L’objet de ce travail de thèse est l’étude en soufflerie du sillage lointain des éoliennes à axe horizontal. La complexité phénoménologique du sillage des éoliennes fait que les mécanismes le régissant sont généralement traités d’un point de vue stationnaire : les principaux paramètres (déficit de vitesse, production de turbulence...) sont alors moyennés dans le temps. Néanmoins, considérer les instationnarités du sillage d’une éolienne placée en écoulement atmosphérique permet d’observer un phénomène appelé meandering, traduisant un battement aléatoire du sillage. Ce travail est construit en deux grandes parties : la vision stationnaire et la vision instationnaire du sillage d’un modèle simplifié d’éolienne, basé sur le principe du disque de Froude, placé dans une couche limite atmosphérique (CLA) modélisée en soufflerie à l’échelle 1/400. La première partie est composée de plusieurs études paramétriques sur l’influence du point de fonctionnement d’une éolienne, sur sa hauteur de mât ou encore sur le type de CLA. Une éolienne en situation de dérapage fait également l’objet d’études paramétriques. Celles-ci visent à étudier le comportement du sillage dans différentes situations et il est alors montré les limites des lois empiriques présentes dans la littérature, concernant notamment l’influence de la turbulence ambiante. La seconde partie se focalise sur la vision instationnaire, jusqu’ici rarement considérée. En utilisant l’anémométrie par fil chaud, les résultats montrent le rôle des grandes échelles de la turbulence atmosphérique sur l’apparition du meandering. L’amplitude du battement et les dimensions du sillage instantané sont ensuite appréhendées de façon quantitative par mesures PIV, montrant l’influence du point de fonctionnement de l’éolienne et de l’intensité de turbulence ambiante. / The aim of this work was to study the far wake of horizontal axis wind turbines in wind tunnels. Aerodynamic phenomena within the wakes are very complex and, most of the time, they are studied from a steady point of view : the main parameters (wake deficit, production of turbulence...) are therefore time averaged. Nevertheless, studying the wake unsteadiness of a wind turbine located in the atmospheric boundary layer (ABL) enables the consideration of the meandering phenomenon, which describes random oscillations of the wake. The present work was constructed in two main parts : firstly the steady vision and secondly the unsteady vision of a wind turbine wake, modelled according to the actuator disk theory and placed in an ABL reproduced in a wind tunnel at a geometric scale of 1/400. Several parametric studies are presented in the first part : on the influence of the wind turbine’s operating point, of its mast height and also on the influence of the ABL characteristics. Wind turbines in yaw are also considered. The wake behaviour is then studied in different configurations and the limits of the empirical laws in the literature are highlighted, especially the lack of the ambiant turbulence intensity consideration. The second part focuses on the unsteady point of view, rarely considered until today. Using hot wire anemometry, the role of the atmospheric large turbulent scales on the meandering phenomenon is proven. The oscillation magnitude and the instantaneous wake dimensions are also investigated using PIV, which leads to quantitative results on the meandering characteristics.

Corrugated steel pipe (CSP) culverts are widely used as an economical alternative for conveying streams and small rivers through road embankments. While passage of the design flow is generally the primary goal for culvert design, consideration must also be given to maintaining connectivity within the aquatic environment for fish and other aquatic organisms. In Canada, the design criteria for fish passage through culverts are generally specified in terms of a maximum mean flow velocity corresponding to the weakest swimming fish expected to be found at a specific location. Studies have shown, however, that the velocity distribution within a CSP culvert may provide sufficient areas of lower velocity flow near the culvert boundary to allow for fish passage, even when the mean flow velocity may exceed a fishs swimming ability. Improved knowledge of the hydraulic conditions within CSP culverts, combined with research into fish swimming capabilities and preferences, may make it possible to better tailor culvert designs for fish passage while at the same time decreasing construction costs. To meet the requirements of regulators, various measures may be taken to reduce culvert flow velocities. Embedding, or setting the invert of a culvert below the normal stream bed elevation, is a simple and inexpensive method of increasing the flow area in a culvert flowing partially full, thereby decreasing flow velocity. Fish traversing through an embedded culvert benefit not only in terms of lower mean flow velocities, but also even lower flow velocities in the near boundary region. In the province of Saskatchewan culvert embedment is regularly used as a means to improve fish passage conditions. In this study, a laboratory scale model was used to study the velocity distribution within a non-embedded and embedded CSP culvert. An acoustic Doppler velocimeter was used to measure point velocities throughout the flow cross section at several longitudinal locations along the culvert. The hydraulic conditions were varied by changing the discharge, culvert slope and depth of embedment. The point velocity data were analyzed to determine patterns of velocity and turbulence intensity at each cross section, as well as along the length of the culvert. The results from the embedded culvert tests were compared with the results from the equivalent non-embedded tests, so that initial conclusions could be made regarding the use of embedment to improve conditions for fish passage. Analysis of the cross section velocity distributions showed that, even the non-embedded culvert had a significant portion of the flow area with flow velocity less than the mean velocity. The results from the embedded tests confirmed that embedding the culvert reduced the flow velocity throughout each cross section, although the effect was most significant for the cross sections located greater than one culvert diameter downstream from the inlet. This variation in effectiveness of embedment at reducing flow velocities is attributed to the length of the M1 backwater profile relative to the culvert length, and thus the differential increase in flow depth that occurred at each measurement location along the culvert. For both the non-embedded and embedded culvert the peak point magnitudes of turbulence intensity were found to be located near the culvert inlet where the flow was contracting. In terms of the cross section average turbulence intensity, in the non-embedded culvert turbulence increased with distance downstream from the inlet and was highest at the cross sections located near the culvert outlet. Embedding the culvert was found to either have no impact, or to slightly increase, the cross section average turbulence intensity near the inlet. Again, a result that is attributed to the tapering out of the M1 backwater profile at locations near the inlet under the flow conditions tested. However, beyond eight culvert diameters downstream from the inlet, embedment did result in lower cross section average turbulence intensity when compared to the non-embedded culvert. The measured velocity profiles for the non-embedded tests were found to compare well to the theoretical log-law velocity distribution using a ks value of between 0.012 m and 0.022 m, or approximately one to two times the corrugation amplitude, when the datum for analysis was considered to be located at the crest of the pipe corrugation. The cross section velocity distributions for the non-embedded tests compared very well to the model proposed by Ead et al. (2000). Based on this assessment, it appears that the Ead et al. model is potentially suitable for use in predicting the amount of the cross sectional area in a non-embedded culvert with flow velocity less than the design target for culvert fish passage design purposes. Overall, the results of the study confirm that, embedding a CSP culvert may be an effective way to improve fish passage conditions in terms of both flow velocity and turbulence intensity.

Corrugated steel pipe (CSP) culverts are widely used as an economical alternative for conveying streams and small rivers through road embankments. While passage of the design flow is generally the primary goal for culvert design, consideration must also be given to maintaining connectivity within the aquatic environment for fish and other aquatic organisms. In Canada, the design criteria for fish passage through culverts are generally specified in terms of a maximum mean flow velocity corresponding to the weakest swimming fish expected to be found at a specific location. Studies have shown, however, that the velocity distribution within a CSP culvert may provide sufficient areas of lower velocity flow near the culvert boundary to allow for fish passage, even when the mean flow velocity may exceed a fishs swimming ability. Improved knowledge of the hydraulic conditions within CSP culverts, combined with research into fish swimming capabilities and preferences, may make it possible to better tailor culvert designs for fish passage while at the same time decreasing construction costs. To meet the requirements of regulators, various measures may be taken to reduce culvert flow velocities. Embedding, or setting the invert of a culvert below the normal stream bed elevation, is a simple and inexpensive method of increasing the flow area in a culvert flowing partially full, thereby decreasing flow velocity. Fish traversing through an embedded culvert benefit not only in terms of lower mean flow velocities, but also even lower flow velocities in the near boundary region. In the province of Saskatchewan culvert embedment is regularly used as a means to improve fish passage conditions. In this study, a laboratory scale model was used to study the velocity distribution within a non-embedded and embedded CSP culvert. An acoustic Doppler velocimeter was used to measure point velocities throughout the flow cross section at several longitudinal locations along the culvert. The hydraulic conditions were varied by changing the discharge, culvert slope and depth of embedment. The point velocity data were analyzed to determine patterns of velocity and turbulence intensity at each cross section, as well as along the length of the culvert. The results from the embedded culvert tests were compared with the results from the equivalent non-embedded tests, so that initial conclusions could be made regarding the use of embedment to improve conditions for fish passage. Analysis of the cross section velocity distributions showed that, even the non-embedded culvert had a significant portion of the flow area with flow velocity less than the mean velocity. The results from the embedded tests confirmed that embedding the culvert reduced the flow velocity throughout each cross section, although the effect was most significant for the cross sections located greater than one culvert diameter downstream from the inlet. This variation in effectiveness of embedment at reducing flow velocities is attributed to the length of the M1 backwater profile relative to the culvert length, and thus the differential increase in flow depth that occurred at each measurement location along the culvert. For both the non-embedded and embedded culvert the peak point magnitudes of turbulence intensity were found to be located near the culvert inlet where the flow was contracting. In terms of the cross section average turbulence intensity, in the non-embedded culvert turbulence increased with distance downstream from the inlet and was highest at the cross sections located near the culvert outlet. Embedding the culvert was found to either have no impact, or to slightly increase, the cross section average turbulence intensity near the inlet. Again, a result that is attributed to the tapering out of the M1 backwater profile at locations near the inlet under the flow conditions tested. However, beyond eight culvert diameters downstream from the inlet, embedment did result in lower cross section average turbulence intensity when compared to the non-embedded culvert. The measured velocity profiles for the non-embedded tests were found to compare well to the theoretical log-law velocity distribution using a ks value of between 0.012 m and 0.022 m, or approximately one to two times the corrugation amplitude, when the datum for analysis was considered to be located at the crest of the pipe corrugation. The cross section velocity distributions for the non-embedded tests compared very well to the model proposed by Ead et al. (2000). Based on this assessment, it appears that the Ead et al. model is potentially suitable for use in predicting the amount of the cross sectional area in a non-embedded culvert with flow velocity less than the design target for culvert fish passage design purposes. Overall, the results of the study confirm that, embedding a CSP culvert may be an effective way to improve fish passage conditions in terms of both flow velocity and turbulence intensity.

While many studies have been conducted in channels that are fully open or entirely covered with ice, little has been examined in channels with border ice. To begin filling in this gap in knowledge, experiments were conducted in the Hydraulics Research Testing Facility at the University of Manitoba in a 1.2m wide, 14m long flume to assess the impact of Froude number, coverage ratio and bed-to-ice roughness ratio on the hydraulic characteristics of channels with border ice cover. Acoustic Doppler velocimetry was used to collect detailed 3D velocity data over a cross section of the flume which facilitated the analysis of velocity and turbulence intensity fields as well as the shear stress distribution. The data collected thus far represents the most complete experimental dataset of measured water velocities in a channel with a border ice cover. / February 2016

This disquisition aims for the study of turbulence intensity influence over the power performance of different sizes of turbines with the intent to validate a hypothesis. The hypothesis formulated for the analysis is the relationship between the rotor diameter (turbine size) and turbulence intensity. The hypothetical relationship is that the smaller turbines tend to experience more influence on the power performance from the turbulence in comparison with larger ones. For this examination, three different wind turbines of models Vestas V90, V100, V126 were chosen from three Swedish wind farms. The power performance of turbines at various levels of turbulence intensity were analyzed and the power deviation from the mean value due to influence of turbulence were assessed. The power deviation values of different turbines were compared at same level of wind speeds and also the power coefficients at same level of tip speed ratios were compared to validate the hypothesis. It was observed that the hypothesis seemed to appear true as higher influence on power curves were observed on V90 compared to others. Nevertheless, there were some obscene results which might be due to several factors such as influence of variation in hub height, site and inadequacy of data. / Detta examensarbete syftar till att studera hur ett vindkraftverks storlek påverkar inflytande från turbulens på effektuttaget. Hypotesen är att vindkraftverk med mindre rotordiameter påverkas mer av turbulens än de större. Tre vindkraftverksmodeller (Vestas V90, V100 och V126) från svenska vindkraftsparker valdes ut. De olika modellernas effektuttag för olika grader av turbulens analyserades och avvikelsen från effektmedelvärdet jämfördes. Effektavvikelserna samt verkningsgradsavvikelsen  för de olika vindkraftverksmodellerna jämfördes vid samma vindhastighet respektive löptal för att kunna testa hypotesen. Hypotesen styrks då den mista modellen (Vestas V90) påverkas mest av turbulens. Resultatet har dock troligtvis påverkats av andra faktorer såsom tornhöjd, terräng och en begränsad mängd data.

Este trabalho apresenta um estudo experimental acerca da influência da razão de espaçamento e da intensidade de turbulência no escoamento biestável. É estudado o escoamento sobre dois cilindros dispostos lado a lado, para o canal aerodinâmico com intensidade de turbulência inferior a 1%, para o canal aerodinâmico com grade promotora de turbulência com intensidade de turbulência de 8 a 7%, e para uma seção do canal aerodinâmico sem paredes laterais. A técnica experimental aplicada consiste na medição das flutuações de velocidade no canal aerodinâmico utilizando a técnica da anemometria de fio quente. As medições são realizadas com duas sondas, uma fixa e outra móvel que foi posicionada em seis pontos diferentes do escoamento. Os dados medidos são tratados através de ferramentas estatísticas, análise espectral e transformadas de ondaletas. Os resultados confirmam a existência do fenômeno da biestabilidade e mostram como o fenômeno sofre alterações quando exposto às condições de maior intensidade de turbulência e redução da razão de bloqueio pela retirada das paredes laterais em uma seção do canal. Para a condição de maior intensidade de turbulência, os resultados mostram que as trocas de modo se intensificam, e em alguns dos pontos de medição os sinais referentes à sonda móvel não apresentam características de biestabilidade. Para o caso sem paredes laterais, os sinais apresentam características de biestabilidade até o momento em que as paredes são removidas. / This work presents an experimental study, about the influence of the pitch to diameter ratio and the turbulence intensity on a bistable flow. The flow through two cylinders side-by-side is studied in an aerodynamic channel with turbulence intensity lower than 1%, in an aerodynamic channel with a turbulence grid, that increased the turbulence intensity to about 8 to 7%, and a section of the aerodynamic channel without lateral walls. The experimental technique consists in the measurements of the fluctuating velocities in the aerodynamic channel, using hot wire anemometry. The measurements are performed with two probes, one static, and one movable, positioned in six different points of the flow. The acquired data are analysed through statistical, spectral and wavelet tools. The results confirmed the occurrence of the bistable phenomenon and shows how it behaves when exposed to different conditions of turbulence intensity, and reduction of the blockage ratio aspect due to removal of the lateral walls of a channel section. The results for the higher turbulence intensity condition shows that the switches are intensified, some data resulting of the measurements of the second probe, do not show bistable characteristics. In the no-walls section case, the signals showed bistable features, until the moment the walls were removed.

Much of the aerosol research completed at the University of South Florida has revolved around evaluating industrial hygiene equipment and instrumentation in environmental chambers. Data collected during these studies has provided valuable baseline data on equipment and instrument performance under calm air conditions. A newly constructed wind tunnel now allows researchers to evaluate industrial hygiene instruments under moving air conditions. Because the wind tunnel is capable of producing wind velocities that a worker could encounter in the occupational setting, researchers may gain insight into instrument performance under simulated field conditions. Because aerosols can be introduced into the new wind tunnel testing section, researchers can also challenge industrial hygiene equipment and instrumentation with aerosols in sizes ranges that are of interest in public health. The purpose of this dissertation research was to develop a new wind tunnel to be used for aerosol research at the University of South Florida. Three specific aims had to be met for this study to be successful. They included: (1) designing a wind tunnel based on best practice information outlined in scientific literature, (2) constructing an operable wind tunnel to be used for aerosol research, and (3) characterizing wind tunnel performance by examining the wind tunnel velocity profile, turbulence intensity, and aerosol introduction/collection. The actual wind tunnel was constructed to a length of approximately 20 feet, a height of approximately 2 feet at its tallest point, and includes an entrance filter housing, a settling chamber, a contraction, a testing section, a diffuser, an exit filter housing, a fan, and exhaust duct. All components were designed and constructed using guidelines and best practices reported in the scientific literature. Velocity profile measurements were the first way that this wind tunnel was characterized. In order to successfully obtain measurements, the wind tunnel cross section was divided into 16 equal quadrants. Five measurements were taken for each quadrant at each wind velocity. Target wind velocities for this research were 0.5 m/s, 1.0 m/s, and 2.0 m/s. Actual average wind velocities of 0.48 m/s, 1.00 m/s, and 2.04 m/s. All were within established limits reported in the scientific literature. Turbulence intensity measurements were the second way that this wind tunnel was characterized. In order to successfully obtain measurements, the wind tunnel cross section was divided into 16 equal quadrants. Five measurements were taken for each quadrant at each wind velocity. Wind tunnels are typically designed to have the lowest turbulence intensity possible, generally below 10%. The overall average turbulence intensities for this wind tunnel at wind velocities of 0.5 m/s, 1.0 m/s, and 2.0 m/s were 9%, 10%, and 8% respectively. Overall turbulence intensity measurements were at or below 10%. Isokinetic sampling was the final method used to characterize this wind tunnel by collecting and detecting aerosols traveling through the wind tunnel testing section. The wind tunnel was operated at wind velocities of 0.5 m/s, 1.0 m/s, and 2.0 m/s with isokinetic sampling flow rates of 15.4 L/min, 30.9 L/min, and 61.7 L/min respectively. Monodisperse fluorescent polystyrene latex spheres were used as the test aerosol because they are uniform in size and shape and can be detected by fluorometry. The Blaustein Multi-Jet Atomizer (BLAM) was used to generate monodisperse fluorescent polystyrene latex aerosol 0.5 µm and 2.0 µm particles from liquid suspensions. The Vilnius Aerosol Generator (VAG) was used to generate monodisperse fluorescent polystyrene latex aerosol of 6.0 µm and 12.0 µm particles from dry powders. Nitrogen gas was used for delivering test aerosols into the wind tunnel. Five experimental runs were completed for each particle size and wind velocity for a total of 60 experimental runs. Fluorescence was detected in all 60 samples with average mass concentrations ranging from 0.000050 ng/ml to 0.002703 ng/ml. Based on velocity profile measurements, turbulence intensity measurements, and isokinetic sampling, the performance of University of South Florida wind tunnel was found to be excellent, indicating that it was designed and constructed appropriately. The wind tunnel can now successfully be used by researchers interested in evaluating industrial hygiene sampling equipment with aerosols ranging from 0.5 µm to 12.0 µm in moving air with velocities ranging from 0.5 m/s to 2.0 m/s.

<p>In the human body, turbulent flow is associated with many complications. Turbulence typically occurs downstream from stenoses and heart valve prostheses and at branch points of arteries. A proper way to study turbulence may enhance the understanding of the effects of stenoses and improve the functional assessment of damaged heart valves and heart valve prostheses.</p><p>The methods of today for studying turbulence in the human body lack in either precision or speed. This thesis exploits a magnetic resonance imaging (MRI) phenomenon referred to as signal loss in order to develop a method for estimating turbulence intensity in blood flow.</p><p>MRI measurements were carried out on an appropriate flow phantom. The turbulence intensity results obtained by means of the proposed method were compared with previously known turbulence intensity results. The comparison indicates that the proposed method has great potential for estimation of turbulence intensity.</p>

In the human body, turbulent flow is associated with many complications. Turbulence typically occurs downstream from stenoses and heart valve prostheses and at branch points of arteries. A proper way to study turbulence may enhance the understanding of the effects of stenoses and improve the functional assessment of damaged heart valves and heart valve prostheses. The methods of today for studying turbulence in the human body lack in either precision or speed. This thesis exploits a magnetic resonance imaging (MRI) phenomenon referred to as signal loss in order to develop a method for estimating turbulence intensity in blood flow. MRI measurements were carried out on an appropriate flow phantom. The turbulence intensity results obtained by means of the proposed method were compared with previously known turbulence intensity results. The comparison indicates that the proposed method has great potential for estimation of turbulence intensity.

An experimental investigation into fully developed high aspect ratio channels was undertaken. A review of the literature reveals that there is a need for accurate measurement of the inner peak value of streamwise turbulence intensity despite the large number of studies already completed. The scattered data on this subject could be attributed either to insufficient channel size (aspect ratio or length) or to hot-wire spatial filtering. A new, high quality, channel flow facility was designed and constructed, considering the most recent geometric limitation provided in the literature. To obtain accurate results, data were acquired using hot-wire probes with constant viscous-scale sensing length and were corrected using the most recent correction formula proposed by Smits et al. (2011). The results show dependence of inner peak value on Reynolds number in channels flow - its magnitude increasing with increasing Reynolds number.