Combustion characterization in a HD lean-burn hydrogen engine by means of CFD modeling

This thesis project aims to develop a computational fluid dynamics (CFD) model for the combustion analysis of a heavy duty hydrogen-fueled engine. Hydrogen is a very interesting alternative to fossil fuels, the most important disadvantages of which are its limited availability and pollutant emissions. Hydrogen, on the other hand, is very attractive thanks to its combustion product, which is water vapor, making it very sustainable. Among the several commercial CFD codes on the market, CONVERGE has been used for this study. This innovative software's most interesting quality is the time saving for mesh creation, which is automatically generated and refined locally. The engine under investigation is a HD hydrogen engine for which experimemental data were available. The combustion models used in this work are ECFM, which is able to simulate the combustion by computing the interaction between the laminar flame speed and the turbulence without reproducing all the chemical kinetics, thus reducing numerical cost, and ISSIM which can simulate the spark plug behavior. Furthermore, laminar flame speed tabels have been generated by means of 1D combustion simulations using the CONVERGE Chemistry Tool, in order to provide the laminar flame speed input to the ECFM combustion model. The ECFM model results are validated against the experimental data and then compared with the detailed chemistry combustion model SAGE ones, which simulations were previously carried out and provided. Furthermore, the effect of different combustion mechanisms of chemical kinetics on the simulations results is also analyzed, in order to identify the mechanism that can reproduce the real engine behavior as efficiently as possible. This thesis can be divided into three main parts: the first part provides an introduction to CFD, the governing equations used by the software to compute the results of the simulations, and a brief description of how the CONVERGE software works. The second part focuses on the engine object of this study, its experimental data, and the case setup modeling in CONVERGE to accurately simualte its behavior, including a description of the models used. In the final part, the results obtained are compared and discussed with appropriate images and plots, focusing on the calibration and validation of the model against the experimental data.