The performance of several turbulence closure models in stagnating turbulent flow is investigated and their deficiencies identified. The simulations are compared with new measurements of the two-dimensional stagnation flow on a flat plate in a wind tunnel obtained using stereo-particle image velocimetry. Numerical solutions to the Poisson equation for the instantaneous pressure allowed all pressure-velocity correlations to be estimated and then separated into their linear (mean strain rate dependent) and nonlinear contributions. Reynolds-averaged Navier-Stokes (RANS) turbulence models were investigated because of their widespread use in, for example, urban wind flow simulations, where impinging flow is common. The models tested are the k - epsilon and some variants as well as a common Reynolds stress model with and without wall reflection. As is well-known from previous studies, the k - epsilon model over-predicted k. Durbin's modified eddy viscosity reduces, but does not correct this over-prediction. The dominance of the turbulent diffusion in the turbulent kinetic energy equation near the wall was not reproduced by any of the gradient diffusion models, suggesting major deficiencies in the modeled terms.

Wind-power ramps are a significant source of uncertainty in wind-energy forecasting and are a challenge to electric-grid stability. In the Canadian province of Alberta, strong westerly winds buffet the Rocky Mountains creating an abundant yet intermittent wind energy resource in the plains of Alberta. In the current study, wind-power ramp events have been detected and correlated to several environmental factors including time-of-day, atmospheric stability, season and a Föhn wind event known locally as a Chinook wind at a field wind measurement station downstream of the Rocky Mountains. Large wind-power ramps (a 50% change in power in less than 4 h) were found to occur on days when a Föhn wind was present over 50% of the time. The result highlights the importance of this meteorological phenomenon to wind energy production locally and also in regions where Föhn winds occur. The detected wind-power ramps were found to vary significantly with season, with the strongest wind-power ramps emanating from the Rocky Mountains in the winter months under stable atmospheric conditions.

This manuscript investigates the role of atmospheric stability on the intermittency of turbulence within the atmospheric boundary layer. Wind turbines are designed for a prescribed level of turbulence intensity but are required to operate under varying atmospheric stability conditions. Longitudinal and vertical velocity and virtual temperature increments are investigated under varying atmospheric stability conditions over numerous multi-day periods. The results indicate a complex relationship between stability and turbulence intermittency. The analysed data is shown to be leptokurtic on all time scales. Importantly, turbulence on small time scales is shown to be highly intermittent irrespective of atmospheric stability and poorly predicted by Gaussian statistics.

A study of the wake behind a Glauert model rotor in a water channel is reported. Phase-locked planar PIV is used to extract signatures of the coherent structures in the wake. The results show for first time both the tip and root vortices behind a wind turbine model operating at the run-away condition. Features of the wake are reported and evolution of the vortical filaments are discussed.

As a first step toward characterizing coherent structures within the atmospheric surface layer (ASL), measurements obtained via a large-scale particle tracking velocimetry (LS-PTV) system were validated against wind-measurement station data as well as canonical turbulent boundary layer studies. The LS-PTV system resolves three-dimensional, Lagrangian tracks over a 16 m3 volume. Mean-velocity measurements, as well as vertical and shear Reynolds-stress measurements, generally agreed with wind-measurement station data and Reynolds-stress profiles referenced from literature. The probability distributions for streamwise, spanwise and vertical velocity-fluctuation components appear normally distributed about zero. Furthermore, the probability distributions for all three components of Lagrangian acceleration were exponential and followed the parametrization curve from LaPorta et al. (Lett Nat 409:1017–1019, 2001). Lastly, the vorticity probability distributions were exponential and symmetric about zero, which matches findings from Balint et al. (Fluid Mech 228:53–86, 1991). The vorticity intensity measured by the LS-PTV system was less than values from Priyadarshana et al. (Fluid Mech 570:307–346, 2007), which is attributed to the low spatial resolution. However, the average spacing of 0.5 m between tracer particles is deemed sufficient for the future characterization of vortical structures within the ASL.

Analysis of the helical vortices measured behind a model wind turbine in a water channel are reported. Phase-locked measurements using planar particle image velocimetry are taken behind a Glauert rotor to investigate the evolution and breakdown of the helical vortex structures. Existing linear stability theory predicts helical vortex filaments to be susceptible to three unstable modes. The current work presents tip and root vortex evolution in the wake for varying tip speed ratio and shows a breaking of the helical symmetry and merging of the vortices due to mutual inductance between the vortical filaments. The merging of the vortices is shown to be steady with rotor phase, however, small-scale non-periodic meander of the vortex positions is also observed. The generation of the helical wake is demonstrated to be closely coupled with the blade aerodynamics, strongly influencing the vortex properties which are shown to agree with theoretical predictions of the circulation shed into the wake by the blades. The mutual inductance of the helices is shown to occur at the same non-dimensional wake distance.

The vortical near wake of a model horizontal axis wind turbine has been investigated experimentally in a water channel. The objective of this work is to study vortex interaction and stability of the helical vortex filaments within a horizontal axis wind turbine wake. The experimental model is a geometrically scaled version of the Tjæreborg wind turbine, which existed in western Denmark in the late 1980s. Here, the turbine was tested in both the upwind and downwind configurations. Qualitative flow visualisations using hydrogen bubble, particle streakline and planar laser-induced fluorescence techniques were combined with quantitative data measurements taken using planar particle image velocimetry. Vortices were identified using velocity gradient tensor invariants. Parameters that describe the helical vortex wake, such as the helicoidal pitch, and vortex circulation, were determined for three tip speed ratios. Particular attention is given here to the root vortex, which has been investigated minimally to date. Signatures of the coherent tip vortices are seen throughout the measurement domain; however, the signature of the root vortex is only evident much closer to the rotor plane, irrespective of the turbine configuration. It is postulated that the root vortex diffuses rapidly due to the effects of the turbine support geometries.

A large-scale particle tracking velocimetry (LS-PTV) system has been developed as a means of quantifying coherent structures within the atmospheric boundary layer. The LS-PTV system resolves three-dimensional, Lagrangian tracks over a volume of approximately 16m3. Mean velocity and Reynolds stresses have been validated and compared with wind-mast measurements and boundary-layer similarity formulations. The probability distributions for streamwise, spanwise and vertical velocity-fluctuation components agree with Gaussian distributions of equal variance. In contrast, the probability distributions for acceleration and vorticity are exponential and symmetric about zero. Thus, extreme acceleration events and vorticity events would be underpredicted by a Gaussian distribution. Finally, examples of particle paths exhibiting a high degree of swirl are presented, which are speculated to be signatures of large vortical structures.

The aerodynamic drag of an arrow is of importance in relation to the arrow’s drift in wind and to its down-range velocity. A significant contributor to that drag is the viscous drag from the arrow shaft, and consequently the nature of air flow over the arrow point and the location of the transition from laminar to turbulent flow are of interest. In this paper the flow was investigated using a scale model in a water channel for two arrow point profiles and for circumferential gaps at the rear of the arrow point. The normal ‘bullet point’ was found to have laminar flow along the front of the shaft and transition at a Reynolds number of approximately 450,000, and that circumferential gaps did not affect the flow. The frequently used ‘short bulge point’ was found to have flow separation at the rear taper of the point and turbulent flow for the full length of the shaft, which would be expected to result in greater drag than for the bullet point or for a bulge point with less aggressive rear taper.

An experimental investigation of the recirculation zone formed downstream of a forward facing step immersed in a turbulent boundary layer has been undertaken using particle image velocimetry. Bluff body flow is observed with the fixed separation point located at the leading edge of the step. The recirculation region dimensions are characterised over a range of Reynolds numbers (1400–19000), with Reh based on the step height and the free stream velocity. Turbulent perturbations are produced in the free shear layer which develops between the recirculating flow close to the step and the free stream flow. Contour maps of amplification factor, streamwise perturbation velocity and Reynolds stresses are constructed, providing insight into optimal placement of structures within such topographical features. The mechanisms affecting the reattachment distance, namely the turbulent mixing within the boundary layer and the velocity deficit in the boundary layer, are discussed.

Flow separation experiments over a forward facing step immersed in a turbulent boundary layer subjected to various levels of freestream turbulence intensity have been undertaken in a water channel. Freestream turbulence was generated using two traditional grids and a third, novel tethered sphere design, which was shown to dramatically increase the turbulence intensity produced. Planar particle image velocimetry was used to characterise the mean recirculation region. The dynamic of the reattachment length was also investigated using instantaneous velocity realisations. It was found that bluff body geometry effects were dominant over the freestream turbulence intensity close to the separation point. Downstream of the separation point, the turbulence level was seen to increase mixing between the high momentum freestream flow and the adverse flow within the recirculation region promoting reattachment.

Flow separation experiments are conducted in two facilities using particle image velocimetry (PIV) and pressure tappings over a wide Reynolds number range for a forward facing step immersed in a turbulent boundary layer. Bluff body flow is observed with the fixed separation point located at the leading edge of the model. The recirculation region dimensions are characterised over the entire Reynolds number range using the PIV technique. Pressure tappings provide insight into the surface pressure distribution in a recirculation region. The mechanisms affecting the reattachment distance, XL, namely the turbulent mixing within the boundary layer and the velocity deficit in the boundary layer are commented on.

The vortical wake of the Tjaereborg wind turbine has been investigated experimentally in a water channel. Both the upwind and downwind configurations were tested. The hydrogen bubble and particle streak techniques were used to visualize the flow prior to obtaining quantitative data using planar particle image velocimetry. Parameters that describe the helical vortex wake, such as the helix pitch, vortex core radius, vortex circulation and vortex meander, were determined for three tip speed ratios, λ=4,7,10. Particular attention was given to the characteristics of the root vortex, which have not previously been measured experimentally. A key finding of the investigation was the rapid diffusion of the coherent root vortex due to the small radius of curvature of the root vortex filament, interaction with the tower section, when in the upwind configuration, and its location near the nacelle boundary layer. The root vortex signal extends further into the wake in the downwind turbine configuration, as there is no tower-vortex interaction, prior to its being diffused in the strong strain field immediately downstream of the nacelle. A coherent tip vortex signal is present over the entire measurement area x/R<2, however the root vortex signal subsides much closer to the rotor plane. The study is part of a more extensive study of the vortex interactions within the wake. The mechanisms affecting wake stability are also briefly commented on.

This thesis investigates two flow fields that a wind turbine in a wind farm might experience. The first is the near-wake of a wind turbine where the focus was on the helical tip and root vortices. Two scale model wind turbines were investigated using PIV. A geometrically scaled rotor was observed to generate a chaotic wake due to poor aerodynamic performance at the experimental Reynolds number. Flow visualisations on a static 3D wing confirmed laminar separation is likely on the geometrically scaled model. A research orientated model was designed based on the optimum Glauert rotor to investigate the stability of the tip and root vortices. A pairing instability was observed in the tip vortices. The onset of this instability was found to be dependent on the tip speed ratio. The root vortices become unstable due to their proximity to the turbine support structures. The influence of freestream turbulence was studied by varying the turbulence intensity using passive turbulence grids including a novel tethered sphere grid. Increased turbulence intensity was observed to hasten the breakdown of the vortices, via turbulent diffusion rather than the pairing instability which was absent in the phase-locked average velocity fields. Tip and root vortices of both turbine models were characterised using Galilean invariant vortex identification schemes. The meander of the vortices was observed to be Gaussian at early vortex ages when interaction between vortices is minimal. Meander magnitude was shown to increase with distance absolutely. However, the magnitude of meander was found to be dependent on tip speed ratio and freestream turbulence intensity. A secondary focus was the flow fields above complex terrain features, a common location for wind farms. The recirculation region which formed downstream of various escarpment geometries was characterised using PIV. The size of the recirculation region was found to be dependent on the escarpment angle, the boundary layer to step height thickness ratio, the Reynolds number and the freestream turbulence intensity. An application of POD phase-averaging revealed the dynamic nature of the recirculation region. The wind speed-up above the escarpment beneficial in a wind energy sense was observed to be coupled to a vertical velocity component. Further, significant turbulence generation in the separated shear was observed which questions the appropriateness of complex terrain as a wind farm location.