No description available.

The automation of airfoil design process is an ongoing effort within the field of turbo-machinery design, with significant focus on developing new reliable and consistent methods that can meet the needs of the engineers. A wide variety of approaches has been in use for inverse airfoil design process which benefit from theoretical inverse design, statistical methods, empirical discoveries and many other ways to solve the design problem. This thesis work also develops a tool in Python to be used in airfoil aerodynamic design process that is simple, fast and accurate enough for initial design of turbo-machinery blades with focus on turbine airfoils used for operation in aircraft engines. To convey the decision-making process during development a simplified case is presented. The underlying considerations are discussed. Other available methods in the literature used for similar problems, are also evaluated and compared to demonstrate the advantages and limitations of the methods used within the tool. The inverse design problem is formulated as a multi-objective optimization problem to handle various different objectives that are relevant for aerodynamic design of turbo-machinery airfoils. Test runs are made and the results are discussed to assess how robust the tool is and how the current capabilities can be modified or extended. After the development process, the tool is verified to be a suitable option for real-life design optimization tasks and can be used as a building block for a much more comprehensive tool that may be developed in the future. / Automatisering av processen för design av vingprofiler kräver fortlöpande insatser inom området turbomaskindesign, med stort fokus på att utveckla nya tillförlitliga och konsekventa metoder som kan tillgodose ingenjörernas behov. Ett stort antal olika tillvägagångssätt har provats för omvänd design av vingprofiler såsom teoretisk invers design, statistiska metoder, empiriska upptäckter och många andra sätt att lösa designproblemet. Detta avhandlingsarbete är också ett lyckat försök att utveckla ett verktyg i Python som ska användas i den aerodynamiska designprocessen; det är enkelt, snabbt och noggrant för den initiala designen av turbomaskinblad med fokus på turbinblad som för användning i flygmotorer. För att förmedla beslutsprocessen under utvecklingen presenteras ett förenklat fall. De underliggande övervägandena diskuteras. Andra tillgängliga metoder i litteraturen som används för liknande problem utvärderas och jämförs för att visa fördelarna och begränsningarna med de metoder som används i verktyget. Det omvända designproblemet formuleras som ett multi-objektivt optimeringsproblem för att hantera olika mål som är relevanta för aerodynamisk design av turbomaskiner. Testkörningar görs och resultaten diskuteras för att bedöma hur robust verktyget är och hur de nuvarande funktionerna kan modifieras eller utökas. Efter utvecklingsprocessen verifieras verktyget som ett lämpligt alternativ för verkliga designoptimeringsuppgifter och kan användas som en byggsten för ett mycket mer omfattande verktyg som kan utvecklas i framtiden.

Magnetic bearings support turbomachinery by regulating their forces exerted in relation to the displacement of the machine supported. The regulating control system must be tuned for stable and safe operation of the rotor. The ultimate goal of this study is to determine the effects of changing controller gains on the behavior of the rotor during operation in its normal speed range with a known unbalance load. We also endeavor to confirm the model of the rotor supported the magnetic bearings, as an additional goal. We first investigate the modelling of rotors supported by magnetic bearings, including the model of the control system. We present a finite element model of a magnetic bearing supported rotor, and perform experiments to determine the characteristics of the control system which governs the magnetic forces on the rotor. The experimental control system characteristics confirm the expected characteristics from theory. With this knowledge, we perform simulations and experiments under the same forcing conditions to determine the accuracy of the model in predicting the experimental behavior of an unbalanced rotor. The model exhibits satisfactory ability in predicting the experimental behavior of the rotor under this loading. Our next step is to determine the effects of variation of proportional and integral controller gains on the behavior of the rotor. Both simulations and experiments show that an increase in the proportional controller gain results in an increase in the rotor’s first critical speed. An increase in the integral gain results in a small decrease in the location of the peak response speed in the speed range tested, while leaving the peak amplitude insignificantly changed. Again, simulations and experiments predict this result. We reach the following three conclusions from this study. First, the finite element model of the rotor/bearing system is a viable model for predicting the behavior of the experimental system. Second, tuning of the proportional gain shows a significant effect on the behavior of the rotor during unbalance loading across its speed range, due to considerable change in bearing stiffness caused by the tuning of this gain. Last, tuning of the integral gain has a small effect on the behavior of the rotor due to the change in bearing damping, too small to be considered significant. / Master of Science

Significant effort in the current technological development of aircraft is aimed at improving engine efficiency, while reducing fuel burn, emissions, and noise levels. One way to achieve these is to better integrate airframe and propulsion system. Tighter integration, however, may also cause adverse effects to the flow entering the engines, such as total pressure, total temperature, and swirl distortions. Swirl distortions are angular non-uniformities in the flow that may alter the functioning of specific components of the turbomachinery systems. To investigate the physics involved in the ingestion of swirl, pre-determined swirl distortion profiles were generated through the StreamVane method in a low-speed wind tunnel and in a full-scale turbofan research engine. Stereoscopic particle image velocimetry (PIV) was used to collect three-component velocity fields at discrete planes downstream of the generation of the distortions with two main objectives in mind: identifying the physics behind the axial development of the distorted flow; and describing the generation of the distortion by the StreamVane and its impact to the flow as a distortion generating device. Analyses of the mean velocity, velocity gradients, and Reynolds stress tensor components in these flows provided significant insight into the driving physics. Comparisons between small-scale and full-scale results showed that swirl distortions are Mach number independent in the subsonic regime. Reynolds number independence was also verified for the studied cases. The mean secondary flow and flow angle profiles demonstrated that the axial development of swirl distortions is highly driven by two-dimensional vortex dynamics, when the flow is isolated from fan effects. As the engine fan is approached, the vortices are axially stretched and stabilized by the acceleration of the flow. The flow is highly turbulent immediately downstream of the StreamVane due to the presence of the device, but that vane-induced turbulence mixes with axial distance, so that the device effects are attenuated for distances greater than a diameter downstream, which is further confirmed by the turbulent length scales of the flow. These results provide valuable insight into the generation and development of swirl distortion for ground-testing environments, and establishes PIV as a robust tool for engine inlet investigations. / Ph. D. / In order to improve performance of the next generation of aircraft, engineers are developing research that aims at reducing fuel consumption, improving the efficiency of engines, and also decreasing the levels of produced noise. There are several ways to achieve these goals, but significant effort has been focused on modifying the position of the engines on the aircraft to improve the properties of the airflow entering them. Computational simulations and small-scale tests have shown that this approach can be beneficial, while also showing that adverse effects to the properties of the air can be caused, affecting the behavior of the propulsion system. This current work makes use of a technique called StreamVane™ to reproduce those modified airflows in laboratory testing environments in order to understand how that flow might behave in the inlet of an engine, and what effects it could cause. This helps scientists and engineers decide if those modifications to the engine would be worth the time and money investments to the aircraft even before a full-scale model of the aircraft is built. More specifically, this work is an experimental investigation of two different types of distortions to the inlet airflow that could be caused by the aforementioned novel aircraft configurations, or by existing ones that have not been fully described yet.

<p dir="ltr">Despite the associated time, cost, and effort, experimental fan research remains necessary to validate computational models and physically develop new technologies. The need for a new fan research facility that would provide high quality experimental fan research at engine-representative speeds using detailed flow measurements was identified by the Office of Naval Research (ONR). The facility would be used to develop stall margin enhancement techniques, namely casing treatments to advance the field. In addition to support by ONR, Honeywell Aerospace donated a transonic fan rig and core exhaust plenum to make this project a reality.</p><p dir="ltr">The new research facility was designed and built around this new fan rig for investigations into casing treatments, inlet distortion, and aeromechanics research, as well as future projects that would make use of the new space. The funding package included a renovation of the build room in ZL1 and two brand new test cells constructed in previously empty space. All necessary equipment was designed, procured, and placed in the correct positions to ensure operability of the fan. The new space necessitated a mechanical checkout and commissioning process before conducting research projects.</p><p dir="ltr">In parallel to the development of the facility, a novel fan casing was designed to make use of rapid prototyping to experimentally test casing treatments. The fan casing assembly is made up of three metal components that remained fixed and six individual 3D printed plastic inserts that make up the flowpath surrounding the rotor. The geometry of each component was developed according to best-practices and computational structural analysis. Following commissioning of the fan test cell, the new fan casing was successfully implemented and tested over the full operating range of the fan.</p>

This thesis presents a joint computational and experimental investigation into the heat transfer to unshrouded turbine blade tips suitable for use in high bypass ratio, large civil aviation turbofan engines. Both the heat transfer to the blade tip and the over-tip leakage flow over the blade tip are characterised, as each has a profound influence on overall engine efficiency. The study is divided into two sections; in the first, computational simulations of a very large scale, low speed linear cascade with a flat blade tip were conducted. These simulations were validated against experimental data collected by Palafox (2006). A thorough assessment of turbulence models and minimum meshing requirements was performed. The standard k-ω and standard k-ϵ turbulence models significantly overpredicted the turbulence levels within the tip gap. The other models were very similar in performance; the SST k-ω and realisable k-ϵ models were found to be the most suitable for the flow environment. The second section documents the development and testing of a novel hybrid blade tip design, the squealet tip, which seeks to combine the known benefits of winglet and double squealer tips. The development of the external geometry was performed primarily through engine-representative CFD simulations at a range of tip gaps from 0.45% to 1.34% blade chord. The squealet tip was found to have a similar aerodynamic sensitivity to tip clearance as a baseline double squealer tip, with a tip gap efficiency exchange rate of 2.03, although this was 18% greater than the alternative winglet tip. The squealet tip displayed higher predicted stage efficiency than the winglet tip over the majority of the range of tip clearances investigated, however. The overall heat load was reduced by 14% compared with the winglet tip but increased by 28% over the double squealer tip, primarily due to the change in wetted surface area. The predicted local heat transfer coefficients were similar across all geometries. A realistic internal cooling plenum and an array of blade tip cooling holes were subsequently added to the squealet tip geometry and the cooling configuration refined by the selective sealing of cooling holes. Film cooling performance was largely assessed by the predicted adiabatic wall temperature distributions. A viable cooling scheme which reduced the cooling air requirement by 38% was achieved, compared to the initial case which had all cooling holes open. This was associated with just a 7% increase in blade tip heat flux and no penalty in peak temperature on the blade tip. Film cooling air ejected from holes on the blade suction side was swept away from the blade tip region, making the squealet rim at the crown of the blade particularly challenging to cool. It was demonstrated that this region could be cooled effectively by ballistic cooling from holes located on the blade tip cavity floor, although this was expensive in terms of the mass flow rate of cooling air required. The computational results were reinforced with experimental data collected in a transonic linear cascade. Downstream aerodynamic loss measurements were taken for a linearised version of the squealet tip design without cooling at nominal tip gaps of 0.45%, 0.89% and 1.34% blade chord, which was compared to similar data taken by O’Dowd (2010) for flat and winglet tips. The squealet was seen to have a similar aerodynamic loss to the flat tip and a reduced loss compared with the winglet tip. Full surface heat transfer measurements were taken for the uncooled squealet tip, at tip gaps of 0.89% and 1.34% blade chord, and for two configurations of the cooled squealet tip, at a tip clearance of 0.89% blade chord. The qualitative similarity between the measured heat transfer distributions and the those predicted by the engine-representative CFD simulations was good. A CFD simulation of the uncooled linear cascade environment at the 1.34% blade chord tip clearance was performed using a single blade with translationally periodic boundary conditions. The predicted size of the over-tip leakage vortex was smaller than had been measured, resulting in a large underprediction in the magnitude of the downstream area-averaged aerodynamic loss. The magnitudes of the predicted blade tip Nusselt number distribution were similar to those produced by the engine-representative CFD simulations and lower than that measured experimentally. Differences in the shape of the Nusselt number distribution were observed in the vicinity of regions of separated and reattaching flow, but other salient features were replicated in the computational data. The squealet tip has been shown to be a promising, viable unshrouded blade tip design with an aerodynamic performance similar to the double squealer tip but is more amenable to film cooling. It is significantly lighter than a winglet tip and incurs a reduced thermal load. The squealet tip design can now be developed into a blade tip geometry for use in real engines to provide an alternative to shrouded turbine blades and current unshrouded blade tip designs. A commercial CFD solver, Fluent 6.3, was shown to capture blade tip heat transfer and over-tip leakage flow sufficiently well to be a useful design guide. However, the sensitivity of the flow structure (and hence, heat transfer) in the forward part of the blade tip cavity suggests that physical testing cannot be eliminated from the design process entirely.

Due to the requirements for high blade loading, modern turbo‐machine blades operate very close to the stall regime. This can lead to flow separation with periodic shedding of vortices, which could lead to self induced oscillations or stall flutter of the blades. Previous studies on stall flutter have focused on flows at high Reynolds number (Re ~ 106). The Reynolds numbers for fans/propellers of Micro Aerial Vehicles (MAVs), high altitude turbofans and small wind turbines are substantially lower (Re < 105). Aerodynamic characteristics of flows at such low Re is significantly different from those at high Re, due in part to the early separation of the flow and possible formation of laminar separation bubbles (LSB). The present study is targeted towards study of stall flutter in a cascade of blades at low Re. We experimentally study stall flutter of a cascade of symmetric NACA 0012 blades at low Reynolds number (Re ~ 30, 000) through forced sinusoidal pitching of the blades about mean angles of incidences close to stall. The experimental arrangement permits variations of the inter‐blade phase (σ) in addition to the oscillation frequency (f) and amplitude; the inter‐blade phase angle (σ) being the phase difference between the motions of adjacent blades in the cascade. The unsteady moments on the central blade in the cascade are directly measured, and used to calculate the energy transfer from the flow to the blade. This energy transfer is used to predict the propensity of the blades to undergo self‐induced oscillations or stall flutter. Experiments are also conducted on an isolated blade in addition to the cascade. A variety of parameters can influence stall flutter in a cascade, namely the oscillation frequency (f), the mean angle of incidence, and the inter‐blade phase angle (σ). The measurements show that there exists a range of reduced frequencies, k (=πfc/U, c being the chord length of the blade and U being the free stream velocity), where the energy transfer from the flow to the blade is positive, which indicates that the flow can excite the blade. Above and below this range, the energy transfer is negative indicating that blade excitations, if any, will get damped. This range of excitation is found to depend upon the mean angle of incidence, with shifts towards higher values of k as the mean angle of incidence increases. An important parameter for cascades, which is absent in the isolated blade case is the inter‐blade phase angle (σ). An excitation regime is observed only for σ values between ‐450 and 900, with the value of excitation being maximum for σ of 900. Time traces of the measured moment were found to be non‐sinusoidal in the excitation regime, whereas they appear to be sinusoidal in the damping regime. Stall flutter in a cascade has differences when compared with an isolated blade. For the cascade, the maximum value of excitation (positive energy transfer) is found to be an order of magnitude lower compared to the isolated blade case. Further, for similar values of mean incidence angle, the range of excitation is at lower reduced frequencies for a cascade when compared with an isolated blade. A comparison with un‐stalled or classical flutter in a cascade at high Re, shows that the inter‐blade phase angle is a major factor governing flutter in both cases. Some differences are observed as well, which appear to be due to stalled flow and low Re.

Approved for public release, distribution is unlimited / The flow around second-generation controlled-diffusion blades in cascade at stall was examined experimentally through the use of a two-component laser-Doppler velocimeter. Blade surface pressure measurements were also preformed at mid span on the blades at various Reynolds numbers. Flow visualization techniques were used to observe and record the flow on the surface of the blade. A correlation between the experimental results and computational fluid dynamic predictions was attempted in order to determine the exact nature of the flow as the blades approached stall, to further assist in the development of advanced blade design. The blade surface pressure measurements showed that the mid-span section of the blade was at a lower loading than previously measured at a smaller inlet flow angle. This indicated that the blade section was at stall. The flow visualization highlighted the extent of the three-dimensional flow over the blades. The LDV measurements documented the mid-span boundary layer and wake profiles. / Ensign, United States Navy

La conception des groupes moto-ventilateurs au sein de Valeo Systèmes Thermiques et la prédiction de leurs performances aérauliques reposent majoritairement sur les méthodes de développement virtuel, i.e. la conception assistée par ordinateur et la simulation numérique de la mécanique des fluides. Dans ce cadre, le présent travail propose une méthodologie de prédiction et de minimisation de la composante tonale du bruit d'un groupe moto-ventilateur. L'approche adoptée est hybride et dissocie la génération et la propagation du bruit. La propagation en champ libre est calculée avec une formulation intégrale de l'analogie de Ffowcs-Williams et Hawkings. Dans un premier temps, les termes-sources à la surface du rotor et du stator sont calculés par une simulation numérique instationnaire. La compacité de la pale ainsi que l'influence du maillage acoustique sur la prédiction sont ensuite investiguées. Finalement, les résultats sont comparés aux mesures expérimentales. Dans un deuxième temps, les sources acoustiques à la surface du stator sont calculées avec le modèle de Sears enrichi avec des données extraites d'une simulation stationnaire du rotor complet. Avant de procéder à la prédiction acoustique, l'influence du modèle de turbulence sur les résultats finaux est évaluée à travers une comparaison entre LES et RANS pour l'écoulement autour de profils extrudés. Enfin, la problématique de minimisation du bruit tonal est traitée en tant que problème d'optimisation où la géométrie d'une aube est paramétrée et où la recherche de l'optimum est conduite par un algorithme génétique. Cette optimisation a permis de concevoir un stator moins bruyant et adapté à l'écoulement en aval du rotor étudié. / The design of fan systems at Valeo Thermal Systems and the prediction of their aerodynamic performances rely mainly on virtual development methods, i.e. computer-aided-design and computational fluid dynamics. Within this context, this dissertation develops a methodology for predicting and minimizing the tonal noise of a fan system. The hybrid approach is used, thus separating noise generation and propagation. The free-field propagation is computed via an integral formulation of the Ffowcs-Williams and Hawkings analogy. In a first step, the source terms located at the surfaces of the rotor and the stator are extracted from an unsteady numerical simulation. The compactness of the blade and the influence of the acoustic mesh on the prediction are then investigated. Finally, the computational results are compared to the experimental ones. In a second step, the acoustic sources at the surface of the stator are computed with Sears' model. Its inputs are extracted from a steady simulation of the whole rotor. Before proceeding to the acoustic prediction, the influence of the turbulence model on the final results is assessed via a comparison between LES and RANS simulations of the flow around airfoils. Finally, minimizing tonal noise is formulated as an optimization problem. The shape of a stator-blade is parametrized and the optimization is conducted with a genetic algorithm. The resulting stator is less noisy and adapted to the flow downstream of the studied rotor.

L’objectif de cette thèse est de faire progresser la modélisation de la transition de couche limite sur des aubes de turbines basse-pression fortement chargées. Cette modélisation repose sur l’utilisation du modèle de transition de Menter et Langtry utilisé pour des calculs RANS dans le code elsA. Une fois les limitations du modèle de transition clairement identifiées par une étude sur la mise en données des calculs, nous avons entrepris de modifier ce dernier. Pour cela, un processus d’optimisation a été développé afin de permettre la recalibration des fonctions de corrélation internes au modèle de transition. Cette nouvelle version du modèle nous permet d’obtenir des gains sur la modélisation d’environ 20 % sur les cas T106C du VKI en capturant mieux la transition au sein du bulbe de décollement. Ces précédents calculs correspondent à des cas idéaux, où l’on peut considérer les surfaces comme étant lisses. Cependant, nous avons aussi un besoin de se rapprocher de surfaces plus réalistes pour lesquelles les rugosités peuvent avoir un impact sur l’écoule- ment. En effet, les rugosités de surface peuvent notamment avoir un effet sur la transition. En particulier, si les rugosités entraînent le déclenchement de la transition en amont du point de décollement laminaire théorique en surface lisse, ce décollement sera supprimé. Vu nos efforts pour améliorer la prévision de la transition par bulbe de décollement par le modèle γ-Rθt, il parait intéressant que celui-ci puisse prendre en compte l’état des surfaces. Pour cela, nous avons implanté une méthode de prévision de la transition sur surfaces rugueuses développée par Stripf et al. au sein du modèle γ-Rθt. Enfin, l’utilisation du modèle de transition γ-Rθt a été étendue au modèle de turbulence k-l de Smith. / The goal of this thesis is to enhance laminar-turbulent transition modeling on high-lift low- pressure turbine blades. The presented transition modeling method relies on the Menter and Langtry transition model used in a RANS framework in the elsA solver. Once the model’s limits were clearly identified through a parametric study, we moved on to modification of the model. To do so, an optimization method was developed that allows recalibration of the model’s inner correlation functions. This new version of the model allows us to obtain modeling gains of about 20% on the VKI T106C cases through better capture of the separation-induced transition process. These previous computations correspond to ideal cases, for which surfaces may be considered as being smooth. However, we also have the need to consider more realistic surfaces for which roughness may influence the flow. Indeed, among those effects, is the potential influence of surface roughness on transition. In particular, if surface roughness induces transition up-stream of the smooth separation point, the separation bubble will be suppressed. Considering our efforts on modeling separation-induced transition with the γ-Rθt model, it seemed natural to add roughness-induced transition modeling capacities to it. To do so, we implemented in the γ-Rθt model a method developed by Stripf et al. to take into account surface roughness. Finally, the use of the γ-Rθt transition model was extended to the k-l of Smith tur- bulence model. Indeed, this turbulence model is widely used in turbomachinery. In order that our works on transition modeling over turbine blades be more widely usable, we have completed this thesis by proposing an evolution of the transition model so that it may be used alongside the k-l model.