Embedding microwires into composite materials is a novel technology that can provide in-situ and remote data on the structural health of the system. Damages such as delamination, disbond, and cracks in the composite structure induce a change in the stresses and strains on the microwires, modifying their electromagnetic response. Analysing this response with artificial intelligence can potentially allow us to not only detect these damages, but also characterise, localise, and quantify the damage, and possibly predict the remaining useful life of the structure. As part of the Horizon Europe project INFINITE, this work aims to develop methodologies for in-service structural health monitoring of composites equipped with microwire sensors. The suitable machine learning algorithms include physics-aware three-dimensional convolutional neural networks, decision trees, logistical regression, etc., all of which require a large amount of training data, as accurate estimation on training including data instances of all type and levels. Since experimental data is scarce due to the low readiness-level of the technology, these algorithms are trained using finite-element numerical computation, which itself has a high computational cost, while fine-tuning is left to experimental data. Currently, maintenance, repair, and overhaul activities represent around 10–15% of an airline’s operational costs, which in 2021 meant a global average of USD 234 million per airline. In-situ monitoring of structural health of composite aircraft components will enable the aerospace industry to move from schedule-based maintenance to condition-based maintenance, especially for inaccessible components, by simplifying diagnostics and efficient logistics of repair operations. This work was funded by the European Union under the Horizon Europe grant 101056884.