseline-free gapped smoothing method applied compute images. indices, represented grayscale images, processed convolutional-neural-network-based automatically identify damaged regions. proposed methodology validated numerical experimental data from panel with different configurations.