Oxy-fuel combustion technology, which is based on burning coal in a mixture of oxygen and recycled flue gas (RFG), is suggested as one of new promising technologies for capturing CO2 from power plants. In this thesis study, the pyrolysis and combustion behaviour of various fuels including imported coal, petroleum coke, two different types of indigenous lignites, olive residue and their blends with different proportions in air and oxy-fuel conditions were investigated by using non-isothermal thermogravimetric method (TGA) coupled with Fourier-transform infrared (FTIR) spectrometer. Pyrolysis tests were carried out in nitrogen and carbon dioxide environments, which are the main diluting gases of air and oxy-fuel environment, respectively. Pyrolysis results reveal that weight loss profiles are similar up to high temperature zone in both pyrolysis environments, indicating that CO2 behaves as an inert gas in this temperature range. However, further weight loss takes place in CO2 atmosphere v after 700oC due to CO2-char gasification reaction which is observed in pyrolysis of all fuel samples. Combustion experiments were carried out in four different atmospheres / air, oxygen-enriched air environment (30 % O2 &ndash / 70 % N2), oxy-fuel environment (21 % O2 &ndash / 79 % CO2) and oxygen-enriched oxy-fuel environment (30 % O2 &ndash / 70 % CO2). Combustion experiments show that replacing nitrogen in the gas mixture by the same concentration of CO2 does not affect the combustion process significantly but leads to slight delay (lower weight loss rate and higher burnout temperature) in combustion. Overall comparison of weight loss profiles shows that higher oxygen content in the combustion environment is the dominant factor affecting the combustion rather than the diluting gas. As O2 concentration increases profiles shift through lower temperature zone, peak and burnout temperatures decrease, weight loss rate increases and complete combustion is achieved at lower temperatures and shorter times. Pyrolysis and combustion behaviour of three different fuel blends were also investigated. Results reveal synergistic interactions in combustion tests of all blends in all combustion environments. During pyrolysis and combustion tests gaseous products CO2, CO, H2O, CH4, SO2 and COS were identified in flue gas and analyzed by using FTIR. Results indicate that higher CO and COS formation take place during pyrolysis tests due to gasification reaction in CO2 atmosphere at high temperature zone. Gaseous species evolution trends in combustion tests are found specific for each fuel. However, evolution trends slightly shift to lower temperatures in oxygen-enriched conditions.

Aujourd’hui, les constructeurs automobiles continuent de chercher les technologies renouvelables face à la pénurie d’énergie et les problèmes d’émission de polluants. Un moyen important pour optimiser l’économie de carburant et réduire les émissions polluantes des moteurs à allumage commandés est le concept ‘downsizing’. Cependant, ce concept est limité par le phénomène de cliquetis dû aux conditions de haute température et haut pression. Dans cette étude, le contrôle de la concentration d’oxygène dans l’air est proposé. Car d’une part, la combustion enrichie en oxygène permet d’améliorer la densité de puissance de moteur avec le même niveau de pression d’admission. Cela permet soit de ‘booster’ la combustion pour augmenter la puissance du moteur ou de l’activer lorsque le moteur fonctionne à faible charge ou dans des conditions de démarrage à froid. D’autre part, une faible concentration en oxygène dans l’air (ou dilution de N2) par un système membranaire peut être considérée comme une alternative à la recirculation des gaz d’échappement. Les expériences ont été effectuées dans un moteur monocylindre ‘downsizing’ avec différents taux d’oxygène et richesse. L’étude de l’impact du contrôle de la concentration d’oxygène sur les caractéristiques de combustion et d’émissions a été effectuée pour plusieurs charges en fonctionnement optimum pour limiter la consommation spécifique de carburant. L’effet de la concentration en oxygène sur les caractéristiques de combustion du moteur a été simulé en utilisant le logiciel commercial AMESim avec le modèle de combustion développé par IFP-EN. En mettant en oeuvre des corrélations de la vitesse de combustion laminaire, déterminées au préalable durant ce travail, et délai d’auto-inflammation, les pressions dans les cylindres sont parfaitement calibrés avec une erreur maximale inférieure à 2% et l’intensité du cliquetis a pu être prédite. / Nowadays, car manufacturers continue to lead researches on new technologies facing to the energy shortage and pollutant emission problems. A major way to optimise fuel economy and reduce pollutant emissions for Spark-Ignition (SI) engines is the downsizing concept. However, this concept is unfortunately limited by ‘knock’ phenomena (abnormal combustion) due to high temperature and high pressure in-cylinder conditions. In the present study, control the oxygen concentration in air is proposed. Indeed, on the one hand, oxygen-enriched combustion can improve engine power density with the same intake pressure level. Thus, oxygen-enriched combustion can be used either as a booster to increase engine output or as a combustion enhancer when the engine operates at low loads or in cold start conditions. On the other hand, low oxygen concentration in air (or N2 dilution) can be considered as an alternative to exhaust gas recirculation (EGR). The experiments were carried out in a downsized single-cylinder SI engine with different rates of oxygen and equivalence ratios. The study of the impact of controlling oxygen concentration on the combustion characteristics and emissions was performed at several loads by optimizing the spark advance and the intake pressure to maintain the load and obtain a minimum value of indicated Specific Fuel Consumption (SFC). The effect of oxygen concentration on the engine combustion characteristics was simulated by using the commercial software AMESim, with the combustion model developed by IFP-EN. By implementing correlations for the laminar burning velocity, determined previously during this study, and auto-ignition delay data base, the in-cylinder pressures were perfectly calibrated with a maximum pressure relative error less than 2%, and the knock intensity was predicted.