Damage detection in composites by AI and high-order modelling surface-strain-displacement analysis

In the recent years, machine learning algorithms have been widely employed for structural health monitoring applications. As an example, Artificial Neu-ral Networks (ANN) could be useful in giving a precise and complete map-ping of damage distribution in a structure, including low-intensity or local-ized defects, which could be difficult to detected via traditional testing tech-niques. In this domain, Convolutional Neural Network (CNN) are employed in this work along with one-dimensional refined models based on the Carrera Unified formulation (CUF) for surface strain\displacement based damage detection in composite laminates. A layer-wise kinematic is adopted, while both an isotropic and orthotropic damage formulation is implemented. In de-tail, CUF-based finite element models have been exploited in combination with Monte Carlo simulations for the creation of a dataset of damage scenar-ios used for the training of the CNN. Therefore, the latter is fed with images of the strain or displacement field in a region of particular interest for each sample, which are subjected to the same boundary conditions. The trained ANN, given the strain\displacement mapping of an unknown structure, is therefore able to detect and classify all the damages within the structure, solving the so-called inverse problem.