Fossil fuel reserves are projected to be decreasing, and emission regulations are becoming more stringent due to increasing atmospheric pollution. Alternative fuels for power generation in industrial gas turbines are thus required able to meet the above demands. Examples of such fuels are synthetic gas, blast furnace gas and coke oven gas. A common characteristic of these fuels is that they are multi-component fuels, whose composition varies greatly depending on their production process. This implies that their combustion characteristics will also vary significantly. Thus, accurate and yet flexible enough combustion sub-models are required for such fuels, which are used during the design stage, to ensure optimum performance during practical operating conditions. Most combustion sub-model development and validation is based on Direct Numerical Simulation (DNS) studies. DNS however is computationally expensive. This, has so far limited DNS to single-component fuels such as methane and hydrogen. Furthermore, the majority of DNS conducted to date used one-step chemistry in 3D, and skeletal chemistry in 2D only. The need for 3D DNS using skeletal chemistry is thus apparent. In this study, an accurate reduced chemical mechanism suitable for multi-component fuel-air combustion is developed from a skeletal mechanism. Three-dimensional DNS of a freely propagating turbulent premixed flame is then conducted using both mechanisms to shed some light into the flame structure and turbulence-scalar interaction of such multi-component fuel flames. It is found that for the multi-component fuel flame heat is released over a wider temperature range contrary to a methane flame. This, results from the presence of individual species reactions zones which do not all overlap. The performance of the reduced mechanism is also validated using the DNS data. Results suggest it to be a good substitute of the skeletal mechanism, resulting in significant time and memory savings. The flame markers commonly used to visualize heat release rate in laser diagnostics are found to be inadequate for the multi-component fuel flame, and alternative markers are proposed. Finally, some popular mean reaction rate closures are tested for the multi-component fuel flame. Significant differences are observed between the models’ performance at the highest turbulence level considered in this study. These arise from the chemical complexity of the fuel, and further parametric studies using skeletal chemistry DNS would be useful for the refinement of the models.

This dissertation addresses the analysis of partially premixed flame configurations and the detection and characterization of their local flame regimes. First, the identification of flame regimes in experimental data is intensively discussed. Current methods for combustion regime characterization, such as the flame index, rely on 3D gradient information that is not accessible with available experimental techniques. Here, a method is proposed for reaction zone detection and characterization, which can be applied to instantaneous 1D Raman/Rayleigh line measurements of major species and temperature as well as to the results of laminar and turbulent flame simulations, without the need for 3D gradient information. Several derived flame markers, namely the mixture fraction, the heat release rate and the chemical explosive mode, are combined to detect and characterize premixed versus non-premixed reaction zones. The methodology is developed and evaluated using fully resolved simulation data from laminar flames. The fully resolved 1D simulation data are spatially filtered to account for the difference in spatial resolution between the experiment and the simulation, and experimental uncertainty is superimposed onto the filtered numerical results to produce Raman/Rayleigh equivalent data. Then, starting from just the temperature and major species, a constrained homogeneous batch reactor calculation gives an approximation of the full thermochemical state at each sample location. Finally, the chemical explosive mode and the heat release rate are calculated from this approximated state and compared to those calculated directly from the simulation data. After successful validation, the approach is applied to Raman/Rayleigh line measurements from laminar counterflow flames, a mildly turbulent lifted flame and turbulent benchmark cases. The results confirm that the reaction zones can be reliably detected and characterized using experimental data. In contrast to other approaches, the presented methodology circumvents uncertainties arising from the use of limited gradient information and offers an alternative to known reaction zone identification methods. Second, this work focuses on the flame structure of partially premixed dimethyl ether (DME) flames. DME flames form significant intermediate hydrocarbons in the reaction zone and are classified as the next more complex fuel candidate in research after methane. To simulate DME combustion processes, accurate predictions by computational combustion models are required. To evaluate such models and to identify appropriate flame regimes, numerical simulations are necessary. Therefore, fully resolved simulations of laminar dimethyl ether flames, defined by different levels of premixing, are performed. Further, the qualitative two-dimensional structures of the partially premixed DME flames are discussed and analyses are carried out at selected slices and compared to each other as well as to experimental data. Further, the flamelet/progress variable (FPV) approach is investigated to predict the partially premixed flame structures of the DME flames. In the context of the FPV approach, a rigorous analysis of the underlying manifold is carried out based on the newly developed regime identification approach and an a priori analysis. The most promising flamelet look-up table is chosen for the fully coupled tabulated chemistry simulations and the results are further compared to the fully resolved simulation data.

Large-eddy simulation (LES) and unsteady Reynolds-averaged Navier-Stokes (URANS) calculations have been performed to investigate the effects of different mathematical models for scalar variance and its dissipation rate as applied to both a non-reacting bluff-body turbulent flow and an extension to a reacting case. In the conserved scalar formalism, the mean value of a thermo-chemical variable is obtained through the PDF-weighted integration of the local description over the conserved scalar, the mixture fraction. The scalar variance, one of the key parameters for the determination of a presumed β-function PDF, is obtained by solving its own transport equation with the unclosed scalar dissipation rate modelled using either an algebraic expression or a transport equation. The proposed approach is first applied to URANS and then extended to LES. Velocity, length and time scales associated with the URANS modelling are determined using the standard two-equation k-ε transport model. In contrast, all three scales required by the LES modelling are based on the Smagorinsky subgrid scale (SGS) algebraic model. The present study proposes a new algebraic and a new transport LES model for the scalar dissipation rate required by the transport equation for scalar variance, with a time scale consistent with the Smagorinsky SGS model.

On s’intéresse ici aux erreurs de modélisation liées à l’usage de modèles de flammelette sous-maille en combustion turbulente non prémélangée. Le but de cette thèse est de développer une stratégie d’estimation d’erreur a posteriori pour déterminer le meilleur modèle parmi une hiérarchie, à un coût numérique similaire à l’utilisation de ces mêmes modèles. Dans un premier temps, une stratégie faisant appel à un estimateur basé sur les résidus pondérés est développée et testée sur un système d’équations d’advection-diffusion-réaction. Dans un deuxième temps, on teste la méthodologie d’estimation d’erreur sur un autre système d’équations, où des effets d’extinction et de réallumage sont ajoutés. Lorsqu’il n’y a pas d’advection, une analyse asymptotique rigoureuse montre l’existence de plusieurs régimes de combustion déjà observés dans les simulations numériques. Nous obtenons une approximation des paramètres de réallumage et d’extinction avec la courbe en «S», un graphe de la température maximale de la flamme en fonction du nombre de Damköhler, composée de trois branches et d’une double courbure. En ajoutant des effets advectifs, on obtient également une courbe en «S» correspondant aux régimes de combustion déjà identifiés. Nous comparons les erreurs de modélisation liées aux approximations asymptotiques dans les deux régimes stables et établissons une nouvelle hiérarchie des modèles en fonction du régime de combustion. Ces erreurs sont comparées aux estimations données par la stratégie d’estimation d’erreur. Si un seul régime stable de combustion existe, l’estimateur d’erreur l’identifie correctement ; si plus d’un régime est possible, on obtient une fac˛on systématique de choisir un régime. Pour les régimes où plus d’un modèle est approprié, la hiérarchie prédite par l’estimateur est correcte. / We are interested here in the modeling errors of subgrid flamelet models in nonpremixed turbulent combustion. The goal of this thesis is to develop an a posteriori error estimation strategy to determine the best model within a hierarchy, with a numerical cost at most that of using the models in the first place. Firstly, we develop and test a dual-weighted residual estimator strategy on a system of advection-diffusion-reaction equations. Secondly, we test that methodology on another system of equations, where quenching and ignition effects are added. In the absence of advection, a rigorous asymptotic analysis shows the existence of many combustion regimes already observed in numerical simulations. We obtain approximations of the quenching and ignition parameters, alongside the S-shaped curve, a plot of the maximal flame temperature as a function of the Damköhler number, consisting of three branches and two bends. When advection effects are added, we still obtain a S-shaped curve corresponding to the known combustion regimes. We compare the modeling errors of the asymptotic approximations in the two stable regimes and establish new model hierarchies for each combustion regime. These errors are compared with the estimations obtained by using the error estimation strategy. When only one stable combustion regime exists, the error estimator correctly identifies that regime; when two or more regimes are possible, it gives a systematic way of choosing one regime. For regimes where more than one model is appropriate, the error estimator’s predicted hierarchy is correct.

Large-eddy simulation (LES) and unsteady Reynolds-averaged Navier-Stokes (URANS) calculations have been performed to investigate the effects of different mathematical models for scalar variance and its dissipation rate as applied to both a non-reacting bluff-body turbulent flow and an extension to a reacting case. In the conserved scalar formalism, the mean value of a thermo-chemical variable is obtained through the PDF-weighted integration of the local description over the conserved scalar, the mixture fraction. The scalar variance, one of the key parameters for the determination of a presumed β-function PDF, is obtained by solving its own transport equation with the unclosed scalar dissipation rate modelled using either an algebraic expression or a transport equation. The proposed approach is first applied to URANS and then extended to LES. Velocity, length and time scales associated with the URANS modelling are determined using the standard two-equation k-ε transport model. In contrast, all three scales required by the LES modelling are based on the Smagorinsky subgrid scale (SGS) algebraic model. The present study proposes a new algebraic and a new transport LES model for the scalar dissipation rate required by the transport equation for scalar variance, with a time scale consistent with the Smagorinsky SGS model.

Depuis quelques années, les modèles 0D trouvent un regain d'intérêt auprès des motoristes. En effet, ces modèles, fournissant aisément un comportement thermodynamique du moteur, peuvent être couplés avec des outils de contrôle moteur. Néanmoins, leur précision doit être augmentée, pour répondre aux enjeux technologiques actuels. Dans les moteurs à allumage commandé, la flamme turbulente prémélangée est modélisée comme un ensemble de flammelettes cohérentes entre elles. Cette approche généraliste nécessite un traitement particulier en proche paroi, motivé par une modification de la structure de flamme due aux couches limites thermique et cinématique. Ce présent travail propose des approches de modélisations 0D de la combustion, en proche paroi et dans la zone réactionnelle de la flamme. Pour modéliser la combustion en proche paroi, la flamme est scindée en une contribution en propagation libre, et une contribution en interaction avec les parois. Chaque contribution est divisée en une zone de transport, dans laquelle l'entraînement des gaz frais est décrit, et une zone de réaction, dans laquelle la réaction de combustion est modélisée. L'ajout d'une zone de réaction en interaction avec les parois permet de modéliser un gradient de température et une réaction de combustion ralentie en proche paroi. Pour modéliser la zone réactionnelle, une discrétisation de la flamme en N zones de réaction indépendantes est proposée. Une plage de fonctionnement moteur a été simulée avec nos approches de modélisation, afin de quantifier la variabilité des paramètres de calibration. Pour ce faire, les modèles sont calibrés sur chaque point de fonctionnement, par une méthode de minimisation de l'erreur quadratique moyenne sur la loi de dégagement d'énergie. Des corrélations aisées de paramètres de calibration peuvent être établies, en fonction de paramètres moteurs. Les résultats de simulations, obtenus à partir de ces corrélations, sont satisfaisants. / Recently, the interest for zero-dimensional models has increased. Indeed, these models provide easily the engines thermodynamic behavior and can be coupled with control tools. However, their accuracy must be improved to meet the current technological challenges. In the spark ignition engines, the premixed turbulent flame is modeled as a set of coherent flamelets. This approach requires special treatment near the walls, motivated by the modifications of the flame structure due to boundary layers. The present work proposes 0D modeling of combustion near the walls and in the reaction zone of the flame. To combustion model near the walls, the flame is divided into a free propagation contribution, and an interacting contribution with the walls. Each contribution is divided into a convective zone, wherein the entrainment of fresh gas is described, and a reaction zone, wherein the combustion reaction is modeled. Adding a reaction zone near the walls allows modeling a thermal gradient and a slower combustion reaction near the walls. To model the reaction zone, a flame discretization is made into several reaction zones. An engine operating range is simulated with our models, for quantifying the calibration parameters variability. To do this, models are calibrated on each operating point, by a method of minimization of the quadratic error on the heat released rate. Linear correlations can be found, depending on engines parameters. A good agreement between experimental data and simulation results is obtained with these parameters correlations.

No description available.