Development of a 3D-CFD predictive combustion model for a heavy-duty Engine fueled by ultra-lean H2 mixture

The potentially null CO2 and reduced pollutants emissions of hydrogen combustion make the research effort on hydrogen Internal Combustion Engines a strategic asset for the green transition, since it can provide on one hand, a bridging technology for low range requirement mobility (passenger vehicles), and on the other hand, it could concur as definitive solution for all the heavy duty transport sectors (heavy trucks, marine and aeronautical), where the current battery storage technologies are not yet relevant to replace the conventional fossil fuel sources. This paper consists of three main parts: the first part discusses the hydrogen adopted as energy vector, in particular the potentialities and the drawbacks of this solution; then its fuel related characteristics are presented and, finally, its adoption in internal combustion engines, both in terms of engine available solutions and combustion and pollutants formation description. The second part introduces the numerical concepts propaedeutic to the numerical model development; in particular the fundamentals of 3D CFD modeling are presented with focus on engine simulation models and sub models. The last part of the project presents the carried out numerical investigations: the most representative engine working points are calibrated against experimental available data and the model performance is compared with a reference SAGE model to evaluate the improvements in prediction capability; then the setup is refined with a study on the best Table of Laminar Flame Speeds TLF, boundary conditions and runtime optimization. In the end, hydrogen combustion instabilities are implemented on the model. The thesis work aims to continue the development of a Computational Fluid Dynamic CFD predictive model for the combustion analysis of a hydrogen fueled internal combustion engine for heavy duty applications. The development process starts from a baseline model, which is the result of previous research projects finalized to characterize the engine combustion process in the 3D CFD environment with the hydrogen fuel. The objective of this work is to enrich the baseline model by improving its prediction capability and covering the most representative engine operating points. The numerical model relies on ECFM for the combustion modeling fed by 1D TLF evaluated with the chemical kinetics mechanism C3V3.5 Hydrogen. The ISSIM is adopted for the spark discharge control sub model, while the turbulence modelling is based on RANS approach, particularly with RNG κ-ε. The simulations are performed with CONVERGE CFD software thanks to its perks when dealing with engine simulations like automatic meshing and adaptive mesh refinement control.