PSI - Issue 52

Sidney Goossens et al. / Procedia Structural Integrity 52 (2024) 647–654 Sidney Goossens / Structural Integrity Procedia 00 (2023) 000 – 000

648

2

1. Introduction Composite materials, such as carbon fibre reinforced polymers (CFRP), are increasingly used in aerospace applications, because of the increased stiffness-to-mass ratio compared to their traditional metallic counterparts. The laminate nature of CFRP components however makes them prone to impact damage. An impact due to a bird- or lightning strike, runway debris, or even a maintenance tool that is accidentally dropped, creates so-called barely visible impact damage (BVID). A BVID consists of a barely visible indentation on the outside of the CFRP that is typically smaller than 0.3 mm, but also of delamination or reinforcement fibre damage inside the composite. Moreover, the BVID can increase in size due to the operational loads that are acting on the damaged structure, which may eventually compromise its load-bearing capabilities. In order to detect and repair BVIDs, airline operators have implemented time-based maintenance (TBM) approaches, adapted from the traditional maintenance strategies for metallic damage. The occurrence of a BVID can however not be predicted by TBM, and therefore overly frequent inspections are required. This increases downtimes and downtime costs associated with the aircraft. A permanently installed structural health monitoring (SHM) network performing condition-based monitoring (CBM) can overcome the drawbacks of TBM for BVID monitoring. Optical fibre sensors (OFS) have been heralded many times as suitable candidates for implementing such an SHM system. Standard optical fibres are made of glass and have an outer diameter of 125 μm, which makes them very small, and light compared to other types of sensors. They can therefore even be embedded within the laminates of a composite structure. Given that the fibre is made from a dielectric and that there is no electrical power supply required at the locations of the sensor, these types of sensors are immune to electromagnetic interference and crosstalk from nearby sensors. It is even possible to create tens, if not hundreds, of sensors in the same optical fibre, which significantly reduces the amount of cabling required for the sensor network. In this work, we use fibre Bragg grating (FBG) sensors that carry out strain measurements to capture the permanent change in residual strain in the composite material that is induced by the BVID (Batte et al., 2018; Okabe et al., 2002, 2004; Peters et al., 2001; Takeda et al., 2002, 2005, 2008; Takeda, Yamamoto, et al., 2007; Yashiro et al., 2017). This allows for performing inspection measurements in between flights, in on-ground conditions, and therefore does not require the data acquisition system to be on board of the aircraft, significantly limiting weight and power consumption. The detection of BVIDs with optical fibre sensors has already been reported earlier in open literature. This however typically happened in laboratory conditions, which did not account for compatibility with harsh aerospace environments, whilst we deploy an optical fibre with an installation method that has been tested against standardized in-flight conditions (Goossens et al., 2018, 2019). We moreover demonstrated the fibre and its installation method to be suited for BVID detection on plate-like structures (Goossens et al., 2020; Goossens, Berghmans, Sharif Khodaei, et al., 2021). Furthermore, only a handful of studies has demonstrated the used of OFS for damage detection on composite structures of higher geometrical complexity, such as e.g. a skin-stiffener (Takeda, Aoki, et al., 2007; Takeda et al., 2012; Tur et al., 2016). In this work we demonstrate, for the first time to the best of our knowledge, the detection of BVIDs with optical fibre sensors on an aerospace-grade real scale composite curved panel of 3 x 5 m. The panel was manufactured with state-of-the-art composite materials and stacking sequences for aerospace applications. The panel is stiffened by composite stringers of the same material and stacking sequence and by metal frames. We instrumented 4 portions of the damage- prone stiffener’s feet with 4 optical fibres, equipped with 30 FBG sensors per fibre, resulting in a total of 120 FBG sensors. We first performed a temperature calibration of the sensors attached to the panel, yielding the sensors response as a function of temperature, which allows to correct for the influence of temperature changes. From this correction, we extracted the temperature compensation error, which allowed us to define a damage detection threshold. Next, we performed controlled impacts on the panel to create calibrated BVIDs under the stiffener’s feet. We obtained the strain response from the FBG se nsors before and after the impact and compared said strain signals to detect the presence and the location of the BVID.