La dégradation de l’environnement ainsi que l’épuisement progressif des énergies fossiles devient très inquiétant et incite les états à définir des limites d’émission polluantes plus strictes. Ceci a conduit les constructeurs automobiles à poursuivre leurs recherches dans le développement de conception propre et efficace des moteurs en utilisant des combustibles alternatifs dans les moteurs à combustion interne.Dans le présent travail, on s’intéresse à l’étude des moteurs fonctionnant en mode DF afin d’améliorer ses performances tout en minimisant les émissions polluantes, en particulier les HC et les CO. Pour ce faire des études expérimentales ont été menées. Une réduction de 77% des émissions de HC a été observée en passant d’une richesse de 0,35 à 0,7. Par ailleurs, Il a été noté aussi qu’une diminution de 20% à 50% des émissions de CO avec une amélioration de 30% du rendement peut être visualisée en variant l’avance à l’injection de 4,5 °V à 6 °V. Concernant la mise en place de la pré-injection, une baisse de 30% des émissions de NOx a été observée avec un gain de 12% à 30% de rendement par rapport à une seule injection. En dernier terme, un modèle thermodynamique à une zone a été développé afin de prédire la température et la pression dans le cylindre. Une bonne concordance a été notée entre les deux résultats avec une erreur moyenne relative inférieure à 5%. / Currently, the environmental degradation due to pollutant emissions and the gradual depletion of fossil fuels, becoming very worrying, are prompting European directives to set pollutant emission limits. These have led manufacturers to continue research in the development of clean and efficient engine designs using alternative fuels in internal combustion engines.In this work, we focus on the study of engines operating in dual-fuel mode to improve its performance while minimizing pollutant emissions, particularly HC and CO. For this, experimental studies were conducted. A reduction of about 77% in the HC emissions was observed as the equivalence ratio was varied from 0.35 to 0.7. Regarding the effect of injection timing, it was noted that the CO emissions decreased about 20% to 50% with an improvement in the brake thermal efficiency by 30% upon varying the injection advance from 4,5 °CA to 6 °CA. On the other hand, the introduction of pre-injection strategy led to a decrease by 30% in NOx emissions with an amelioration of brake thermal efficiency of 12% to 30% compared to a single injection. Lastly, a single zone thermodynamic model was developed to predict the in-cylinder temperature and pressure. A good agreement was noted between the predicted and experimental results. The average relative error was less than 5%.

<p>One of the major challenges in the field of internal combustion engines is keeping up with the advancements in electrification and hybridization. Automakers are striving to design environment – friendly and highly efficient engines to meet stringent emission standards worldwide. Improving engine efficiency and reducing heat losses are critical aspects of this development. Therefore, accurate heat transfer prediction capabilities play a vital role in engine design process. Current methods rely on computationally intensive 3D numerical analyses, there is a growing interest in reliable simplified models. </p> <p>In this study, a 1D diesel engine model featuring predictive combustion was integrated with a detailed finite element thermal primitive based on the 3D meshing feature available in GT Suite. Coolant and oil hydraulic circuits were incorporated in the model. The model proves to be an effective means to assess the impact on heat rejection and engine heat distribution given by an engine calibration and operating conditions. </p> <p>This work also contributes to the advancement of virtual IC engine development methods by focusing on the design and tuning of complex engine system models using GT Power for accurate prediction of engine performance. The current processes in engine simulations are assessed to identify sources of errors and opportunities for improvements. The methods discussed in this work include isolated sub system level calibration and model evolution specifically address the issue of identifying noise factors and issues in smaller parts. Additionally, the study aims on improving the model’s trustworthiness by computing 1st law sanity checks, replicating real-life compressor map calculations and refining GT’s existing global convergence criteria. </p>