Au cours des dernières années, le transport aérien de passagers connaît un développement sans cesse croissant et continue ainsi d’accroire sa contribution aux émissions mondiale de CO2. Par conséquent, un effort commun entre les avionneurs est fait pour diminuer les émissions de CO2 et de polluants. Pour encourager cet effort, les réglementations deviennent de plus en plus drastiques en terme d'émissions et de polluants tels que le CO2, les NOx mais aussi le bruit. Ces nouvelles limitations sont à la fois définies à court et moyen-long termes pour inciter les motoristes à travailler sur les technologies de plus en plus efficientes.Pour concevoir des moteurs toujours plus performants tout en respectant ces réglementations à court terme, les motoristes travaillent sur l'optimisation de leurs technologies conventionnelles, en améliorant des leviers bien identifiés comme l'augmentation du taux de compression. Cependant, cette optimisation des turbomachines actuelles a déjà atteint un niveau de maturité très élevé. Il semble ainsi difficile de continuer indéfiniment leurs optimisations. Par conséquent, pour atteindre les objectifs à moyen-long terme, les motoristes sont dès aujourd'hui en train d'étudier des nouveaux systèmes propulsifs avancés comme les chambres de Combustion à Volume Constant (CVC) qui peuvent accroître le rendement thermique. Contrairement aux chambres de combustion traditionnelles, qui fonctionnent à flux continu, les chambres CVC opèrent de façon cyclique afin de créer un volume constant pendant la phase de combustion et libérer les gaz chauds dans les étages de turbines.Pendant cette thèse, une approche numérique permettant d'évaluer ce type de chambres est développée. Tout l'enjeu est de pouvoir étudier des chambre de combustion intégrant des parties mobiles, qui permettent de créer le volume constant dédié à la combustion tout en évitant les fuites à travers ces systèmes mobiles lors de l'élévation de la pression dans la chambre. Cette modélisation doit aussi prédire correctement les phases transitoires comme l'admission des gaz frais, qui pilote la phase de combustion. Cette étude utilise des objets immergés pour modéliser les parties mobiles. Les objectifs de cette thèse sont de rendre ces objets immergés imperméables et adapter la méthode aux différents modèles utilisés pour étudier les milieux réactifs tels que le modèle de combustion ECFM-LES ou encore l'injection liquide Lagrangienne utilisée pour résoudre l'injection du fuel.Dans cette étude, une nouvelle formulation est développée puis testée sur différents cas tests de plus en plus représentatifs des chambres CVC. Cette approche numérique est ensuite évaluée sur une chambre réel étudiée expérimentalement au laboratoire PPRIRME de Poitiers. Dans cette dernière étude, deux cas non réactifs permettent de comparer les évolutions de pression à deux endroits dans la dispositif expérimental, ainsi que les champs de vitesse au sein de la chambre de combustion, aux simulations réalisées. Pour ces cas complexes, l'utilisation des objets immergés permet de prédire les résultats expérimentaux à un coût attractif.Un des cas non réactif est ensuite carburé et allumé pour confronter l'évolution pression et les champs de vitesse dans la chambre de combustion des résultats numériques obtenus aux mesures expérimentales. L'approche numérique développée a permis d’enrichir les données expérimentales, d'analyser les variabilités cycle-à-cycle rencontrées au banc et d'identifier les leviers qui permettraient d'optimiser ce type d’architecture. / Over the past few years, aircrafts have become a common means of transport, thus continuously increasing their contribution to global CO2 emissions. Consequently, there is a common effort between aircraft manufacturers to reduce CO2 and pollutant emissions. To encourage this effort, regulations are becoming more and more stringent on the emissions and pollutants like CO2, NOx and noise. These regulations are both defined in the short and medium-long terms to urge aircraft manufacturers to work on more and more efficient technologies.In order to design more efficient engines while respecting the short term objectives, engine manufacturers are working on the improvement of conventional architectures by using well-known levers like the increase of the Overall Pressure Ratio (OPR). However, the optimization of the present turbomachinery has already reached a high level of maturity and it seems difficult to continuously enhance their performances. Consequently, to reach the medium-long term objectives, engine manufacturers are working on new advanced propulsion systems such as the Constant Volume Combustion (CVC) chambers, which can increase the thermal efficiency of the system. Contrary to present turbomachinery which are burning fresh gases continuously, CVC chambers operate cyclically so as to create the constant vessel dedicated to the combustion phase and to expand the burnt gases into turbine stages.In this PhD thesis, a numerical approach is developed to allow the evaluation of these kind of combustors. The challenge is to be able to evaluate CVC chambers by taking into account the moving parts which create the constant volume and avoid mass leakages through these moving parts during the increase of the combustion chamber pressure when the combustion occurs. This approach also has to correctly predict unsteady phases like the intake, which directly controls the combustion process.These moving parts are modeled with a Lagrangian Immersed Boundary (LIB) method .The main goals of this thesis is to make the LIB as airtight as possible and to render this approach compatible with the different models which are adapted to analyse reactive flows such as the ECFM-LES combustion model or Lagrangian liquid injection, used for fuel sprays. In this study, a new formulation is developed and tested on several test cases from very simple ones to cases more representative of CVC chambers.Then, this approach is evaluated on a real chamber experimentally analysed in PPRIME laboratory in Poitiers. Two non-reactive operating points are used to compare the experimental pressure at two positions in the apparatus and the experimental velocity fields in the combustion chamber with the numerical results. In this complex configuration, the LIB method allows the prediction of the experimental results with a low CPU cost. As in the experiment, one non-reactive case is carburized and ignited to compare the measured pressure and the velocity fields in the combustion chamber with the simulations. The proposed numerical approach allows the data enhancement of the experiment and then the analysis of the cycle-to-cycle variability encountered during the experimental measurements. Last but not least, this method enables the identification of the different levers that could decrease the variability and then could improve operability of this type of combustors.

Face à l’explosion du trafic aérien attendue ces prochaines années, l’impact de l’aviation civile sur l’environnement est un enjeu majeur. Les instances environnementales internationales comme l’ACARE (Conseil Consultatif pour la Recherche Aéronautique en Europe), en partenariat avec les grands groupes aéronautiques internationaux, ont fixé des objectifs drastiques pour préserver l’environnement : une réduction des émissions de CO2de 75 %et une réduction de 90 % des rejets d’oxydes d’azote dans l’atmosphère sont attendues d’ici 2050 par rapport aux avions fabriqués au début du 21èmesiècle. Les turbomachines actuelles possédant un degré de maturité très élevé ne permettront pas d’atteindre ces objectifs. Les motoristes cherchent donc à étudier de nouveaux concepts en rupture technologique pour les horizons 2050, comme les moteurs à détonation, ou encore les moteurs de type combustion à volume constant. Actuellement, les phénomènes physiques associés à la combustion du kérosène dans ce type de moteur sont encore mal documentés. L’objectif de cette thèse est donc de contribuer à l’amélioration de la connaissance et de la compréhension de ces phénomènes physiques.Au cours de cette étude, les flammes de pré-mélanges de kérosène et d’air sont étudiées expérimentalement grâce à des diagnostics optiques (strioscopie,PIV) et métrologiques. Le processus de combustion est notamment étudié dans des conditions thermodynamiques semblables à celles rencontrées dans un moteur aéronautique. La phase de propagation est dans un premier temps analysée dans des conditions laminaires et adiabatiques à travers la détermination de la vitesse fondamentale de flamme non-étirée, grandeur qui pilote le processus de combustion. Puis la sensibilité du front de flamme à l’étirement et la formation des instabilités de combustion sont dans un second temps examinées. Enfin, la phase d’allumage des pré-mélanges de kérosène et d’air dans des conditions aérodynamiques critiques est elle aussi traitée.Un second point abordé au cours de cette étude concerne la reproduction d’un kérosène réel par un substitut constitué d’un nombre d’espèces limité pour simplifier les problématiques industrielles et les études amont. En effet, la composition d’un kérosène commercial est complexe et variée et l’utilisation d’un représentant permet de modéliser numériquement le phénomène de combustion plus facilement. La pertinence de quelques surrogates plus ou moins représentatifs, formulés dans la littérature et élaborés au cours de différents travaux est notamment traitée dans cette étude en comparant les résultats obtenus avec ceux d’un kérosène commercial. De plus, la modélisation de ces kérosènes de substitution par un schéma cinétique valide estégalement analysée.Ce travail prend place dans le cadre de la chaire industrielle CAPA sur la combustion alternative pour la propulsion aérobie financée par SAFRANTech, MBDA et l’ANR. / With air traffic expected to soar in the next few years, the impact of civil aviation on the environment is a major issue. International environmental organizations such as ACARE (the Advisory Council for Aeronautical Research and Innovation in Europe), in partnership with the main international aeronautical groups, have set drastic objectives to preserve the environment: a reduction of 75 % of CO2emissions and a reduction of 90 % of nitrogen oxide emissions into the atmosphere are sought by 2050, with reference to aircraft produced at the beginning of the 21st century. Current turboshaft engines have a very high degree of maturity and may not achieve these objectives. Engineers are therefore aiming to study new concepts that will become technological breakthroughs at the 2050 horizon, such as detonation engines or constant volume combustion engines. Currently, the physical phenomena associated with the combustion of kerosene in those kinds of engines are still poorly documented. The objective of this PhD thesis is to contribute to the improvement of the knowledge and understanding of these physical phenomena. In this work, premixed flames of kerosene and air are experimentally studied with optical diagnostics (Schlieren, PIV) and metrology techniques. The combustion process is here studied in thermodynamic conditions similar to those encountered in an aeronautical engine. First, the propagation phaseis analyzed in laminar and adiabatic conditions through the determination of the unstretched laminar burning velocity, which drives the combustion process. Then, in a second stage, the sensitivity of the flame front to stretch and the formation of combustion instabilities are examined. Finally, the ignition phase of premixed flames of kerosene and air under critical aerodynamic conditions is also investigated. A second issue tackled in this work is the reproduction of a real kerosene by a surrogate made up of a limited number of species, to simplify industrial problems and initial studies. Indeed, the composition of a commercial kerosene is complex and can vary, and the use of a surrogate allows an easier numerical simulation of the combustion process. The relevance of some more or less representative surrogates, formulated in the literature and elaborated all through different studies, is also studied in this thesis, by comparing the results obtained with those of a commercial kerosene. In addition, the modelling of those surrogates by a valid chemical kinetic mechanism is also analyzed. This research was conducted within the CAPA industrial Chair project dedicated to innovative combustion modes for air-breathing propulsion, financially supported by SAFRAN Tech, MBDA and France’s ANR national research agency.

Un des principaux enjeux de l'industrie aéronautique est la recherche du moteur au meilleur rendement possible, pour satisfaire des contraintes économiques, techniques et environnementales. Les turbomachines bénéficient d'un constant perfectionnement depuis plus de 60 ans, et cette technologie semble avoir atteint un plateau. Une rupture technologique est aujourd'hui nécessaire, comme la combustion à volume constant (CVC). Le gain attendu est suffisant pour tenter de remplacer les systèmes actuels où la combustion se fait à pression constante. La combustion à isovolume fait appel à des mécanismes encore rarement maitrisés dans le contexte aéronautique. Sa compréhension passe par des expérimentations et des modèles théoriques et numériques. L’objectif de cette thèse est de développer une théorie et un outil de simulation LES (Large Eddy Simulation) appliqué au cas du concept ‘thermoréacteur’. Ainsi, la première étape a consisté à mettre en place un outil de simulation 0D traduisant l’évolution d’un cycle moteur de type CVC (Combustion à Volume Constant). Certains modèles utilisés dans cet outil 0D sont basés sur des corrélations expérimentales. D'autres présentent des paramètres à déterminer à partir de simulations numériques. La simulation 3D d’un système de type CVC est envisageable aujourd’hui grâce aux progrès récents des méthodes LES. Ainsi, des simulations du thermoréacteur ont pu être réalisées, et confrontées aux résultats expérimentaux obtenus au laboratoire Pprime sur trois points de fonctionnement. Les variabilités cycle à cycle observées expérimentalement ont été analysées dans les calculs LES. Les vitesses importantes au niveau de l'allumage et le taux de résidus du cycle précédent semblent être les principaux facteurs à l'origine de ces variations cycle à cycle. / A major challenge for the aircraft industry is to improve engine efficiency and to reduce pollutant emissions for economic, technical and environmental reasons. Aeronautical gas turbines have enjoyed a constant improvement for more than 60 years. This technology seems to have reached such efficiency levels that a technological breakthrough is necessary. Constant Volume Combustion (CVC) offers significant gain in consumption and could replace classical constant pressure combustion technologies, currently used in aeronautical engines. Mechanisms involved in isovolume combustion are not accurately controlled in the context of aeronautical chambers. Experimental, theoretical and numerical studies should provide a better understanding of CVC devices. The objective of this thesis is to develop simulation tools to study the thermoreacteur concept. First, a zero-dimensional (0D) simulation tool is developed to describe the evolution of a CVC cycle. Models based on experimental correlations are used to build the 0D tool. Parameters have to be determined from numerical simulations. Today, the 3D simulation of a CVC system is possible thanks to the recent progress of the LES (Large Eddy Simulation) methods developed at CERFACS. Simulations of the thermoreacteur concept have been carried out, and compared to experimental results obtained at the Pprime laboratory. Three operating points have been calculated. The main conclusion is the existence of significant cyclic variations which are observed in the experiment and analyzed in the LES: the local flow velocity at spark timing and the level of residuals gases are the major factors leading to cyclic variations.

Indiana University-Purdue University Indianapolis (IUPUI) / A chemically reactive turbulent traversing hot-jet issued from a pre-chamber to a relatively long combustion chamber is experimentally investigated. The long combustion chamber represents a single channel of a wave rotor constant-volume combustor. The issued jet ignites the fuel-air mixture in the combustion chamber. Fuel-air mixtures are prepared with different hydrocarbon fuels of different reactivity, namely, methane, propane, methane-hydrogen blend, methane-propane blend and methane-argon blend. The jet acts as a rapid, distributed and moving source of ignition, traversing across one end of the long combustion chamber entrance, induces complex flow structures such as a train of counter rotating vortices that enhance turbulent mixing. In general, a stationary hot-jet ignition process lack these structures due to absence of the traversing motion. The ignition delay of the fuels and fuel blends are measured in order to obtain insights about constant-volume pressure-gain combustion process initiated by a moving source of ignition and also to glean useful data about design and operation of a wave rotor combustor. Reactive hot-jets are useful to ignite fuel-air mixtures in internal combustion engines and novel wave rotor combustors. A reactive hot-jet or puff of gas issued from a suitably designed pre-chamber can act as rapid, distributed and less polluting ignition source in internal combustion engines. Each cylinder of the engine is provided with its own pre-chamber. A wave rotor combustor has an array of circumferentially arranged channels on a rotating drum. Each channel acts as a constant-volume combustor and produces high pressure combustion products. Implementation of hot-jet igniter in a wave rotor combustor offers utilization of available high temperature and high pressure reactive combustion products residing in each of the wave rotor channels as a distributed source of ignition for other channels, thus requiring no special pre-chamber in ultimate implementation. Such reactive products or partially combusted and radical-laden gases can be issued from one or more channels to ignite the fuel-air mixture residing in another channel. Due to the rotation of the rotor channels, the issued hot-jet would have relative motion with respect to one end of the channels and traverse across it. This thesis aims to investigate the effects of jet traverse time experimentally on ignition delay along with other important factors that affect the hot-jet ignition process such as fuel reactivity, fuel-air mixture preparation quality and stratification and equivalence ratio. In this study, the traversing motion of the hot-jet at one end of the main combustion chamber is implemented by keeping the main combustion chamber stationary and rotating a pre-chamber at speeds of 400 RPM, 800 RPM and 1200 RPM. The rotational speeds correspond to jet traverse times of 16.9 ms, 8.4 ms and 5.6 ms respectively. The fuel-air mixture inside the channel is at room temperature and pressure initially and its equivalence ratio is varied from 0.4 to 1.3. The cylindrical pre-chamber is initially filled with a 50%-50% methane-hydrogen blend fuel and air mixture at room pressure and temperature and at an equivalence ratio of 1.1. These conditions were chosen based on prior evidence of ignition rapidity with the jet properties. The hot-jet is issued by rupturing a thin diaphragm isolating the chambers. Using high frequency dynamic pressure transducer pressure histories, the diaphragm rupture moment and onset of ignition is measured. Pressure traces from two transducers are employed to measure the initial rupture shock speed and ignition delay. Schlieren images recorded by a high speed camera are used to identify ignition moment and validate the measured ignition delay times. Ignition delay is defined as time interval from the rupture moment to onset of ignition of fuel-air mixture in the main combustion chamber. The ignition system is designed to produce diaphragm rupture at almost exactly the moment when jet traverse begins. Ignition delay times are measured for methane, propane, methane-hydrogen blends, methane-propane blend and methane-argon blend. The equivalence ratio of the fuel-air mixtures varied from 0.4 to 1.3 in steps of 0.1 for stationary-hot jet ignition experiments and in steps of 0.3 for traversing hot-jet ignition experiments. Hot-jet ignition delay of fuel-air mixtures, for both stationary hot-jet ignition process and traversing hot-jet ignition process, generally increased with increasing equivalence ratio. For stationary hot-jet ignition delay, the minimum ignition delay occurs between Ф = 0.4 to Ф = 0.6 for the tested fuel-air mixtures. Both stationary and traversing hot-jet ignition delay depended on fuel reactivity. In particular, the shortest ignition delay times were observed for a fuel blend containing hydrogen. Among pure fuels propane exhibited slightly shorter ignition delay times, on average, compared to pure methane fuel. The addition of argon to pure methane, intended to control fuel density and buoyancy, increased the ignition delay. The traversing hot-jet ignition delay generally increased with increasing jet traverse times. To explain the variations in the measured hot-jet ignition delay and investigate qualitatively the effect of buoyancy on flame propagation and mixture stratification, the fuel-air mixture inside the main combustion chamber was ignited using a spark plug to generate a propagating laminar flame. The laminar flame propagated within the flammable regions of the channel in ways that sensitively reveal variations in local fuel-air mixture equivalence ratio. Flame luminosity images from a high speed camera and schlieren images revealed the fuel-air mixture being highly stratified depending on the density difference between the fuel and air and provided mixing time (0 s, 10s ,30s for most fuels). The lack of buoyancy-driven spreading caused the fuel to remain in the vicinity of the fuel injector resulting in significant longitudinal stratification of the fuel-air mixture. Lighter fuels stratified to the top of the chambers and heavier fuel stratified to the bottom of the chamber. Increasing the mixing time, which is defined as the time interval from end of fuel injection into the chamber to the triggering of the spark plug, improved the buoyancy-driven spreading and extended the flammable region as evidenced by the schlieren and flame luminosity images. The maximum pressure developed in the combustor for the three ignition processes, namely, stationary hot-jet ignition, traversing hot-jet ignition and spark ignition process in laminar flame propagation experiments were compared. Stationary hot-jet ignition process generally exhibited the highest pressure being developed in the chamber. Variations in heat loss, fuel-air mixture leakage and mass addition mechanisms reduced the maximum pressure for spark ignition and traversing hot-jet ignition process.

Indiana University-Purdue University Indianapolis (IUPUI) / Investigations of a constant-volume combustor ignited by a penetrating transient jet (a puff) of hot reactive gas have been conducted in order to provide vital data for designing wave rotor combustors. In a wave rotor combustor, a cylindrical drum with an array of channels arranged around the axis spins at a high rpm to generate high-temperature and high-pressure product gas. The hot-gas jet ignition method has been employed to initiate combustion in the channels. This study aims at experimentally investigating the ignition delay time of a premixed combustible mixture in a rectangular, constant-volume chamber, representing one channel of the wave rotor drum. The ignition process may be influenced by the multiple factors: the equivalence ratio, temperature, and the composition of the fuel mixture, the temperature and composition of the jet gas, and the peak mass flow rate of the jet (which depends on diaphragm rupture pressure). In this study, the main mixture is at room temperature. The jet composition and temperature are determined by its source in a pre-chamber with a hydrogen-methane mixture with an equivalent ratio of 1.1, and a fuel mixture ratio of 50:50 (CH4:H2 by volume). The rupture pressure of a diaphragm in the pre-chamber, which is related to the mass flow rate and temperature of the hot jet, can be controlled by varying the number of indentations in the diaphragm. The main chamber composition is varied, with the use of four equivalence ratios (1.0, 0.8, 0.6, and 0.4) and two fuel mixture ratios (50:50, and 30:70 of CH4:H2 by volume). The sudden start of the jet upon rupture of the diaphragm causes a shock wave that precedes the jet and travels along the channel and back after reflection. The shock strength has an important role in fast ignition since the pressure and the temperature are increased after the shock. The reflected shock pressure was examined in order to check the variation of the shock strength. However, it is revealed that the shock strength becomes attenuated compared with the theoretical pressure of the reflected shock. The gap between theoretical and measured pressures increases with the increase of the Mach number of the initial shock. Ignition delay times are obtained using pressure records from two dynamic pressure transducers installed on the main chamber, as well as high-speed videography using flame incandescence and Schileren imaging. The ignition delay time is defined in this research as the time interval from the diaphragm rupture moment to the ignition moment of the air/fuel mixture in the main chamber. Previous researchers used the averaged ignition delay time because the diaphragm rupture moment is elusive considering the structure of the chamber. In this research, the diaphragm rupture moment is estimated based on the initial shock speed and the longitudinal length of the main chamber, and validated with the high-speed video images such that the error between the estimation time and the measured time is within 0.5%. Ignition delay times decrease with an increase in the amount of hydrogen in the fuel mixture, the amount of mass of the hot-jet gases from the pre-chamber, and with a decrease in the equivalence ratio. A Schlieren system has been established to visualize the characteristics of the shock wave, and the flame front. Schlieren photography shows the density gradient of a subject with sharp contrast, including steep density gradients, such as the flame edge and the shock wave. The flame propagation, gas oscillation, and the shock wave speed are measured using the Schlieren system. An image processing code using MATLAB has been developed for measuring the flame front movement from Schlieren images. The trend of the maximum pressure in the main chamber with respect to the equivalence ratio and the fuel mixture ratio describes that the equivalence ratio 0.8 shows the highest maximum pressure, and the fuel ratio 50:50 condition reveals lower maximum pressure in the main chamber than the 30:70 condition. After the combustion occurs, the frequency of the pressure oscillation by the traversing pressure wave increases compared to the frequency before ignition, showing a similar trend with the maximum pressure in the chamber. The frequency is the fastest at the equivalence ratio of 0.8, and the slowest at a ratio of 0.4. The fuel ratio 30:70 cases show slightly faster frequencies than 50:50 cases. Two different combustion behaviors, fast and slow combustion, are observed, and respective characteristics are discussed. The frequency of the flame front oscillation well matches with that of the pressure oscillation, and it seems that the pressure waves drive the flame fronts considering the pressure oscillation frequency is somewhat faster. Lastly, a feedback mechanism between the shock and the flame is suggested to explain the fast combustion in a constant volume chamber with the shock-flame interactions.

Indiana University-Purdue University Indianapolis (IUPUI) / This study contributes to a better working knowledge of the equipment being used in a well-established combustion lab. In particular, several constant-volume combustion properties (e.g., time ignition delay, flame propagation, and more) are examined to deduce any buoyancy effects between fuel and air mixtures and to develop a method aimed at minimizing such effects. This study was conducted on an apparatus designed to model the phenomena occurring within a single channel of a wave rotor combustor, which consists of a rotating cylindrical pre-chamber and a fixed rectangular main combustion chamber. Pressure sensors monitor the internal pressures within the both chambers at all times, and two slow-motion videography techniques visually capture combustion phenomena occurring within the main chamber. A new recirculation pump system has been implemented to mitigate stratification within the chamber and produce more precise, reliable results. The apparatus was used in several types of experiments that involved the combustion of various hydrocarbon fuels in the main chamber, including methane, 50%-50% methane-hydrogen, hydrogen, propane, and 46.4%-56.3% methane-argon. Additionally, combustion products created in the pre-chamber from a 1.1 equivalence ratio reaction between 50%-50% methane-hydrogen and air were utilized in the issuing pre-chamber jet for all hot jet ignition tests. In the first set of experiments, a spark plug ignition source was used to study how combustion events travel through the main chamber after different mixing methods were utilized – specifically no mixing, diffusive mixing, and pump circulation mixing. The study reaffirmed that stratification between fuel-air mixtures occurs in the main chamber through the presence of asymmetrical flame front propagation. Allowing time for mixing, however, resulted in more symmetric flame fronts, broader pressure peaks, and reduced combustion time in the channel. While 30 seconds of diffusion helped, it was found that 30 seconds of pumping (at a rate of 30 pumps per 10 seconds) was the most effective method at reducing stratification effects in the system. Next, stationary hot jet ignition experiments were conducted to compare the time between jet injection and main chamber combustion and the speed of the resulting shockwaves between cases with no mixing and 30 seconds of pump mixing. Results continued to show an improvement with the pump cases; ignition delay times were typically shorter, and shock speeds stayed around the same, if not increased slightly. These properties are vital when studying and developing wave rotor combustors, and therefore, reducing stratification (specifically by means of a recirculation system) should be considered a crucial step in laboratory models such as this one. Lastly, experiments between a fueled main chamber and rotating pre-chamber helped evaluate the leakage rate of the traversing hot jet ignition experimental setup paired with the new pump system. In its current form, major leaks are inevitable when attempting traversing jet experiments, especially with the pump’s suction action drawing sudden large plumes of outside air into the main chamber. To minimize leaks, gaps between the pre-chamber and main chamber should be reduced, and the contact surface between the two chambers should be more evenly distributed. Also, the pump system should only be operated as long as needed to evenly distribute the fuel-air mixture, which approximately happens when the main chamber’s total volume has been circulated through the system one time. Therefore, a new pump system with half of the original system’s volume was developed in order to decrease the pumping time and lower the risk of leaks.

When studying pressure-gain combustion and wave rotor combustors, it is vital that any experimental model accurately reflect the real world conditions/applications being studied; this not only confirms previous computational and analytical work, but also provides new insights into how these concepts and devices work in real life. However, mixture stratification can have a noticeable effect on multiple combustion properties, including flame propagation, pressure, ignition time delay, and more, and this is especially true in constant-volume combustion chambers. Because it is beneficial to model wave rotor systems using constant-volume combustors such as what is employed in the IUPUI Combustion and Propulsion Research Laboratory, these stratification effects much be taken into account and reduced if possible. This study sought to find an effective method to reduce stratification in a rectangular constant-volume combustion chamber by means of manual recirculation pump. Spark-ignited flames were first produced in the chamber itself and studied using schlieren and color videography techniques as well as quantitative pressure histories. After determining the pump's effectiveness in reducing stratification, it was next employed when a hot jet of combustion products from a separate combustion chamber was used as an ignition source instead of the spark plug - a process typically employed in real wave rotor combustors. Lastly, the pump was used to study the leakage from the system for future test cases in order to offer further recommendations on how to effectively use the recirculation system. This process found that key properties significant to wave rotor development, such as time ignition delay, were affected by these stratification effects in past studies that did not account for this detail. As such, the pump has been permanently incorporated into the wave rotor model, as stratification is a vital. Additionally, significant fuel leakage is possible during rotational pre-chamber cases, and this should be address before proceeding with such experiments in the future. To combat this, the pump system has been reduced in volume, and suggestions have been provided on how to better seal the main rectangular chamber in the future.

La propagation de flammes turbulentes dans des milieux réactifs inhomogènes concerne un grand nombre d’applications pratiques, y compris celles qui reposent sur des cycles de combustion à volume constant. Les hétérogénéités de composition (richesse, température,dilution par des gaz brûlés, etc.) sont issues de plusieurs facteurs distincts tels que la dispersion du spray de gouttelettes de combustible et son évaporation, la topologie de l’écoulement ainsi que la présence éventuelle de gaz brûlés résiduels issus du cycle précédent. La structure des flammes partiellement prémélangées qui en résultent est significativement plus complexe que celles des flammes plus classiques de diffusion ou de prémélange. L’objectif de ce travail de thèse est donc de contribuer à l’amélioration de leur connaissance, en s’appuyant sur la génération et l’analyse de base de données de simulations numériques directes ou DNS (Direct Numerical Simulation). Celles-ci sont conduites avec le code de calcul Asphodele qui est basé sur l’approximation de faible nombre de Mach. Le combustible de référence retenu est l’iso-octane.La base de données est structurée suivant cinq paramètres qui permettent de caractériser l’écoulement turbulent ainsi que l’hétérogénéité de composition du milieu réactif. Dans un premier temps, des configurations bidimensionnelles ont été considérées en raison du coût élevé induit par la description détaillée de la cinétique chimique. L’étude des ces différents cas de calcul a permis de mettre en lumière plusieurs mécanismes fondamentaux de propagation dans les milieux hétérogènes en composition. Une réduction significative des coûts de calcula pu ensuite être obtenue grâce au développement d’un modèle chimique simplifié optimisé.Son utilisation a permis d’étendre les analyses à de / The propagation of turbulent flames in non-homogeneous reactive mixtures of reactants concerns a large number of practical applications, including those based on constant volume combustion cycles. The composition heterogeneities (equivalence ratio, temperature, dilution by burnt gases, etc.) result from several distinct factors such as the dispersion of the spray of fuel droplets and its evaporation, the flow field topology as well as the possible presence of residual burnt gases issued from the previous cycle. The resulting partially premixed flames structure is significantly more complex than the one of more conventional diffusion or premixed flames.The aim of this thesis work is therefore to contribute to the improvement of their understanding, by proceeding to the generation and analysis of a new set of direct numerical simulations (DNS) databases. The present computations are performed with the low-Mach number DNS solver Asphodele. The database is structured according to five parameters that characterize the turbulent flow as well as the composition heterogeneity of the reactive mixture. First, because of the high numerical costs induced by the detailed description of chemical kinetics, two-dimensional configurations were considered. The study of these various simulations highlights several fundamental mechanisms of flame propagation in heterogeneous mixtures. Then, a significant computational cost saving has been achieved through the development of an optimized simplified chemistry model. The use of the latter allowed to overcome the major bottleneck of high CPU costs related to chemical kinetics description and thus to extend the analysis to three-dimensional configurations. Some of the conclusions obtained previously were reinforced.

Les turbomachines actuelles ont atteint un niveau de maturité technique très élevé. De nouvelles architectures reposant sur des cycles thermodynamiques basés sur une combustion à gain de pression, comme la combustion à volume constant (CVC), ont le potentiel d’augmenter leur efficacité. Dans cette étude,une solution qui repose sur l’intégration dans une turbomachine de chambres de combustion à volume constant sans piston (CVCSP) est considérée. Les objectifs de ces travaux de thèse sont doubles : dans un premier temps de développer et de caractériser extensivement un nouveau dispositif (CV2) dédié à la Combustion à volume constant sans piston sur un cas de référence et, dans un second temps, de proposer à travers plusieurs études, une analyse de l’influence de l’aérodynamique et de la dilution sur les processus d’allumage et, plus généralement de combustion. Le dispositif CV2 permet la combustion aérobie en allumage commandé d’un mélange de propane ou de n-décane, injecté directement dans la chambre. Un point de référence est caractérisé en détail via : des mesures de champs de vitesse par PIV, de chimiluminescence pendant la combustion, une analyse 0D développée dans cette étude. La caractérisation détaillée de ce point de référence montre que le dispositif CV2 reproduit correctement une combustion à volume constant turbulente dans un mélange faiblement hétérogène en température et stratifié en composition, et ce sur un nombre de cycles permettant d’établir une convergence statistique raisonnable. Ces diagnostics et analyses sont employés dans 2 cas d’études pour caractériser successivement : l’influence de l’aérodynamique, via une variation de l’instant d’allumage, l’influence des gaz brûlés résiduels sur la combustion en allumage commandé et la stabilité cyclique, via une variation de la pression d’échappement.Dans un fonctionnement sans balayage, on montre que cette variabilité cyclique est liée au premier ordre à la variation de la dilution en gaz brûlé résiduel du mélange et à la vitesse locale. On montre notamment que, pour un mélange donné, il existe une corrélation statistique entre une vitesse statistique limite et la probabilité d’allumage moyenne. Pour représenter l’effet de pression dans un plénum en amont d’une turbine, on réalise une étude paramétrique sur la pression d’échappement. La dilution résultante, croissant avec la pression d’échappement, diminue la vitesse fondamentale de flamme et ralentit donc la combustion. Les niveaux de températures des gaz brûlés résiduels résultent des échanges de chaleur qui ont lieu sur toute la durée du cycle, de l’allumage du cycle N à celui du cycle N+1 suivant. Des extrapolations sur des cycles à température de paroi plus élevée et à échappement plus court montrent que l’adiabaticité du cycle est améliorée (de 20 %) et que l’effet de dilution en température est alors favorable à une vitesse de flamme turbulente qui est alors plus élevée. Un phénomène d’allumage par gaz brûlé résiduel est observé sur certains cycles de combustion. Ce phénomène est caractérisé dans des conditions favorables, i.e. faible richesse (0.66), allumage tardif et cycle plus court. Lors d’un allumage par gaz brûlés résiduels, un noyau de flamme se développe dans les zones présentant des gaz brûlés résiduels chauds et à basse vitesse autour du jet d’admission et se propage ensuite au reste du mélange identiquement à celui qui serait généré par allumage commandé.Ce travail prend place dans le cadre de la chaire industrielle CAPA sur la combustion alternative pour la propulsion aérobie financée par SAFRAN Tech, MBDA et l’ANR. / Current turbomachines have reached a very high level of technical maturity. Thermodynamic cycles based on pressure-gain combustion, such as constant volume combustion (CVC), feature a clear potential for efficiency improvement. The present study considers the integration in a turbomachine of piston-lessCVC chambers. The thesis work is twofold. First, a new experimental setup (CV2) dedicated to cyclic piston-less CVC is developed and thoroughly characterized on a reference operating point. Second, the influence of the aerodynamics and dilution on the processes of ignition and, in a larger sense, on combustion is discussed through dedicated studies. The CV2 device allows for the spark-ignited air-breathing combustion of a mixture of either propane orn-decane, directly injected into the chamber. A reference condition is characterized in details using: PIV velocity field measurements, chemiluminescence of combustion and a 0D modeling of the device. This detailed characterization evidenced that the CV2 combustion chamber successfully replicates, on a number of cycles allowing a reasonable statistical convergence, a turbulent deflagrative constant-volume combustion in a mixture stratified in composition. Those diagnostics and analyses are applied to 2 cases of study to characterize successively : the influence of the aerodynamics, through a variation of the ignition timing, the influence of the residual burnt gases on spark-ignited combustion and the cyclic stability, through a variation of the exhaust backpressure.Operating the device without scavenging of the combustion chamber, we show that the cyclic variability correlates strongly with both the variation of residual burnt gases dilution and the local velocity. Particularly, we show that for a given mixture, a correlation exists between a statistical velocity limit and the average probability of ignition. The effect of a plenum backpressure upstream of a turbine, downstream of the combustion chamber, is simulated by varying the exhaust system backpressure. The resulting dilution, which increases with the exhaust backpressure, diminishes the fundamental flame velocity of the mixture and slows down the combustion. The residual burnt gases temperature results from the integrated heat exchanges that happen during the total cycle duration starting from the end of combustion of cycle N, to the ignition of cycle N+1. Enhanced cycles, with an increased wall temperature and reduced exhaust duration, are extrapolated by 0D analysis. Those cycles evidence a reduction of the cumulated heat exchanges of up to 20 %. The resulting dilutionis more favorable to higher turbulent flame velocity thus to shorter combustion duration. A phenomenon of ignition induced by the residual burnt gases is observed on certain combustion cycles. This phenomenon is characterized in favorable conditions, i.e. fuel-lean equivalence ratio (0.66), late ignition and shortcycles. During an ignition by residual burnt gases, a flame kernel is ignited in areas where the still hot residuals burnt gases meet fresh gases in low-velocity areas around the intake jet. The ignition kernel then propagates to the rest of the mixture in a similar manner as if it was spark-ignited.This work is part of the CAPA Chair research program on Alternative Combustion modes for Air-breathing Propulsion supported by SAFRAN Tech, MBDAFrance and ANR (French National Research Agency).

No description available.