This study is part of the Project Democrite, an international cooperation between four research institutes: CIRA (Italy), Łukasiewicz – Institute of Aviation ( Poland), ONERA (France) and INCAS (Romania) under EREA coordination.

One of the target of the project is the analysis and investigation of SHM system for carbon fiber based composite cylinder shaped coupons, aimed at demonstrating an innovative hydrogen storage approach. Such elements, representative of outer tank, experience biaxial loading caused by internal pressure, in turn able to suffer multidirectional delaminations. The presence of these structural failure can play a remarkable impact both in terms of load-bearing capacity and leakage resistance. For this reason, an integrated monitoring system must be considered as a necessary tool for safety and reliability purposes. The experimental campaign presented in this paper is an intermediate stage of analysis focusing on the thermo-mechanical test of cylinders shaped coupons under ultimate load. The monitoring of the response at cryogenic temperature is entrusted by fiber optic and strain gauges sensors bonded on the surface. Data are then processed by an algorithm based on cross-correlation in time and space domain. The effects of temperature environment over the mechanical load is analyzed and possible faults are monitored to estimate the grade of sensitivity of the system. This work offers an empirical foundation for improving the mechanical performance and leakage resistance toward the final target of the composite cryogenic storage systems.

Additively manufactured (AM) polymer composites are increasingly entering structural applications, yet the failure mechanisms and the role of embedded sensing for health monitoring remain incompletely characterized. This work presents a sequential multiscale framework, linking micromechanics homogenization to mesoscale numerical analysis to study the dominant failure modes under uniaxial tension for AM polylactic acid carbon fiber reinforced polymer. Effective homogenized properties are obtained from a representative volume element (RVE) with periodically distributed fibers, reflecting realistic microstructures. These properties are used to parameterize the mesoscale ABAQUS model. Intralaminar failure is predicted with the Hashin and LaRC05 criteria, while interlaminar behavior is captured via cohesive zone elements. The cohesive formulation is further used to probe the influence of interfacial adhesion between the impregnated polymer surrounding the fiber filaments and the matrix on the onset and propagation of delamination. The finite element model results are validated against tensile tests of coupons embedded with distributed fiber optic sensors (DFOS) with two different embedding patterns and post-fracture Scanning Electron Microscopy (SEM). The simulations reproduce the observed failure sequences such as matrix cracking, fiber-matrix debonding, fiber kinking, etc. Both the Hashin and LArC05 criteria capture the intralaminar damage and the most aiding factors in matrix and fiber-dominated modes. DFOS strain data agree with extensometer measurements for both embedding layouts while revealing local damage initiation not resolved by point sensors. The framework clarifies the choice of failure criteria for AM composites, highlights the impact of interface quality on structural performance, and investigates utility sensor embedding for effective structural-health monitoring (SHM).

Transition towards sustainability in the current industry landscape is enforcing the development of strategies to reduce waste during production and define roadmaps for the end of life of products that consider recycling or repurposing for other applications. The composite materials industry (Fiber-reinforced plastics, FRPs) faces particular challenges because of the complexity for separation of materials and the toxicity of the materials involved. In addition, there is still a lack of standardization for waste management in composite materials, both during production and at the end of life. Composite materials in Wind Energy industry pose a more pressing challenge since there are already tonnes of materials that are reaching their end of life. In Aeronautics industry, the majority of waste is set to come in 20 to 30 years, when the first full composite aircrafts reach their end of life, but there is a more urgent situation to manage the composite waste generated everyday in factories. In particular, cutoff of prepreg tapes made of carbon fibre impregnated with uncured epoxy that are produced during cutting of the laminate preforms.

Most efforts up to now focused on strategies for recycling FRP composite material, but there are much fewer works on reusing uncured prepreg cutoffs. To this end, the present study proposes the use of prepreg cutoffs for manufacturing of added-value components for the factory like layup toolings or drilling templates. However, since the composite material is going to be exposed to high pressure and temperatures and the tooling requires high dimensional stability, there is a need to monitor the degradation of the material with time. Fibre Bragg Grating technology has been used for developing a strategy for in-situ monitoring deformation during service of these tooling.

A strategy for embedding fibre optics sensors within the composite laminate has been conceived based on integrating the sensor during the layup process, creating an exit channel through the layers. The embedment process presents several challenges: dealing with high temperature and pressure. Fibre optic sensor materials need to be suitable to manage service conditions of the autoclave consolidation process (up to 200 °C and 7 bar pressure). Factors like material degradation due to high temperature and physical damage, birefringence or signal loss due to high pressure have been taken into account for the selection of the fibre optics materials. As a result, this study has explored the integration of a FBG based fibre optic sensor, coated with a PEEK/PTFE protection tube, into the layup tooling made of reused composite material for real-time deformation monitoring. As a validation strategy, once embedded in the laminate and consolidated, the tooling has been subjected to several autoclave cycles to simulate real service conditions. Deformations have been monitored before and after each cycle in order to detect permanent deformation of the material coming from microcracking of the composite matrix. In addition, data analysis procedures have been implemented to account for the effect of autoclave pressure during laminate consolidation that lead to permanent deformation of the fibre optic sensor.

As the most widely used foundation type in offshore wind farms, monopile foundations are highly vulnerable to scour, which reduces soil support and threatens structural stability. Although various monitoring techniques have been developed, achieving continuous and remote detection of scour-induced structural changes remains challenging. This study proposes a smart pipe concept integrating distributed fiber-optic strain sensing with three-dimensional shape reconstruction for scour monitoring. To evaluate the feasibility of the proposed monitoring concept, numerical simulations were performed using finite element analysis. Scour is represented by modifying the boundary conditions of the pipe to reflect the reduction of soil support. Under identical loading conditions, the change in support conditions produced distinguishable deformation patterns in the pipe. The strain distributions obtained from the finite element model are then used as input for the proposed shape reconstruction algorithm. The reconstructed shapes showed strong agreement with the deformation obtained directly from the finite element results. These results demonstrate that the smart pipe system can effectively detect boundary condition changes associated with scour and has strong potential for continuous monitoring of monopile foundation stability.

Composite unmanned aerial vehicles (UAVs) are particularly susceptible to barely visible impact damage, and they lack an onboard pilot who can sense or report low-velocity impacts during flight. As a result, it is important for UAV platforms to autonomously detect impact events and estimate their locations in real time to ensure structural integrity, maintain mission readiness, and reduce unforeseen failures. In composite wing structures, low-energy impacts can induce internal delamination or matrix cracking without producing clear surface indications, which makes early detection even more essential for practical maintenance planning.

This study proposes an impact localization approach based on Fiber Bragg Grating (FBG) sensors, developed with a focus on practical applicability for real composite UAV wings. The system uses only commercially available FBG interrogators and standard optical fiber sensing components, eliminating the need for complex hardware modifications or specialized instrumentation. This design choice reduces integration effort and allows the sensing system to be applied to existing UAV platforms without significant changes to the aircraft structure.

To validate the proposed method, experiments were conducted using an actual composite wing section from a small UAV. A single optical fiber line containing four to six multiplexed FBG sensors was attached along selected regions of the wing skin to capture strain responses induced by impact events. Low-velocity impacts were introduced using an instrumented impact hammer, enabling controlled excitation and repeated tests under consistent conditions. The multiplexed FBG configuration allowed multiple sensing points to be implemented with minimal additional mass and simplified cabling, which is advantageous for weight-sensitive UAV systems.

The developed algorithm processed the strain signals in real time and demonstrated consistently high accuracy in identifying the locations of impact events across the tested area of the wing. Although specific numerical accuracy values are not presented here, the experimental results confirm that the sensing configuration and localization logic are effective for estimating impact positions on composite wing structures. The capability for real-time analysis, combined with the lightweight and straightforward sensor layout, suggests that the approach can support more efficient maintenance decisions by providing immediate diagnostic information following impact events.

Overall, the results indicate that an FBG-based sensing system can be applied to composite UAV wings without excessive system complexity, while still offering meaningful localization performance. This work highlights the potential for integrating lightweight optical fiber sensing into UAV structural health monitoring systems, contributing to improved maintenance efficiency, reduced inspection workload, and enhanced operational readiness during field operations.

Continuous traffic monitoring is critical for accurate structural load assessment and fatigue life estimation of highway bridges. However, conventional vision-based methods suffer from limitations such as line-of-sight restrictions, susceptibility to adverse weather and lighting conditions, and limited spatial coverage. Distributed acoustic sensing (DAS) offers a robust alternative by repurposing existing telecommunications dark fiber into dense, kilometer-scale sensor arrays. Nevertheless, interpreting the complex DAS signals generated by vehicular traffic remains challenging due to overlapping dynamic signatures and the scarcity of ground-truth data for model training. To overcome this, we present a cross-modal (vision-to-optic) supervision framework that transforms existing fiber infrastructure into a traffic monitoring system. We deployed a synchronized camera–DAS testbed along a roadway segment served by dark fiber. Video data is processed using modern computer vision foundation models SAM3 to automatically extract vehicle trajectories and classifications. These camera-derived labels supervise a deep sequence learning model trained on the corresponding DAS strain data. Once trained, the fiber-optic system independently achieves accurate vehicle detection, classification, localization, and speed estimation. Finally, we demonstrate how these continuous, DAS-derived traffic metrics can be directly translated into dynamic load profiles, providing a scalable, continuous monitoring solution for bridge fatigue and structural health assessment.