Development of shape sensing and damage monitoring algorithms for digital twins of aeronautical structures

Growing demands for lighter, safer, and more efficient aeronautical structures are driving research toward condition-based and digital twin–oriented structural management, rather than relying on scheduled inspections. In this context, the ability to estimate the structural state in real time, in terms of both performance and integrity, represents a fundamental requirement. In this thesis, a Structural Health Monitoring framework for shape sensing and damage monitoring of aeronautical structures is developed, combining physics-based and data-driven approaches using a single network of Fiber Bragg Grating sensors. In the first part, shape sensing methodologies based on the inverse Finite Element Method are developed. Since such a method requires a high number of sensors to accurately reconstruct the full displacement field, different strategies are proposed to preserve reconstruction accuracy while reducing the required sensor count, thus enabling practical applicability in real industrial scenarios. The proposed frameworks are validated through both numerical simulations and experimental testing on typical aeronautical structures. In the second part, methods for structural damage monitoring are investigated, with a particular focus on detection and localization. The adopted approaches are vibration-based and rely on variations in strain mode shapes between baseline and current structural configurations. Fully automated methodologies are developed, both through the use of damage indices and neural networks. These frameworks are validated using numerical case studies and experimental data from a composite scale glider. Overall, the work carried out in this thesis represents a relevant advancement toward the development of structural digital twins, supporting the transition toward next-generation aircraft.