The present thesis is an experimental study of some aspects of unsteady two dimensional boundary layers subject to adverse pressure gradient. An adverse pressure gradient usually leads to boundary layer separation or an instability which may result in transition to turbulence. Unsteady boundary layer separation is not yet fully understood and there is no specific criterion proposed in literature for its occurrence. The details of separation depend on the Reynolds number, the geometry of the body (streamlined or bluff) and the type of imposed unsteady motion (impulsive, oscillatory etc.). Similarly there are many unknowns with respect to instability and transition in unsteady boundary layers, especially those having a streamwise variation. For unsteady flows it is useful to break up the pressure gradient term in the unsteady boundary layer equation into two components:(Formula) is the velocity at the edge of the boundary layer. The first term of the right hand side of this equation may be called the temporal component (Πt) which signifies acceleration or deceleration in time of the free stream and the second term is the spatial component (Πx) which represents the spatial or convective acceleration of the free stream. Many of the studies on instability in unsteady flows found in literature are carried out in straight tubes or channels, where the Πx term is absent. However, in many cases, especially in biological systems both terms are present. An example is the unsteady flow over the moving body of a fish. To study the effects of Πt and Πx on unsteady separation and instability we have built an unsteady water tunnel where the two components can be systematically varied. The flow is created by a controlled motion of a piston. By a suitable combination of the geometry of the model and the piston motion, different types of separation bubbles may be generated. In our studies the piston motion follows a trapezoidal variation: constant acceleration from rest, followed by constant velocity and then deceleration to zero velocity. We have chosen two geometries. One is a bluff body and thus has a high value of Πx and other is a small angle diffuser with a divergence angle 6.2° and thus having a small value of Πx. Upstream and downstream of the diffuser are long lengths of constant cross section. We have performed experiments with the above mentioned geometries placed in the tunnel test section. Flow is visualized using the laser induced fluorescence technique by injecting a thin layer of fluorescein dye on the test wall. Numerical simulations have been done using the software FLUENT. Boundary layer parameters like boundary layer, displacement and momentum thicknesses are calculated from the simulations and used to analyze the experimental results. For the flow in the diffuser, quasi-steady stability analysis of the instantaneous velocity profiles gives a general idea of stability behavior of the flow. Two types of experiments have been done with the bluff body. One is the unsteady boundary layer separation and the formation of the initial vortex for a flow that is uniformly accelerated from rest. We have found some scalings for the formation time (tv) of the separation vortex. The second type of experiment was to study the vortex shedding from the separating shear layer after the boundary layer has fully separated. At high enough Reynolds number shear layer vortices are seen to shed from the separation bubble. The Strouhal number based on the momentum thickness and the velocity at the edge of the boundary layer just upstream of the separation point is found to vary between 0.004 and 0.008. This value is close to the Strouhal number value of 0.0068 found in laminar separation bubbles on a flat plate. The second part of the study concerns with the evolution of the flow in the small angle diffuser with a mild variation of the spatial component of the pressure gradient. From the experimental visualizations we have found that the ratio of Πx and Πt at the start of the deceleration phase of the piston motion is an important parameter that determines the type of instability. This value of Πx/Πt is controlled by controlling the piston deceleration: a large deceleration gives a low Πx/Πt value and a low deceleration gives a large Πx/Πt value. Three types of instabilities have been observed in our experiments. In Type I, the first vortex forms at the maximum pressure gradient point (MPGP) and which grows disproportionately with time. However, instability vortices are seen later at other locations around the MPGP. In type II an array of vortices over a certain length are observed; the vortices grow with time. In Type III, which we observe for low decelerations, we observe initial vortices only in the diffuser section in the deceleration phase of the piston motion. Type III instability is similar to the one observed in dynamic stall experiments. In all cases the instability is very localized - it occurs only over some length of the boundary layer. Transition to turbulence, which is also localized, is observed at higher Reynolds numbers. The non-dimensional time for vortex formation is not very different from that found in straight channel experiments. Quasi-steady linear stability analyses for the boundary layer at the MPGP both for the top and the bottom walls show that the flow is absolutely unstable for some cases. In summary, the thesis looks at in a unified way the separation and instability of unsteady boundary layers with reverse flow. It is hoped that the results will be useful in predicting and understanding onset of separation and instability in practically occurring unsteady flows.

Dans le contexte économique et environnemental actuel, la prochaine génération de moteurs d’avion devra offrir opérabilité, compacité et hauts rendements. Les compresseurs demeurent une des pièces critiques de ces moteurs, et leur conception un challenge. À débit réduit, leur plage de fonctionnement est contrainte par la limite de pompage, phénomène hautement instable et dangereux. À ce jour, peu d’études expérimentales sur un compresseur en situation de pompage ont été réalisées, étant donné le danger inhérent pour les installations. Dans ce cadre, la simulation numérique peut apporter des informations sur le développement des instabilités aérodynamiques et aider à la prévision de la limite de pompage. L’objectif du travail présenté dans cette thèse est de mettre en place une méthode afin de simuler numériquement l’entrée en pompage et un cycle complet de l’instabilité avec le code elsA. Le cas test retenu est le compresseur de recherche axial multi-étage CREATE dessiné par Snecma, et étudié expérimentalement par le LMFA. Des études antérieures ont montré le rôle joué par les volumes entourant le compresseur ; l’originalité de cette étude réside donc dans l’inclusion des volumes du banc d’essai dans la simulation du compresseur. Une des difficultés inhérentes à la simulation de ces instabilités est leur temps caractéristique, qui représente plus d’une centaine de rotations de la machine. Le calcul a donc nécessité le recours à une approche massivement parallèle ; environ un million d’heures CPU ont été utilisées pour décrire le cycle. Enfin, compte tenu du retournement de l’écoulement dans le compresseur, les conditions aux limites ont été modifiées pour pouvoir s’adapter aux changements de sens de l’écoulement. La simulation a permis de décrire l’entrée en pompage et un cycle complet de l’instabilité. La comparaison avec les données expérimentales montre que les caractéristiques du cycle sont correctement prédites (phénomènes physiques précurseurs de l’instabilité, durée du cycle..). En parallèle, une étude acoustique a été menée afin de mettre en évidence les modes propres du banc d’essai. L’analyse de ces résultats a notamment montré le rôle de l’acoustique dans le déclenchement du pompage. Les différentes phases du cycle de pompage sont ensuite étudiées, et caractérisées (déclenchement, débit inversé, récupération et recompression). Ce travail a généré une base de données qui permet de mieux comprendre les instabilités qui se développent dans ce type de machine. À terme, ces résultats pourront être utilisés pour élaborer et valider des modélisations du phénomène de pompage moins coûteuses, pouvant intervenir dans un cycle de conception. / In order to deal with the current economical and environmental context, the next engine generation will need to offer great operability, compactness and high efficiency. In aircraft engines, the compressor remains one of the critical components, and its design is still a challenging task. At low massflow rate, their operability is bounded by the surge limit, surge being a highly unstable and dangerous phenomenon. Today, few experimental studies on compressor surge are available because of the inherent threat to the facility. In that context, numerical simulation can bring about information on the onset of aerodynamic instabilities and help to predict the surge limit. The work presented in this PhD thesis aims at setting up a method to perform the numerical simulation of surge inception and of an entire cycle of the instability with the CFD code elsA. The chosen test case is the axial multistage research compressor CREATE designed and built by Snecma, and experimentally studied at LMFA. Previous studies have pointed out the role of the volumes adjacent to the compressor ; the originality of this work is thus the inclusion of the volumes of the test-rig in the simulation of the compressor. One of the difficulties inherent to the simulation of those instabilities is their characteristic time of at least one hundred revolutions of the machine. Hence the computation has required a massively parallel approach and about one million CPU hours. Finally, given that the flow reverses during a surge cycle, the boundary conditions have been modified to be able to cope with the flow inversions. The simulation was able to capture surge inception and the entire cycle of the instability. The comparison with the experimental data showed that the main patterns of the cycle are correctly predicted (precursor phenomena of surge, duration of the cycle...). In the meantime, an acoustic study has been performed in order to isolate the eigenmodes of the test-rig. The analysis of the results pointed out the role of acoustic phenomena in surge inception. The different phases of the cycle are then studied and characterized (surge inception, reversed-flow phase, recovery and repressurization). This work has incremented a database that allows a better understanding of the instabilities that develop in this kind of machine. From now on, those results may help to elaborate and validate cheaper models of the surge phenomenon to be used in the design process.

Ce travail est le produit d’une collaboration entre le LMFA (Laboratoire de Mécanique des Fluides et d’Acoustique, École Centrale de Lyon – France), Snecma et le Cerfacs. Il vise à étudier l’écoulement se développant dans le compresseur haute-vitesse axial de 3.5 étages CREATE (Compresseur de Recherche pour l’Etude des effets Aérodynamique et TEchnologique – vitesse de rotation : 11543 tr/min, vitesse en tête de rotor 1 : 313 m/s), conçu et construit par Snecma et étudié au LMFA sur un banc d’essai de 2 MW. Pour étudier l’écoulement, des mesures stationnaires de pression et température, de la vélocimétrie laser et des mesures rapides de pression statique et totale ont été utilisées. L’analyse se concentre sur deux aspects principaux : l’étude de l’écoulement aux points de fonctionnement stables, avec un intérêt tout particulier pour les interactions rotor-stator, et l’étude des instabilités apparaissant dans la machine à faibles débits.La description de l’écoulement instationnaire aux points stables est faite par le biais de mesures de pression statique en parois, de pression totale et de vitesse, mais également de température totale, entropie et angle d’incidence du fluide. Il est montré que la complexité et l’instationnarité de l’écoulement dans un compresseur multiétagé augmente fortement à l’arrière de la machine à cause des interactions entre les roues fixes et mobiles. Ainsi, une méthode d’analyse modale basée sur la décomposition de Tyler et Sofrin a été développée pour analyser ces interactions. Elle est d’abord appliquée aux mesures de pression afin d’extraire les contributions de chaque roue. Il est ainsi montré que les interactions complexes de pression dans CREATE peuvent être réduites à trois principaux types d’interactions. La méthode de décomposition est enfin appliquée au champ d’entropie dans toute la machine extrait de calculs CFD URANS réalisés par le Cerfacs, afin d’évaluer l’impact des interactions sur les performances de CREATE en terme de génération de pertes.La dernière partie de ce travail est dédié à l’analyse des instabilités apparaissant dans CREATE à faible débit. Il est montré que des ondes de pression tournantes apparaissent aux points stables et augmentent en amplitude à mesure qu’on se rapproche de la ligne de pompage, jusqu’à atteindre une taille critique induisant l’apparition d’une cellule de décollement tournant sur toute la hauteur de veine. Cette cellule entraîne la machine en pompage en seulement quelques tours. L’étude de ces ondes de pression, et la compréhension de leur véritable nature sont réalisées grâce à l’application d’un modèle analytique aux mesures expérimentales. Une description précise du déclenchement et du cycle du pompage est également faite grâce aux mesures de pression statique au dessus des rotors. Un système de contrôle anti-pompage développé au laboratoire et basé sur la détection de l’amplitude des ondes de pression est finalement décrit. / The present work is a result of collaboration between the LMFA (Laboratoire de Mécanique des Fluides et d’Acoustique, Ecole Centrale de Lyon – France), Snecma and the Cerfacs. It aims at studying the flow in the 3.5-stages high-speed axial compressor CREATE (Compresseur de Recherche pour l’Etude des effets Aérodynamique et TEchnologique - rotation speed: 11543 RPM, Rotor 1 tip speed: 313 m/s), designed and built by Snecma and investigated at LMFA on a 2-MW test rig. Steady measurements, as well as laser velocimetry, fast-response wall static and total pressure measurements have been used to experimentally investigate the flow. The analysis focuses on two main aspects: the study of the flow at stable operating points, with a special interest on the rotor-stator interactions, and the study of the instabilities arising in the machine at low mass flow rates.The description of the unsteady flow field at stable operating points is done through measurements of wall-static pressure, total pressure and velocity, but also total temperature, entropy and angle of the fluid. It is shown that the complexity and unsteadiness of the flow in a multistage compressor strongly increases in the rear part of the machine, because of the interactions between steady and rotating rows. Therefore, a modal analysis method developed at LMFA and based on the decomposition of Tyler and Sofrin is presented to analyze these interactions. It is first applied to the pressure measurements, in order to extract the contributions of each row. It shows that all the complex pressure interactions in CREATE can be reduced to three main types of interactions. The decomposition method is then applied to the entropy field extracted from URANS CFD calculations performed by the Cerfacs, in order to evaluate the impact of the interactions on the performance of the machine in term of production of losses.The last part of this work is devoted to the analysis of the instabilities arising in CREATE at low mass flows. It shows that rotating pressure waves appear at stable operating points, and increase in amplitude when going towards the surge line, until reaching a critical size provoking the onset a full span stall cell bringing the machine to surge within a few rotor revolutions. The study of these pressure waves, and the understanding of their true nature is achieved through the experimental results and the use of some analytical models. A precise description of the surge transient through wall-static pressure measurements above the rotors is also provided, as well as a description of a complete surge cycle. An anti-surge control system based on the detection of the amplitude of the pressure waves is finally proposed.

The development of numerical methods for the incompressible Navier-Stokes equations received much attention in the past 50 years. Finite element methods emerged given their robustness and reliability. In our work, we choose the P2-P1 finite element for space approximation which gives 2nd-order accuracy for velocity and 1st-order accuracy for pressure. Our research focuses on the development of several high-order semi-implicit time-stepping methods to compute unsteady flows. The methods investigated include backward difference formulae (SBDF) and defect correction strategy (DC). Using the defect correction strategy, we investigate two variants, the first one being based on high-order artificial compressibility and bootstrapping strategy proposed by Guermond and Minev (GM) and the other being a combination of GM methods with sequential regularization method (GM-SRM). Both GM and GM-SRM methods avoid solving saddle point problems as for SBDF and DC methods. This approach reduces the complexity of the linear systems at the expense that many smaller linear systems need to be solved. Next, we proposed several numerical improvements in terms of better approximations of the nonlinear advection term and high-order initialization for all methods. To further minimize the complexity of the resulting linear systems, we developed several new variants of grad-div splitting algorithms besides the one studied by Guermond and Minev. Splitting algorithm allows us to handle larger flow problems. We showed that our new methods are capable of reproducing flow characteristics (e.g., lift and drag parameters and Strouhal numbers) published in the literature for 2D lid-driven cavity and 2D flow around the cylinder. SBDF methods with grad-div stabilization terms are found to be very stable, accurate and efficient when computing flows with high Reynolds numbers. Lastly, we showcased the robustness of our methods to carry 3D computations.

Ces travaux de recherche ont pour but d’évaluer la méthode dite de la "Simulation auxEchelles Adaptées" (SAS pour Scale-Adaptive Simulation). Cette approche coïncide avec uneapproche RANS classique dans les zones pariétales attachées et adapte le niveau de viscositéturbulente dans les zones décollées pour y permettre une résolution partielle des structures turbulentes.Dans une première partie, une analyse théorique du modèle SAS original a été menéeet a permis de développer une correction visant à favoriser l’adaptation du niveau de viscositéturbulente dans les zones sièges d’instabilités de type Kelvin-Helmholtz. Le modèle ainsi corrigéest nommé SAS-αL. Les modèles SAS et SAS-αL ont été implantés dans le code de calculNavier-Stokes elsA de l’ONERA. À l’issue de cette étape, trois cas académiques d’écoulementsturbulents instationnaires, cylindre à grand nombre de Reynolds, marche descendante et cavitétranssonique, ont été simulés grâce aux trois modèles de turbulence SST, SAS et SAS-αL. Outreune comparaison aux bases de données expérimentales disponibles, une attention particulièrea été portée à l’influence de paramètres numériques tels que des schémas numériques d’ordreélevé. Enfin, afin d’étudier la viabilité de l’approche SAS dans un contexte industriel, les troismodèles de turbulence ont été testés sur une configuration issue de l’industrie aéronautique etcorrespondant à la sortie d’air chaud d’un système de dégivrage des nacelles d’avion. La comparaisondes prévisions obtenues avec les modèles SST, SAS et SAS-αL aux données expérimentalesobtenues à l’ONERA a permis de montrer un gain de précision grâce à l’emploi de l’approcheSAS et ce pour un coût de calcul compatible avec un cycle de conception industrielle. / This research work is meant to assess an upgraded URANS approach, namely the Scale-Adaptive Simulation (SAS). This method is similar to a conventional RANS approach (namelythe SSTmodel) in attached areas and is able to adapt the eddy-viscosity level in detached areas toensure the resolution, at least partially, of the turbulent structures. In a first part of this researchwork, an improvement of the SAS approach is suggestedto allowa better sensitivity of themodelto instabilities such as Kelvin-Helmholtz ones. This "improved" model is referred to as SAS-αLmodel. Both SAS and SAS-αL models were implemented in the ONERA Navier-Stokes solverelsA and both of themaswell as the SSTmodelwere tested on academic test cases : a cylinder in acrossflowat a high Reynolds number, a backward-facing step flowcorresponding to theDriver&Seegmiller experiment and the transonic flow over the M219 cavity experimentally investigatedby de Henshaw. The influence of the numerical parameters was deeply investigated and particularattention was paid to the high-order space-discretization schemes effects. The reliabilityof the SAS approach in an industrial framework was assessed on an aeronautic configurationnamely a nacelle de-icing device. Comparisons between the threemodels (SST, SAS and SAS-αL)and an experimental database available at ONERA - The French Aerospace Lab have shown thebetter accuracy of the SAS approach as well as the high potential of the SAS-αL model.

This thesis deals with computing the pressure distribution around a swimming slender fish and the thrust generated by its flapping motion. The body of the fish is modeled as a missile like slender body to which a tail is attached that is modeled as a swept wing. The tail is attached to the tip of the slender body and maintains its slope with it. The motion for the swimming fish is prescribed. The fluid flow is modeled as an unsteady potential flow problem with the flow around the slender body modeled as flow over an array of cylinders of varying radii and the flow over the swept wing modeled using the vortex panel method. The pressure distribution is computed using the unsteady Bernoulli equation. The overall thrust & drag for different parameters are studied and compared

<p align="justify">This work deals with the development of numerical methods for compressible flow simulation with application to the interaction of fluid flows and structural bodies.</p><p><p><p align="justify">First, we develop numerical methods based on multidimensional upwind residual distribution (RD) schemes. Theoretical results for the stability and accuracy of the methods are given. Then, the RD schemes for unsteady problems are extended for computations on moving meshes. As a second approach, cell centered and vertex centered finite volume (FV) schemes are considered. The RD schemes are compared to FV schemes by means of the 1D modified equation and by the comparison of the numerical results for scalar problems and system of Euler equations. We present a number of two and three dimensional steady and unsteady test cases, illustrating properties of the numerical methods. The results are compared with the theoretical solution and experimental data.</p><p><p><p align="justify">In the second part, a numerical method for fluid-structure interaction problems is developed. The problem is divided into three distinct sub-problems: Computational Fluid Dynamics, Computational Solid Mechanics and the problem of fluid mesh movement. The problem of Computational Solid Mechanics is formulated as a system of partial differential equations for an anisotropic elastic continuum and solved by the finite element method. The mesh movement is determined using the pseudo-elastic continuum approach and solved again by the finite element method. The coupling of the problems is achieved by a simple sub-iterative approach. Capabilities of the methods are demonstrated on computations of 2D supersonic panel flutter and 3D transonic flutter of the AGARD 445.6 wing. In the first case, the results are compared with the theoretical solution and the numerical computations given in the references. In the second case the comparison with experimental data is presented.</p> / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished