Dans cette étude nous considérons l'interaction entre une onde de choc et une couche limite turbulente sur un écoulement transsonique sur une bosse d'un point de vue expérimentale et théorique.Des mesures expérimentales ont permis de montrer que l'interaction est caractérisée par la coexistence de deux fréquences caractéristiques distinctes, mais l'origine des oscillations est controversée. Des simulations numériques permettent une description de l'écoulement moyen, mais ne sont pas capables de reproduire le comportement instable de l'interaction. Nous proposons d'abord une étude de stabilité globale: une décomposition en valeurs propres de l'opérateur de Navier-Stokes linéarisé indique que l'interaction est un phénomène stable, et la dynamique de l'écoulement ne peut pas être décrite par un mode global instable.Nous considérons ensuite une approche linéarisée, où la réceptivité de l'écoulement à un forçage externe est analysée à travers une décomposition en valeurs singulières du Résolvant global. Cette nouvelle approche est proposée afin d'expliquer le processus de sélection de fréquence dans cet écoulement, et montre que l'interaction filtre et amplifie le bruit résiduel existant.La même approche est enfin appliquée sur un cas d'écoulement transsonique autour d'un profil d'aile, qui peut présenter des oscillations périodiques de l'onde de choc. La décomposition en valeurs propres de opérateur de Navier-Stokes linéarisé est capable de décrire la dynamique du choc, tandis que la décomposition en valeurs singulières du Résolvant global peut indiquer la présence des instabilité convectives. / A transonic interaction between a shock wave and a turbulent boundary layer is experimentally and theoretically investigated. The configuration is a channel flow over a bump, where a shock wave causes the separation of the boundary layer and a recirculating bubble is observed downstream of the shock foot.The mean flow is experimentally investigated by means of PIV, then different techniques allows to identify the main unsteadiness of this shock-wave/boundary-layer interaction. As recognised in similar configurations, the flow presents two distinguished characteristic frequencies, whose origins are still unknown.Numerical simulations are performed solving RANS equations. Results are in good agreement with the experimental mean flow, but the approach fails to predict the unsteady. The solution of RANS equations is then considered as a base flow, and a global stability analysis is performed. Eigenvalue decomposition of the Navier-Stokes operator indicates that the interaction is stable, and the dynamics cannot be described by unstable global modes.A linearised approach based on a singular-value decomposition of the Resolvent is then proposed: the noise-amplifier behaviour of the flow is highlighted by the linearised approach. Medium-frequency perturbations are shown to be the most amplified in the mixing layer, whilst the shock wave behaves as a low-pass filter.The same approach is then applied to a transonic flow over a profile, where the flow can present high-amplitude shock oscillations. The stability analysis can describe both the buffet phenomenon when an unstable mode is present, and the convective instabilities responsible of medium-frequency unsteadiness.

The thesis gives overview of complex aerodynamic optimization approach applied to business-jet aircraft wing design. Response surface method (RSM) potential was explored particularly. The efficiency of RSM approach for CFD based aerodynamic optimization was demonstrated. Basic structural requirements were successfully integrated to optimization – real multidisciplinary problem was solved. Some methods for evaluation of forces distribution along wingspan were explored. Thesis was solving within the frame of 6th EU FP integrated project CESAR.

The purpose of this study was to develop a robust fluid-structure-interaction (FSI) technology that can accurately model non-linear flutter responses for sub- and transonic fluid flow. The Euler equation set governs the fluid domain, which was spatially discretised by a vertex-centred edge-based finite volume method. A dual-timestepping method was employed for the purpose of temporal discretisation. Three upwind schemes were compared in terms of accuracy, efficiency and robustness, viz. Roe, HLLC (Harten-Lax-Van Leer with contact) and AUSM+-up Advection Up-stream Splitting Method). For this purpose, a second order unstructured MUSCL (Monotonic Upstream-centred Scheme for Conservation Laws) scheme, with van Albada limiter, was employed. The non-linear solid domain was resolved by a quadratic modal reduced order model (ROM), which was compared to a semi-analytical and linear modal ROM. The ROM equations were solved by a fourth order Runge-Kutta method. The fluid and solid were strongly coupled in a partitioned fashion with the information being passed at solver sub-iteration level. The developed FSI technology was verified and validated by applying it to test cases found in literature. It was demonstrated that accurate results may be obtained, with the HLLC upwind scheme offering the best balance between accuracy and robustness. Further, the quadratic ROM offered significantly improved accuracy when compared to the linear method. / Dissertation (MEng)--University of Pretoria, 2011. / Mechanical and Aeronautical Engineering / unrestricted

Vehicles with internal combustion (IC) engines fueled by hydrocarbon compounds have been used for more than 100 years for ground transportation. During the years and in particular in the last decade, the environmental aspects of IC engines have become a major political and research topic. Following this interest, the emissions of pollutants such as NOx, CO2 and unburned hydrocarbons (UHC) from IC engines have been reduced considerably. Yet, there is still a clear need and possibility to improve engine efficiency while further reducing emissions of pollutants. The maximum efficiency of IC engines used in passenger cars is no more than $40\%$ and considerably less than that under part load conditions. One way to improve engine efficiency is to utilize the energy of the exhaust gases to turbocharge the engine. While turbocharging is by no means a new concept, its design and integration into the gas exchange system has been of low priority in the power train design process. One expects that the rapidly increasing interest in efficient passenger car engines would mean that the use of turbo technology will become more widespread. The flow in the IC-engine intake manifold determines the flow in the cylinder prior and during the combustion. Similarly, the flow in the exhaust manifold determines the flow into the turbine, and thereby the efficiency of the turbocharging system. In order to reduce NOx emissions, exhaust gas recirculation (EGR) is used. As this process transport exhaust gases into the cylinder, its efficiency is dependent on the gas exchange system in general. The losses in the gas exchange system are also an issue related to engine efficiency. These aspects have been addressed up to now rather superficially. One has been interested in global aspects (e.g. pressure drop, turbine efficiency) under steady state conditions.In this thesis, we focus on the flow in the exhaust port and close to the valve. Since the flow in the port can be transonic, we study first the numerical modeling of such a flow in a more simple geometry, namely a bump placed in a wind tunnel. Large-Eddy Simulations of internal transonic flow have been carried out. The results show that transonic flow in general is very sensitive to small disturbances in the boundary conditions. Flow in the wind tunnel case is always highly unsteady in the transonic flow regime with self excited shock oscillations and associated with that also unsteady boundary-layer separation. To investigate sensitivity to periodic disturbances the outlet pressure in the wind tunnel case  was varied periodically at rather low amplitude. These low amplitude oscillations caused hysteretic behavior in the mean shock position and appearance of shocks of widely different patterns. The study of a model exhaust port shows that at realistic pressure ratios, the flow is transonic in the exhaust port. Furthermore, two pairs of vortex structures are created downstream of the valve plate by the wake behind the valve stem and by inertial forces and the pressure gradient in the port. These structures dissipate rather quickly. The impact of these structures and the choking effect caused by the shock on realistic IC engine performance remains to be studied in the future. / CICERO

Endwall contouring has been widely studied as an effective measure to improve aerodynamic performance by reducing secondary flow strength. The effects of endwall contouring with discrete holes and an upstream purge slot for a high turning (127") airfoil passage under transonic conditions are investigated. The total pressure loss and secondary flow field were measured for two endwall geometries. The non-axisymmetric endwall was developed through an optimization study [1] to minimize secondary losses and is compared to a baseline planar endwall. The blade inlet span increased by 13 degrees with respect to the inlet in order to match engine representative inlet/exit Mach number loading in a HP turbine.  The experiments were performed in a quasi-2D linear cascade with measurements at design exit Mach number 0.88 and incidence angle. Four cases were analyzed for each endwall -- the effect of slot presence (with/without coolant) and the effect of discrete holes (with/without coolant) without slot injection. The coolant to mainstream mass flow ratio was set at 1.0% and 0.25% for upstream purge slot and discrete holes, respectively.  Aerodynamic loss coefficient is calculated with the measured exit total pressure at 0.1 Cax downstream of the blade trailing edge. CFD studies were conducted in compliment. The aero-optimized endwall yielded lower losses than baseline without the presence of the slot. However, in presence of the slot, losses increased due to formation of additional vortices. For both endwall geometries, results reveal that the slot has increased losses, while the addition of coolant further influences secondary flow development. / Master of Science

We examine steady transonic flows of Bethe-Zel\'dovich-Thompson (BZT) fluids over thin turbine blades or airfoils. BZT fluids are ordinary fluids having a region of negative fundamental derivative over a finite range of pressures and temperatures in the single phase regime. We derive the transonic small disturbance equation (TSDE) capable of capturing the qualitative behavior of BZT fluids. The shock jump conditions, and shock existence conditions consistent with the derived TSDE are presented. The flux function is seen to be quartic in the pressure or density perturbation rather than the quadratic (convex) flux function of the perfect gas theory. We show how this nonconvex flux function can be used to predict and explain the complex flows possible in transonic BZT fluids. Numerical solutions using a successive line relaxation (SLR) scheme are presented. New results of interest include shock-splitting, collisions between expansion and compression shocks, the prediction and observation of two compressive bow shocks in supersonic flows, and the observation of as many as three normal stern shocks following an oblique trailing edge shock. / Master of Science

The measurement of wind tunnel flow quality for the transonic flow regime has been plagued by the inability to interpret complex unsteady flow field information obtained in the free stream. Traditionally hot wire anemometry and fluctuating pressure techniques have been used to quantify the unsteady characteristics of a wind tunnel. This research focuses on the application of these devices to the transonic flow regime. Utilizing hot wire anemometry, one can decompose the unsteady flow field with a three sensor technique, to obtain fluctuations associated with the velocity, density, and total temperature. Implementing thermodynamic and kinematic equations, new methods for expanding the measured velocity, density, and total temperature fluctuations to obtain additional fluctuations are investigated. The derived static pressure fluctuations are compared to the static pressure fluctuations obtained with a conventional fluctuating static pressure probe. The results of this comparison are good, which implies that the individual velocity, density, and total temperature components are time accurate. In the process of obtaining a high quality fluctuating flow field information, it was necessary to evaluate the calibration of the hot wire sensors. A direct calibration approach was compared to a conventional non-dimensional technique. These two calibration techniques should have resulted in the same hot wire sensitivities. There were significant differences in the hot wire sensitivities as obtained from the two approaches. The direct approach was determined to have less errors due to the added heat transfer information required of the indirect approach. Both calibration techniques demonstrated that the velocity and density sensitivities were in general not equal. This suggests that the velocity and density information cannot be combined to form a mass flow. A comparison of several hot wire techniques was included to highlight the errors obtained when assuming that these sensitivities are the same. An evaluation of the free stream flow quality associated with a Laminar Flow Control experiment was carried out in the Langley Research Center 8-Foot Transonic Pressure Tunnel (8' TPT). The facility was modified with turbulence manipulators and a liner that provided a flow field around a yawed super-critical airfoil that is conducive to transition research. These devices are evaluated to determine the sources of disturbances associated with the LFC experiment. / Ph. D.

<p>Improvements in modern jet engines to obtain lower specific fuel consumption and emission levels have pushed towards the optimization of every component in these machines. One current approach in minimizing engine weight is using high lift blading to reduce blade count and therefore the overall turbine weight while maintaining similar levels of efficiency. The goal of this work was to evaluate the performance of six different high pressure turbine geometries to assess the viability of using high lift designs. To do so, this study begins through the development of measurement techniques to aid the aerodynamic characterization of the flow field in a turbine environment: five-hole probe calibration maps were produced, hot wire sensors were calibrated, and an oil visualization technique was developed. The airfoils in this work were tested in a rainbow annular cascade in a stationary rig and a flow conditioning gauze was used to replicate the rotor inlet profile in the blade reference frame. The next step taken was characterizing and validating the flow field behind the gauze. Subsequently, the six turbine geometries were tested at two transonic Mach number conditions and at different Reynolds numbers. Continuous total pressure and flow angle traverses were carried out to evaluate the pressure and kinetic loss coefficients across the turbines; blade loading measurements and oil visualizations were used to quantify the power extraction of the different geometries and to study the flow structures within a turbine passage. </p>

No description available.

No description available.