PSI - Issue 80

Francisco J.G. de Oliveira et al. / Procedia Structural Integrity 80 (2026) 1–10 Author name / Structural Integrity Procedia 00 (2023) 000–000

2

2

1. Introduction

Fluid-structure interactions (FSI) arise whenever turbulent flows impose unsteady loads on solid bodies, coupling flow dynamics with structural response. These interactions underpin the performance of many engineered systems, ranging from reactor fuel rods to aircraft, bridges, and wind turbines. Understanding the mechanisms by which turbu lence imprints itself on structures is critical for designing systems that can withstand variable operating environments and avoid premature fatigue. In wind energy, this issue is particularly pressing: the majority of installed capacity now exists in farms where downstream turbines are unavoidably exposed to the wakes of their upstream neighbours. These wakes are characterised by strong velocity deficits, elevated turbulence intensities, and complex coherent structures, which combine to increase unsteady loads and reduce turbine lifetimes (Thomsen and Sørensen, 1999; Barthelmie et al., 2009; Veers et al., 2019). The role of free-stream turbulence (FST) in altering both flow development and structural response has long been recognised in canonical blu ff -body flows. For instance, the wake of a circular cylinder — one of the most fundamental test cases in fluid mechanics — exhibits modifications in vortex shedding, spanwise coherence, and load spectra when subjected to turbulent inflow (Bearman and Morel, 1983a; Williamson, 1996; de Oliveira et al., 2025). These changes result in enhanced fluctuating loads, reduced vortex formation length, and increased coherence of large-scale shedding structures. Such e ff ects directly impact engineering systems, where FST often governs unsteady loads (Chamorro and Porte´-Agel, 2011; Stevens and Meneveau, 2017). Traditionally, experimental studies of FSI have relied on load cells, sparse strain gauges, or pressure taps to quantify the mechanical imprint of turbulent flows. While useful, these methods provide either global load measurements or limited local information, making it di ffi cult to resolve how specific flow structures couple to distributed loading across complex geometries. Structural health monitoring (SHM) techniques provide insightful ways to retrieve temporally and spatially resolved data (Pan et al., 2025), capable of resolving / capture fine details imprinted in the structure from the surrounding flow. From the available SHM techniques, fibre-optic sensing o ff ers a powerful alternative (Zhou et al., 1999; de Oliveira et al., 2024; Li and Sharif-Khodaei, 2025). In particular, Rayleigh backscattering (RBS) fibre sensors o ff er millimetre-scale distributed strain measurements along a single fibre, with negligible impact on the flow or structure (Zhou et al., 1999; Xu and Sharif-Khodaei, 2020; Li and Sharif-Khodaei, 2025). By combining such measurements with temporally resolved flow diagnostics, one can directly relate turbulent structures to their structural imprint (de Oliveira et al., 2024). In this paper, we unify results from two complementary experimental campaigns that exploit this methodology. The first considers a cantilevered cylinder in a water flume, serving as a canonical blu ff -body benchmark. By com bining distributed RBS strain sensing with time-resolved particle image velocimetry (PIV), we resolve the influence of FST on vortex shedding, tip-vortex formation, and their imprint on the fluctuating strain experienced by an acrylic cantilevered cylinder. The second extends this approach to an upscaled wind turbine model tested in the 10 × 5wind tunnel at Imperial College London. One blade was instrumented with a sinusoidal fibre-optic layout, connected via an optical slip-ring at the hub, enabling continuous spanwise strain measurements under rotating conditions. Simulta neous hot-wire anemometry mapped the wake velocity field over multiple tip-speed ratios ( λ ) and inflow turbulence intensities. These measurements reveal how blade dynamics couple to wake structures, highlighting the dominant role of tip-vortex harmonics and turbulence-enhanced coherence in governing the blade response. The remainder of this paper is structured as follows. Section 2 introduces the experimental methodologies for both the cylinder and turbine campaigns, including fibre-optic sensor layout, flow diagnostics, and operating conditions. Section 3 presents results from the cylinder case, establishing the sensitivity of RBS to canonical turbulent structures, and strain as a surrogate to flow dynamics at the cylinder’s wake. Section 4 extends the analysis to the turbine case, quantifying the spanwise heterogeneity of blade response and its coupling with wake dynamics. Finally, Section 5 synthesises the findings to demonstrate the broader potential of fibre-optic sensing for advancing FSI research and guiding wind energy applications.

2. Experimental Methodology

The experimental programme consisted of two complementary campaigns: (i) a cantilevered cylinder in a turbulent cross-flow, serving as a canonical blu ff -body case for developing and validating the fibre-optic framework, and (ii)