Dr Marco Placidi (he/him)

Dataset description associated with the article “Turbulence statistics estimation across a step change in roughness via interpretable network-based modelling”, (2024), G. Iacobello, M. Placidi, S. Ding, M. Carpentieri. Files description:- ‘u_timetraces.txt’, ‘v_timetraces.txt’, and ‘w_timetraces.txt' include the instantaneous velocity values for the three velocity components u, v, and w in [m/s].- ‘XYZ_LDA_coordinatees.txt' includes the measurement coordinates (in [m[) with respect to the reference system.- ‘x_coordinatees.txt' and ‘z_coordinatees.txt' include the unique x and z coordinates (streamwise and vertical coordinates, respectively) in [m].

A novel fast-running model is developed to predict the three-dimensional (3D) distribution of turbulent kinetic energy (TKE) in axisymmetric wake flows. This is achieved by mathematically solving the partial differential equation of the TKE transport using the Green's function method. The developed solution reduces to a double integral that can be computed numerically for a wake prescribed by any arbitrary velocity profile. It is shown that the solution can be further simplified to a single integral for wakes with Gaussian-like velocity-deficit profiles. Model results are compared and validated against detailed 3D laser Doppler anemometry data measured within the wake flow of a model wind turbine in a laboratory environment. This shows a remarkably good agreement in both the magnitude and shape of the radial TKE profiles at the turbine hub height. The wind-tunnel data also provide insights into the evolution of important turbulent flow quantities such as turbulent viscosity, mixing length, and the TKE dissipation rate within a wind turbine wake.

Wind tunnel experiments were conducted to understand the effect of building array size (N), aspect ratio (AR), and the spacing between buildings (W S) on the mean structure and decay of their wakes. Arrays of size 3×3, 4×4,and 5×5, AR = 4, 6, and 8, and W S = 0.5W B , 1W B , 2W B and 4W B (where W B is the building width) were considered. Three different wake regimes behind the building clusters were identified: near-, transition-, and far-wake regimes. The results suggest that the spatial extent of these wake regimes is governed by the overall array width (W A). The effects of individual buildings are observed to be dominant in the near-wake regime (0 < x/W A < 0.45) where individual wakes appear behind each building. These wakes are observed to merge in the transition-wake region (0.45 < x/W A < 1.5), forming a combined wake in which the individual contributions are no longer apparent. In the far-wake regime (x/W A > 1.5), clusters' wakes are akin to those developing downwind of a single isolated building. Accordingly, new local and global scaling parameters in the near-and far-wake regimes are introduced. The decay of the centreline velocity deficit is then modelled as a function of the three parameters considered in the experiment.

A series of wind tunnel experiments were conducted in the University of Surrey's Environmental Flow wind tunnel with a 1:50 scale of a typical London street canyon to assess the exposure of cyclists riding in a group to the emissions of polluting vehicles. A propane source emitted from an Ahmed body was used to model a car exhaust and a fast flame ionisation detector was used to measure pollutant concentration around four cyclists for multiple configurations of the source, cyclists, and wind directions. Two cases were investigated with a vehicle driving in front of a line of cyclists and adjacent to them (as if it were overtaking them). In the first case, for small wind incidence, findings confirm that the cyclists exposure decreases exponentially with their distance from the source with a small dependence on wind direction but largely independently of the riders position within the group. For large wind incidences, typical of urban canyons, the rider position within the group becomes more important. For the second set of experiments, with the vehicle positioned adjacent to the riders, it was found to be preferable for a rider to be in front of the group regardless of the distance from the source, as this results in lower exposure to pollutants. This is likely linked with the complex aerodynamic field generated by the group of riders that can trap the vehicle exhaust fumes amongst the cyclists, hence increasing the exposure. This research suggests that group riding should be considered when designing mitigation strategies to minimise cyclists exposure to road traffic pollution within urban environments, where busy and narrow cycle lanes often results in cyclists riding in line.

Wind tunnel experiments are performed in both neutral and stable boundary layers to study the effect of thermal stability on the wake of a single turbine and on the wakes of two axially aligned turbines, thereby also showing the influence of the second turbine on the impinging wake. In the undisturbed stable boundary layers, the turbulence length scales are significantly smaller in the vertical and longitudinal directions (up to 50 % and ≈≈40 %, respectively), compared with the neutral flow, while the lateral length scale is unaffected. The reductions are larger with the imposed inversion of a second stable case, except in the near-wall region. In the neutral case, the length scales in the wake flow of the single turbine are reduced both vertically and laterally (up to 50 % and nearly 40 %, respectively). While there is significant upstream influence of a second turbine (on mean and turbulence quantities), there is virtually no upstream effect on vertical length scales. However, curiously, the presence of the second turbine aids length-scale recovery in both directions. Longitudinally, each turbine contributes to successive reduction in coherence. The effect of stability on the turbulence length scales in the wake flows is non-trivial: at the top of the boundary layer, the reduction in the wall-normal length scale is dominated by the thermal effect, while closer to the wall, the wake processes strongly modulate this reduction. Laterally, the turbines’ rotation promotes asymmetry, while stability opposes this tendency. The longitudinal coherence, significantly reduced by the wake flows, is less affected by the boundary layer's thermal stability.

The time trial bicycle, helmet and mannequin digital models were obtained from a cloud-based open-source library of computer aided design files (Mundy, 2012; Nestell, 2015; Vinayagamoorthy, 2017, respectively), and were modified using SolidWorks 2022. The bicycle geometry features an open five-spoke front wheel, disk rear wheel, vortex generators at the seat tube and a standard time-trial handlebar without aero bar. Certain details of the bicycle, such as the crank, drive train and chain stay were not included since they were considered to have a negligible impact on the flow field, and to avoid part failure during the additive manufacturing process. The wheel hub and pedals were simplified. To guarantee the model was strong and stiff enough to survive the experimental environment, the seat stay and front fork were thickened. Vertical reinforcement features were added to the bicycles to fix the model to a thin structural base. Further details are contained in Arbelo Romero (2023).Please use any CAD software to open the step or stl files. We used a Prusa MINI+ (PLA with 60% infill) to 3D print the model in our work. Mundy, B. (2012). Time-trial bicycle. https://grabcad.com/library/tt-bike. Nestell, W. (2015). Time-trial helmet. https://grabcad.com/library/tt-race-helmet-1. Vinayagamoorthy, D. (2017). Flexi-robot assembled. https://grabcad.com/library/flexi-robot-assembled-1. Arbelo Romero, J. M. (2023). Strategies to obtain an aerodynamic advantage in time trials. MEng Thesis. University of Surrey.

Abstract from manuscript: "Wind tunnel experiments are performed in both neutrally and stable boundary layers in order to study the effect of thermal stability on the wake of a single turbine and on the wakes of two axially aligned turbines, thereby also showing the influence of the second turbine on the impinging wake. In the undisturbed stable boundary layer, the turbulence length scales are significantly smaller in the vertical and longitudinal directions (up to 50% and 30%, respectively), compared with the neutral flow, while the lateral length scale is unaffected. The reductions are larger still with the imposed inversion of a second stable case, except in the near-wall region. In the neutral case, the length scales in the wake flow of the single turbine are reduced both vertically and laterally (up to 50% and 40% respectively). While there is significant upstream influence of a second turbine (on mean and turbulence quantities), there is virtually no upstream effect on vertical length scales. However, curiously, the presence of the second turbine aids length-scale recovery in both directions. Longitudinally, each turbine contributes to successive reduction in coherence. The effect of stability on the turbulence length scales in the wake flows is non-trivial: at the top of the boundary layer, the reduction in the wall-normal length scale is dominated by the thermal effect, while closer to the wall, the wake processes strongly modulate this reduction. Laterally, the turbines’ rotation promotes asymmetry, while stability opposes this tendency. The longitudinal coherence, significantly reduced by the wake flows, is less affected by the boundary layer’s thermal stability.

Experiments were conducted in the fully rough regime on surfaces with large relative roughness height (h/δ ≈ 0.1, where h is the roughness height and δ is the boundary layer thickness). The surfaces were generated by distributed LEGOr bricks of uniform height, arranged in different configurations. Measurements were made with both floating-element drag balance and high-resolution particle image velocimetry on six configurations with different frontal solidities, λF, at fixed plan solidity, λP, and vice versa, for a total of twelve rough-wall cases. The results indicated that the drag reaches a peak value λF ≈ 0.21 for a constant λP = 0.27, while it monotonically decreases for increasing values of λP for a fixed λF = 0.15. This is in contrast to previous studies in the literature based on cube roughness which show a peak in drag for both λF and λP variations. The influence of surface morphology on the depth of the roughness sublayer (RSL) was also investigated. Its depth was found to be inversely proportional to the roughness length, y0. A decrease in y0 was usually accompanied by a thickening of the RSL and vice versa. Proper orthogonal decomposition (POD) analysis was also employed. The shapes of the most energetic modes calculated using the data across the entire boundary layer were found to be self-similar across the twelve rough-wall cases. However, when the analysis was restricted to the roughness sublayer, differences that depended on the wall morphology were apparent. Moreover, the energy content of the POD modes within the RSL suggested that the effect of increased frontal solidity was to redistribute the energy towards the larger scales (i.e. a larger portion of the energy was within the first few modes), while the opposite was found for variation of plan solidity.

High-resolution particle image velocimetry data obtained in rough-wall boundary layer experiments are re-analysed to examine the influence of surface roughness heterogeneities on wind resource. Two different types of heterogeneities are examined: (i) surfaces with repeating roughness units of the order of the boundary layer thickness (Placidi & Ganapathisubramani. 2015 J. Fluid Mech. 782, 541–566. (doi:10.1017/jfm.2015.552)) and (ii) surfaces with streamwise-aligned elevated strips that mimic adjacent hills and valleys (Vanderwel & Ganapathisubramani. 2015 J. Fluid Mech. 774, 1–12. (doi:10.1017/jfm.2015.228)). For the first case, the data show that the power extraction potential is highly dependent on the surface morphology with a variation of up to 20% in the available wind resource across the different surfaces examined. A strong correlation is shown to exist between the frontal and plan solidities of the rough surfaces and the equivalent wind speed, and hence the wind resource potential. These differences are also found in profiles of Ū2 and Ū3 (where U is the streamwise velocity), which act as proxies for thrust and power output. For the second case, the secondary flows that cause low- and high-momentum pathways when the spacing between adjacent hills is beyond a critical value result in significant variations in wind resource availability. Contour maps of Ū2 and Ū3

Experiments on the receptivity of two-dimensional boundary layers to acoustic disturbances from two-dimensional roughness strips were performed in a low-turbulence wind tunnel on a at plate model. The freestream was subjected to a plane acoustic wave so that a Stokes Layer (SL) was created on the plate, thus generating a Tollmien-Schlichting (T-S) wave through the receptivity process. An improved technique to measure the T-S component is described based on a retracting two-dimensional roughness, which allowed for phase-locked measurements at the acoustic wave frequency to be made. This improved technique enables both protuberances and cavities to be explored in the range 30m < jhj < 750m (equivalent to 0:025 < jhj=B < 0:630 in relative roughness height to the local unperturbed Blasius boundary layer displacement thickness). These depths are designed to cover both the predicted linear and non-linear response of the T-S excitation. Experimentally, cavities had not previously been explored. Results show that a linear regime is identifiable for both positive and negative roughness heights up to 150 m (jhj=B 0:126). The departure from the linear behaviour is, however, dependent on the geometry of the surface imperfection. For cavities of signicant depth, the non-linear behaviour is found to be milder than in the case of protuberances - this is attributed to the flow physics in the near field of the surface features. Nonetheless, results for positive heights agree well with previous theoretical work which predicted a linear disturbance response for small-height perturbations.

Wind-tunnel experiments were carried out on fully-rough boundary layers with large roughness (δ/h≈10 δ/h≈10, where h is the height of the roughness elements and δ δ is the boundary-layer thickness). Twelve different surface conditions were created by using LEGO™ bricks of uniform height. Six cases are tested for a fixed plan solidity (λ P λP) with variations in frontal density (λ F λF), while the other six cases have varying λ P λP for fixed λ F λF. Particle image velocimetry and floating-element drag-balance measurements were performed. The current results complement those contained in Placidi and Ganapathisubramani (J Fluid Mech 782:541–566, 2015), extending the previous analysis to the turbulence statistics and spatial structure. Results indicate that mean velocity profiles in defect form agree with Townsend’s similarity hypothesis with varying λ F λF, however, the agreement is worse for cases with varying λ P λP. The streamwise and wall-normal turbulent stresses, as well as the Reynolds shear stresses, show a lack of similarity across most examined cases. This suggests that the critical height of the roughness for which outer-layer similarity holds depends not only on the height of the roughness, but also on the local wall morphology. A new criterion based on shelter solidity, defined as the sheltered plan area per unit wall-parallel area, which is similar to the ‘effective shelter area’ in Raupach and Shaw (Boundary-Layer Meteorol 22:79–90, 1982), is found to capture the departure of the turbulence statistics from outer-layer similarity. Despite this lack of similarity reported in the turbulence statistics, proper orthogonal decomposition analysis, as well as two-point spatial correlations, show that some form of universal flow structure is present, as all cases exhibit virtually identical proper orthogonal decomposition mode shapes and correlation fields. Finally, reduced models based on proper orthogonal decomposition reveal that the small scales of the turbulence play a significant role in assessing outer-layer similarity.

An experimental investigation on the influence of the spatial frequency content of roughness distributions on the development of crossflow instabilities has been carried out. From previous research it is known that micro roughness elements can have a large influence on the crossflow development. When the spanwise spacing is chosen such that it is the most unstable wavelength (following linear stability analysis), stationary crossflow waves are amplified. While in earlier studies the focus was on the height or spanwise spacing of roughness elements, in the present study it is chosen to vary the shape of the elements. Through the modification of the shape the forcing at the critical wavelength is increased, while the forcing at the harmonics of the critical wavelength is damped. Experiments were carried in the low turbulence wind tunnel at City University London (Tu=0.006%) on a swept flat plate in combination with displacement bodies to create a sufficiently strong favourable pressure gradient. Hot wire measurements across the plate tracked the development of stationary and travelling crossflow waves. Initially, stronger crossflow waves were found for the elements with stronger forcing, while further downstream the effect of forcing diminished. Spatial frequency spectra showed that the stronger forcing at the critical wavelength (via the roughness shape) dominates the response of the flow while low forcing at the harmonics has no notable effect. Additionally, high resolution streamwise hot wire scans showed that the onset of secondary instability is not significantly influenced by the spatial frequency content of the roughness distribution.

In this work, we study the development of the internal boundary layer (IBL) induced by a surface roughness discontinu-ity, where the downstream surface has a roughness length greater than that upstream. The work is carried out in the EnFlo meteorological wind tunnel, at the University of Surrey, in both thermally neutral and stable cases with varying degrees of stability. For the neutrally-stratified boundary layer, the IBL development in the log-law region shows good agreement with the diffusion model proposed by Panofsky and Dutton (Atmospheric turbulence, Wiley, New York, 1984) provided that a modified origin condition is introduced and its growth rate is dictated by a constant diffusion term. However, the model over-predicts the growth of the IBL in the outer layer, where the IBL depth grows slowly with fetch following a power function with exponent n being 0.61 (whereas the original model prescribes n ≈ 0.8). For the stably-stratified boundary layers, n is found to further reduce as the bulk Richardson number, Ri b , increases. The analysis of the top region of the IBL shows that the slow growth rate is due to a combination of the decay of the diffusion term and a significantly negative mean wall-normal velocity, which transports fluid elements towards the wall. Considering these two effects, a modified diffusion model is proposed which well captures the growth of the IBL for both neutrally and stably-stratified boundary layers. Graphical abstract 1 Introduction

Wind tunnel experiments on the receptivity of three-dimensional boundary layers were performed in a range of freestream turbulence intensities, Tu, from 0.01%—the lowest level ever achieved in this type of work—up to 0.41%. This work confirms that for Tu=0.01%, and presumably below this level, the transition process is dominated by stationary modes. These are receptive to surface roughness and generate Type-I and Type-II secondary instabilities that eventually cause the transition to turbulence. The saturation amplitude of these stationary waves is highly sensitive to the level of environmental disturbances; the former is here recorded to be the highest in the literature, with the latter being the lowest. Travelling modes are still present; however, their influence on the transition process is marginal. At matched surface roughness levels, when the level of environmental disturbance is enhanced to Tu≥0.33%, the travelling modes acquire more importance, strongly influencing the laminar/turbulent transition process, whilst the initial amplitude and growth of the stationary modes are hindered. For this level of Tu, is the interaction of steady and unsteady disturbances that produces highly amplified waves (Type-III), that quickly lead to nonlinear growth and anticipated turbulence. Finally, a simple rule of thumb is proposed, where the transition front was found to move forward by roughly 10% chord for an increase in one order of magnitude in the Tu levels.

Wind-tunnel experiments were carried out on four urban morpholo-6 gies: two tall canopies with uniform height and two super-tall canopies with a 7 large variation in element heights (where the maximum element height is more 8 than double the average canopy height, h max =2.5h avg). The average canopy 9 height and packing density are fixed across the surfaces to h avg = 80 mm, 10 and λ p = 0.44, respectively. A combination of laser doppler anemometry and 11 direct-drag measurements are used to calculate and scale the mean velocity 12 profiles within the boundary-layer depth δ. In the uniform-height experiment, 13 the high packing density results in a 'skimming flow' regime with very little 14 flow penetration into the canopy. This leads to a surprisingly shallow rough-15 ness sublayer (z ≈ 1.15h avg), and a well-defined inertial sublayer above it. 16 In the heterogeneous-height canopies, despite the same packing density and 17 average height, the flow features are significantly different. The height het-18 erogeneity enhances mixing, thus encouraging deep flow penetration into the 19 canopy. A deeper roughness sublayer is found to exist extending up to just 20 above the tallest element height (corresponding to z/h avg = 2.85), which is 21 found to be the dominant length scale controlling the flow behaviour. Results 22 point toward the existence of a constant stress layer for all surfaces considered 23 herein despite the severity of the surface roughness (δ/h avg = 3 − 6.25). This 24 contrasts with previous literature. 25 Keywords Laser doppler anemometry · Turbulent boundary layers · Urban 26 roughness · Wind-tunnel experiments

Wind tunnel experiments were conducted in a low-turbulence environment (Tu < 0:006%) on the stability of 3D boundary layers. The effect of two different distributions of discrete roughness elements (DREs) on crossfl ow vortices disturbances and their growth was eval uated. As previously reported, DREs are found to be an effective tool in modulating the behaviour of crossfl ow modes. However, the effect of 24μm DREs was found to be weaker than previously thought, possibly due to the low level of environmental disturbances here with. Preliminary results suggest that together with the height of the DREs and their spanwise spacing, their physical distribution across the surface also intimately affects the stability of 3D boundary layers. Finally, crossfl ow vortices are tracked along the chord of the model and their merging is captured. This phenomena is accompanied by a change in the critical wavelength of the dominant mode.

Introduction and background Vegetation in both fresh and sea waters is not only ubiquitous in natural habitats but also instrumental for a variety of reasons. It provides the foundation for many food chains [4], contributes to the thriving of fish and corals [6], plays a role in reducing coastal erosion [1] and drastically improves the water quality by producing oxygen [3]. Furthermore, many engineering applications rely upon and would benefit from a better understanding of the flow physics characterising these problems. Despite the numerous reviews [2, 5, 6] that have attempted to capture different aspects of canopy flows over flexible vegetation, a satisfactory understanding of this topic is still elusive. For this reason, a simple controlled experiment aimed at comparing wall-bounded flows over rigid and flexible roughness was designed and carried out. Experimental facility and details Three different surfaces are considered in this work: a smooth wall and two rough-wall cases. The first rough surface is characterised by rigid roughness (i.e. conventional rough wall), while in the second case the flow develops over flexible roughness elements (i.e. aquatic vegetation). Experiments were designed to compare the statistical properties of flexiblerough beds as opposed to their rigid counterpart when the roughness height under wind loading, heff , is matched. The tests were carried out in the Donald Campbell wind tunnel at Imperial College London (freestream turbulence Tu < 0:5%U1). The tunnel working section measures 2:98 m in length, with a 1:37 m x 1:12 m cross section. The conditions were set to represent a nominally zero-pressure gradient at a freestream velocity of 12 ms

This work was presented at WESC 2019 in Cork.

An experimental investigation on the influence of the spatial frequency content of roughness distributions on the development of crossflow instabilities has been carried out. From previous research it is known that micro roughness elements can have a large influence on the crossflow development. When the spanwise spacing is chosen such that it is the most unstable wavelength (following linear stability analysis), stationary crossflow waves are amplified. While in earlier studies the focus was on the height or spanwise spacing of roughness elements, in the present study it is chosen to vary the shape of the elements. Through the modification of the shape the forcing at the critical wavelength is increased, while the forcing at the harmonics of the critical wavelength is damped. Experiments were carried in the low turbulence wind tunnel at City University London (Tu=0.006%) on a swept flat plate in combination with displacement bodies to create a sufficiently strong favourable pressure gradient. Hot wire measurements across the plate tracked the development of stationary and travelling crossflow waves. Initially, stronger crossflow waves were found for the elements with stronger forcing, while further downstream the effect of forcing diminished. Spatial frequency spectra showed that the stronger forcing at the critical wavelength (via the roughness shape) dominates the response of the flow while low forcing at the harmonics has no notable effect. Additionally, high resolution streamwise hot wire scans showed that the onset of secondary instability is not significantly influenced by the spatial frequency content of the roughness distribution.

Interest in laminar flow flight due to both economic and environmental factors has recently seen a resurgence (Tufts et al., 2017). Within this topic, the study of the detailed effect of surface excrescences on laminar/turbulent transition has received significant attention (Fage, 1943; Schlichting, 1979). However, most of the previous work has focused on the effect of steps in 2D environments (i.e. in the absence of a pressure gradient), while the effect of steps on a 3D wings has received less attention (Bender et al., 2005). Therefore, experiments on the stability of 3D boundary layers were performed in a very low turbulence wind tunnel by examining the effect of different excrescences, of a height of approximately one-third of the local displacement thickness, δ*, located at 20% chord. Three different stepped geometries (see figure 1) are considered in order to mimic the leading edge to wing box joint characterising new concepts of laminar flow wings. Results show, as expected, that all surface imperfections reduce the extent of the laminar flow region when compared to the case in the absence of a step. However, despite the severity of the excrescences, this reduction is very moderate, which suggests scope to relax current laminar flow wing tolerances. The pyramidal geometry (in figure 1c), with more gradual forward- and aft-facing steps is it found to be optimum, as the performance degradation is the lowest. Results also suggest that the different step geometries have an influence on both the excitation of the primary modes (and its harmonics) and the onset of the nonlinear phase of the instability. Further analysis will follow in the full paper.

Experiments on the stability of a 3D boundary layer were performed in a very low turbulence wind tunnel (Tu ͌ 0:006%U¥). The effect of different shapes of surface steps (of h = 200 mm) located at 20% chord were investigated by looking into the crossflow modes evolution and growth. Stable crossflow vortices were generated by the means of discrete roughness elements (DREs) positioned upstream of the steps. Preliminary results seem to suggest that the different step geometries have a severe influence on both the maximum disturbance growth and the excitation of the primary mode and its harmonics. These different surface imperfections also seem to play a critical role on the appearance of the non-linear phase of the instability. Finally, the different step geometries are shown to influence the transition front location by up to 9%, which results in performance degradation. The softer and more gradual geometrical disturbance (i.e. Pyramid-type step) was found to minimise the performance loss, providing that each step comprising the complex geometry is designed to be conservatively subcritical.

A triple slotted aerofoil following the Handley Page 44F design was tested at City University London T-2 wind tunnel. The model allowed the study of a fixed triple slotted wing as well as investigation of the effects of isolated slots at different locations along the chord. PIV measurements were performed within the chord Reynolds number range in between approximately 200,000-400,000. The model was tested at an angle of attack of 22o. Measurements of mean streamwise velocity, velocity fluctuations and shear stress were analysed. The study shows how an isolated slot is more favourable when it is placed closest to the leading edge, although slow moving fluid regions can still be found close to the trailing edge. Fully attached flow was only achievable by using all three slots. In addition, the fully slotted profile is shown to generate channel exit velocities in the order of 1.4U∞, which highly energise the boundary layer on the suction side.