This thesis deals with ways to suppress turbulent flow in pipelines. In the first part various methods of laminarization are presented, when the turbulent flow is transformed into laminar flow, including the results of experiments published by the authors. The next part presents the results from CFD. The calculations were performed for one of the methods mentioned in the first part and the results were compared with each other. In addition, several options have been suggested to improve the original method.

Le sujet de cette thèse est l'hydrodynamique de corps minces (feuilles) et élancés (filamenteux) de fluide visqueux immergés dans un second fluide ayant une viscosité différente. Nous nous concentrons sur deux exemples : la subduction de la lithosphère océanique et le flambage de fils visqueux dans microcanaux divergents, les deux ont un nombre de Reynolds caractéristique Re<<1. Pour le cas de la subduction d'une feuille mince, nous proposons une hybride méthode «boundary integral & thin sheet» (BITS). Après la validation en comparant ses prévisions avec celles de la boundary-element méthode, deux solutions instantanées et dépendant du temps sont effectués pour analyser la subduction avec la méthode BITS. L'analyse à l'échelle de la vitesse d'immersion normalisée en fonction de «la rigidité en flexion» de la feuille est confirmée par nos prédictions numériques. Pour des rapports de viscosité modérée (≈100), la feuille amincit sensiblement quand elle coule, mais pas assez pour conduire à la «rupture de la dalle» que l'on observe dans plusieurs zones de subduction sur Terre. Ensuite, le code BLEU parallèle pour écoulements polyphasiques est utilisé à simuler pliage visqueux tridimensionnel dans des microcanaux divergent. Nous avons réalisé une étude paramétrique comprenant cinq simulations dans lequel le rapport de débit volumétrique, le rapport de viscosité, le nombre de Reynolds, et la forme de la chaîne ont été modifiées par rapport à un modèle de référence. Le fil devient instable à une instabilité de pliage en raison de la contrainte de compression longitudinale. L'axe de pliage initial peut être parallèle ou perpendiculaire à la dimension étroite de la chambre. Dans le premier cas, le pliage transforme lentement au pliage perpendiculaire au moyen d'une torsion, ou peut disparaître totalement. / The hydrodynamics of thin (sheet-like) and slender (filamentary) bodies of viscous fluid immersed in a second fluid with a different viscosity is studied. Here we focuses on two examples: the subduction of oceanic lithosphere and the buckling of viscous threads in diverging microchannels, both have a characteristic Reynolds number Re<<1. A hybrid boundary integral & thin sheet method (BITS) is build for the subduction of 2D immersed sheet. After the validation by comparing with the results of full boundary elements method, both instantaneous and time-dependant soloutions are done to analyze the subduction with the BITS method. The scaling analysis of the normalized sinking speed V/V_Stokes as a function of the sheet's 'flexural stiffness' is confirmed by our numerical predictions. For moderate viscosity ratios (≈100), the sheet thins substantially as it sinks, but not enough to lead to the ‘slab breakoff’ that is observed in several subduction zones on Earth. Next, the parallel code BLUE for multi-phases flows is used to simulate the 3-dimensional viscous folding in diverging microchannels. We performed a parameter study comprising five simulations in which the flow rate ratio, the viscosity ratio, the Reynolds number, and the shape of the channel were varied relative to a reference model. The thread becomes unstable to a folding instability due to the longitudinal compressive stress. The initial folding axis can be either parallel or perpendicular to the narrow dimension of the chamber. In the former case, the folding slowly transforms via twisting to perpendicular folding , or may disappear totally.

Small-scale wind turbines were not considered viable in the past due to their poor efficiencies, mainly because of their aerodynamic effects around the irfoil shape. Recently researchers have renewed interest in enhancing the aerodynamic performances of the blades’ designs inspired by the aerodynamic pattern of biological characteristics of insects and marine mammals such as locusts, dragonflies, damselflies, Humpback Whales etc. Bioinspired wing designs have advantages compared to conventional smooth irfoil blades as they can counter the bending forces that the wings experience during flapping. Bio-inspired corrugated airfoil based on dragonfly wing geometries have been reported to perform well compared to conventional airfoil at low Reynolds numbers. Corrugated airfoils reduce flow separation and enhance aerodynamic performance by trapping vortices in the corrugations thus drawing flow towards the airfoil’s surface. This results in the higher lift whilst incurring only marginally higher drag. Such airfoils also have an advantage when it comes to span-wise structural stiffness due to the corrugated cross-sections. Replacing conventional turbine blades by tubercles or corrugated blades could enhance turbine performance by reducing the pressure gradient along the leading edge; however, the aerodynamic effects at the leading edge will depend on the variations of wavelength and amplitude. In this study, two types of computational studies were investigated: Optimising a corrugated airfoil and investigating the aerodynamic effects of a sinusoidal shape at the leading edge of a blade. Previous studies used an idealized geometry based on the dragonfly wing cross-section profile but did not attempt to optimize the geometry. In the present study: a two-dimensional CFD model is constructed using ANSYS Fluent Workbench-Design Explorer to determine the optimal corrugated blade profile for four angles of attack (AOA) from 5° to 20° corresponding to typical AOA of small-scale wind turbine blades. Two modified blades with variations of wavelength and amplitude at the leading edge were studied to investigate the aerodynamic effects. Three-dimensional models were constructed using Qblade software and 3D points were exported to AutoCAD Inventor to generate the CAD model. The governing equations used are continuity and Navier-Stokes equations written in a frame reference rotating with the blade. The CFD package used is ANSYS FLUENT 19.0. The simulation was run under steady-state, using SST-k omega turbulence model. The modifications have improved the aerodynamic performance. The optimised corrugated blade produced a maximum increase of CL and L/D. Both modified blades (1 and 2) had their performances measured separately and compared to that of baseline blade SG6042 (Conventional blade). Modified blade 1 had a lower wavelength and amplitude at the leading edge of 14.3 % and 4 % respectively of the chord. It was noted that the aerodynamic performance decreased by 6%. Modified model 2, on the other hand had a higher wavelength and amplitude at the leading edge. of 40.4 % and 11.9 % respectively of the chord. It was also noted the aerodynamic performance increased by 6%. From the empirical evidence highlighted above, it can be observed that there is a direct correlation between wavelength, amplitude, and aerodynamic performance of the blade. / Electrical and Mining Engineering / M. Tech. (Engineering)

<p>Highly loaded airfoils in low-pressure turbines (LPTs) suffer from laminar flow separation from the suction side of the airfoils aft of the throat of the passages. This separation harms the performance of the engine by reducing the power extraction from the turning air and ultimately reduces the overall turbine efficiency. Flow control techniques have been investigated to eliminate flow separation in aerodynamic surfaces to abate the losses associated with it. This Master of Science Thesis investigates the design, implementation and testing of pulsated injection actuation in a low-Reynolds flow over a wall-mounted hump.</p> <p>Furthermore, this Thesis expands on the existing expertise in the infrared (IR) thermography measurement technique at the Purdue Experimental Turbine Aerothermal Lab. This is done through an investigation of the factors affecting the IR measurement technique and the development of an optical instrument (borescope) to implement in an annular cascade wind tunnel. IR thermography is used on the wall-mounted hump blowdown tests to detect the separation point in the boundary layer using two techniques: by an investigation of the surface temperature distribution and an investigation of the heat transfer behavior at the surface. Finally, the borescope is commissioned through the first testing campaign of the LPT airfoils, and are processed to thermally investigate the passage.</p> <p>This thesis succeeds in expanding the IR capabilities within PETAL, and at demonstrating pulsated injection as an effective method to eliminate flow separation. Furthermore, IR successfully detects flow separation on the wall-mounted hump through the two methods presented, as well as detecting the boundary layer reattachment caused by the flow control technique. The limitations of the thermal methodology, as well as those of the optical probe are addressed, and the uncertainties in the measurements are quantified. Finally, steps to continue the studies are suggested at the end of each methodology chapter, including the potential redesign of the IR borescope to improve the quality of measurements. </p>

The presence of vegetation in open channel flow has a significant influence on flow resistance, turbulence structures and sediment transport. This study will evaluate flow resistance and scale velocity profile in depth limited flow conditions, specifically investigating the impact of vegetation on the flow resistance under submerged flow conditions. The resistance induced by vegetation in open channel flows has been interpreted differently in literature, largely due to different definitions of friction factors or drag coefficients and the different Reynolds numbers. The methods utilized in this study are based on analytical and numerical models to investigate the effects of vegetation presence on flow resistance in open channel flows. The performing strategy approach was applied by three-dimensional computational fluid dynamics (CFD) simulations, using artificial cylinders for the velocity profile. This is to estimate the average flow velocity and resistance coefficients for flexible vegetation, which results in more accurate flow rate predictions, particularly for the case of low Reynolds number. This thesis shows different formulas from previous studies under certain conditions for a length scale metric, which normalises velocity profiles of depth limited open channel flows with submerged vegetation, using both calculated and simulated model work. It considers the submerged vegetation case in shallow flows, when the flow depth remains no greater than twice the vegetation height. The proposed scaling has been compared and developed upon work that have been influenced by logarithmic and power laws to present velocity profiles, in order to illustrate the variety of flow and vegetation configurations.

No description available.

No description available.

No description available.

No description available.

Modelling solid-fluid multiphase flows are crucial to many applications such as fluidized beds, pyrolysis and gasification, catalytic cracking etc. Accurate modelling of the fluid-particle forces is essential for lab-scale and industry-scale simulations. Fluid-particle system solutions can be obtained using various techniques including the macro-scale TFM (Two fluid model), the meso-scale CFD-DEM (CFD - Discrete Element Method) and the micro-scale PRS (Particle Resolved Simulation method). As the simulation scale decreases, accuracy increases but with an exponential increase in computational time. Since fluid forces have a large impact on the dynamics of the system, this study trains deep learning models using micro-scale PRS data to predict drag forces on ellipsoidal particle suspensions to be applied to meso-scale and macro-scale models. Two different deep learning methodologies are employed, multi-layer perceptrons (MLP) and 3D convolutional neural networks (CNNs). The former trains on the mean characteristics of the suspension including the Reynolds number of the mean flow, the solid fraction of the suspension, particle shape or aspect ratio and inclination to the mean flow direction, while the latter trains on the 3D spatial characterization of the immediate neighborhood of each particle in addition to the data provided to the MLP. The trained models are analyzed and compared on their ability to predict three different drag force values, the suspension mean drag which is the mean drag for all the particles in a given suspension, the mean orientation drag which is the mean drag of all particles at specific orientations to the mean flow, and finally the individual particle drag. Additionally, the trained models are also compared on their ability to test on data sets that are excluded/hidden during the training phase. For instance, the deep learning models are trained on drag force data at only a few values of Reynolds numbers and tested on an unseen value of Reynolds numbers. The ability of the trained models to perform extrapolations over Reynolds number, solid fraction, and particle shape to predict drag forces is presented. The results show that the CNN performs significantly better compared to the MLP in terms of predicting both suspensions mean drag force and also mean orientation drag force, except a particular case of extrapolation where the MLP does better. With regards to predicting drag force on individual particles in the suspension the CNN performs very well when extrapolated to unseen cases and experiments and performs reasonably well when extrapolating to unseen Reynolds numbers and solid fractions. / M.S. / Multiphase solid-fluid flows are ubiquitous in various industries like pharmaceuticals (tablet coating), agriculture (grain drying, grain conveying), mining (oar roasting, mineral conveying), energy (gasification). Accurate and time-efficient computational simulations are crucial in developing and designing systems dealing with multiphase flows. Particle drag force calculations are very important in modeling solid-fluid multiphase flows. Current simulation methods used in the industry such as two-fluid models (TFM) and CFD-Discrete Element Methods (CFD-DEM) suffer from uncertain drag force modeling because these simulations do not resolve the flow field around a particle. Particle Resolved Simulations (PRS) on the other hand completely resolve the fluid flow around a particle and predict very accurate drag force values. This requires a very fine mesh simulation, thus making PRS simulations many orders more computationally expensive compared to the CFD-DEM simulations. This work aims at using deep learning or artificial intelligence-based methods to improve the drag calculation accuracy of the CFD-DEM simulations by learning from the data generated by PRS simulations. Two different deep learning models have been used, the Multi-Layer Perceptrons(MLP) and Convolutional Neural Networks(CNN). The deep learning models are trained to predict the drag forces given a particle's aspect ratio, the solid fraction of the suspension it is present in, and the Reynolds number of the mean flow field in the suspension. Along with the former information the CNN, owing their ability to learn spatial data better is additionally provided with a 3D image of particles' immediate neighborhood. The trained models are analyzed on their ability to predict drag forces at three different fidelities, the suspension mean drag force, the orientation mean drag, and the individual particle drag. Additionally, the trained models are compared on their abilities to predict unseen datasets. For instance, the models would be trained on particles of an aspect ratio of 10 and 5 and tested on their ability to predict drags of particles of aspect ratio 2.5. The results show that the CNN performs significantly better compared to the MLP in terms of predicting both suspension mean drag force and also mean orientation drag force, except a particular case of extrapolation where the MLP does better. With regards to predicting drag force on individual particles in the suspension, the CNN performs very well when extrapolated to unseen cases and experiments and performs reasonably well when extrapolating to unseen Reynolds numbers and solid fractions.