Wood, Jens Nikolas

In a former study carried out by Wood et al. (2022) and Wood and Breuer (2023) a novel approach for the generation of artificial wind gusts in a wind tunnel setup was presented denoted “the paddle”. The device generates wind gusts by dynamically blocking the nozzle outlet area with a rigid plate inducing a sudden rise and drop of the velocity signal. This procedure leads to highly reproducible wind gusts within a certain region of the test section depending on the kinematic settings of the paddle. However, outside of this restricted region the flow is dominated by a highly fluctuating flow regime. This effect limits the space of the test section, which can be used for experimental investigations. In order to reduce the negative effects of the paddle, a modified design of the gust generator is presented in this contribution. For this purpose, a second wind tunnel is designed using the same automation equipment as the paddle. However, the new setup comprises a nozzle with a fully movable upper contour in order to generate smoother wind gusts. The working principle and a comarison between the gusts generated by the paddle and the new device are presented based on a closely matched motion pattern of the servo driver unit and an identical blocking ratio. In summary, the adaptive nozzle reduces the drawbacks of the paddle such as the large flow separation and the velocity undershoot on the falling flank of the gust. Additionally, previously not attainable gust shapes can be generated leading to a greater variety of flow conditions for experimental studies on fluid-structure interaction (FSI).

The paper is concerned with a novel approach to generate horizontal wind gusts in classical wind tunnels. Assuming an open test section of an Eiffel- or Göttingen-type wind tunnel, the new wind gust generator denoted ‘‘The Paddle’’ can be easily retrofitted to such experimental setups at low costs. The device is constructed based on commercially available components including a programmable software tool that allows to adjust all relevant kinematic parameters. Five predefined motion patterns of the paddle are investigated which differ according to the achieved blocking ratio of the wind tunnel nozzle as well as the velocity and acceleration of the downward and upward motion of the paddle. It is shown that the shape and intensity of the generated gusts can be fully controlled by these parameters. That guarantees a strong individual adjustability and customization of the induced gusts. Furthermore, a synchronization of the wind gust generator with measurement devices such as laser-Doppler anemometer or particle-image velocimetry is easy to implement. An extensive measurement campaign has shown that the generated horizontal gusts are highly reproducible for the presently investigated laminar boundary layer and vary in the core area of the gust measurements solely according to free-stream turbulence of the wind tunnel.

© 2020 The Authors The present paper investigates the fluid–structure interaction (FSI) of a wing with two degrees of freedom (DOF), i.e., pitch and heave, in the transitional Reynolds number regime. This 2-DOF setup marks a classic configuration in aeroelasticity to demonstrate flutter stability of wings. In the past, mainly analytic approaches have been developed to investigate this challenging problem under simplifying assumptions such as potential flow. Although the classical theory offers satisfying results for certain cases, modern numerical simulations based on fully coupled approaches, which are more generally applicable and powerful, are still rarely found. Thus, the aim of this paper is to provide appropriate experimental reference data for well-defined configurations under clear operating conditions. In a follow-up contribution these will be used to demonstrate the capability of modern simulation techniques to capture instantaneous physical phenomena such as flutter. The measurements in a wind tunnel are carried out based on digital-image correlation (DIC). The investigated setup consists of a straight wing using a symmetric NACA 0012 airfoil. For the experiments the model is mounted into a frame by means of bending and torsional springs imitating the elastic behavior of the wing. Three different configurations of the wing possessing a fixed elastic axis are considered. For this purpose, the center of gravity is shifted along the chord line of the airfoil influencing the flutter stability of the setup. Still air free-oscillation tests are used to determine characteristic properties of the unloaded system (e.g. mass moment of inertia and damping ratios) for one (pitch or heave) and two degrees (pitch and heave) of freedom. The investigations on the coupled 2-DOF system in the wind tunnel are performed in an overall chord Reynolds number range of 9.66×103≤Re≤8.77×104. The effect of the fluid-load induced damping is studied for the three configurations. Furthermore, the cases of limit-cycle oscillation (LCO) as well as diverging flutter motion of the wing are characterized in detail. In addition to the DIC measurements, hot-film measurements of the wake flow for the rigid and the oscillating airfoil are presented in order to distinguish effects originating from the flow and the structure.

Membrane structures are of increasing interest for modern civil engineering due to their adaptable application. The safe assembly and operation of membranous buildings in urban regions is a challenging task due to permanently changing environmental conditions. A critical issue is the dynamic response of the flexible membrane to wind loads which has to be addressed as a primary design criterion for this type of structure in the future. The influence of wind loads on the deformable structure form a multi-physical problem since fluid and structure mechanics have to be considered simultaneously to encompass the whole problem. This leads to the motivation of this thesis in which the fluid-structure interaction (FSI) of a thin-walled and air-inflated membranous structure immersed in a turbulent boundary layer is investigated. The highly flexible structure has the shape of a hemisphere. In order to maintain its hemispherical form and to attain a resistance against wind loads, the flexible structure is pressurized by a slight gauge pressure pre-stressing the membrane. In this configuration, the membranous hemisphere is considered as an air-inflated building often seen as roofing for temporary facilities. The fluid-structure interaction of the flexible hemisphere immersed in a turbulent flow is experimentally investigated in a wind tunnel. For this purpose, an appropriate flexible model is manufactured using a casting procedure, where the structure of the hemisphere is based on a silicone material. A second fully rigid hemispherical model is manufactured out of aluminum serving as a reference for the flow field studies. The first investigation focuses on the turbulent flow around the rigid hemisphere at Re = 50,000. Large effort is put into the generation of the required thick turbulent boundary layer at the inlet of the test section of the wind tunnel, since it is an essential boundary condition of the experimental case. The flow around the solid bluff body is measured by laser-Doppler (LDA) and constant-temperature anemometry (CTA). Two characteristic vortex shedding processes are observed in the wake of the hemisphere: An asymmetric von Karman and an arch-type symmetric type. Both vortex patterns alternate in an irregular manner in time, where only one is present in the wake during a certain period of time. The time-averaged data reveal the characteristic phenomena forming around the hemisphere such as the horseshoe vortex system, the free shear layer and the recirculation region. All data are furthermore used for the successful validation of a large-eddy simulation which is taken from the literature. After this initial flow field study, the coupled problem is observed. For this purpose, the flexible hemisphere is once more exposed to a turbulent boundary layer at three Reynolds numbers (50,000, 75,000 and 100,000). This setup is used to examine the interaction between the flow and the pressurized membrane at different free stream velocities. Furthermore, the flow field around the rigid hemisphere is measured again in order to maintain comparability between the used measurement equipment and the extended Reynolds number range. The experiments are carried out by combining particle-image-velocimetry (PIV) for the flow field and high-speed digital-image correlation (DIC) measurements for the deformation of the oscillating membranous structure. Moreover, a constant-temperature anemometer is utilized in order to evaluate the velocity spectra at locations close to the wall. This is necessary to connect the non-synchronized fluid and structure measurements. Afterwards the spectra of the velocity fluctuations (CTA) and the structure oscillations (DIC) are compared. This procedure leads to the characterization of the underlying FSI mechanisms. As before, the two main vortex shedding types (von Karman and symmetric arch-type) are observed at all Reynolds numbers. These are also identified in the unsteady structure excitations. With increasing Re number the time-averaged deformations of the structure as well as the observed amplitudes of the oscillation increase. The displacements of the structure strongly influence the time-averaged flow field revealing a significant difference in the wake. A thorough analysis of the comprehensive data sets for the fluid flow and the displacements of the structure leads to the characterization of the behavior of the flexible structure under changing flow conditions. Again, the experimental results are supported by complementary numerical investigations based on large eddy simulations for the fluid and a finite-element solver for the structure taken from the literature.

The present study investigates the fluid-structure interaction (FSI) of a highly flexible membranous hemisphere immersed into a fully turbulent boundary layer flow. The experiments are carried out in a wind tunnel considering three Reynolds numbers (50,000, 75,000 and 100,000) defined by the diameter of the hemisphere and the free-stream velocity. A gauge pressure inside the hemisphere is used to stabilize the thin-walled structure against the outer wind loads. This setup is comparable to air-inflated structures often used in civil engineering. A combination of mono particle-image-velocimetry (PIV) measurements for the fluid field and three-dimensional digital-image correlation (DIC) measurements for the structure deformation are used to evaluate the occurring FSI phenomena. Additionally, a constant-temperature hot-wire probe is applied at specific locations in the flow field in order to link the independently gathered PIV and DIC data. Furthermore, all measurements are compared with the results of a rigid hemisphere leading to a characterization of the FSI between the turbulent flow field and the flexible structure. Besides the analysis and investigation of the occurring FSI phenomena, the measurements build the basis for validating complementary numerical high-fidelity simulations of the coupled problem.