Injection and combustion modelling in hydrogen internal combustion engines

This thesis focuses on the numerical analysis of the injection, mixing, and combustion in direct injection internal combustion engines fueled by hydrogen. Most likely, the current transition towards a sustainable mobility will be achieved only through a combination of different technologies. For specific applications, such as heavyduty road transport, hydrogen-powered internal combustion engines—particularly those based on direct injection — represent a competitive powertrain solution in terms of cost and reliability compared to more innovative alternatives like fuel cells. Although internal combustion engines have been in use for over a century, the unique chemical and physical properties of hydrogen necessitate a renewed analysis of the fundamental processes governing engine operation, such as injection, fuel-air mixing, and combustion. For this reason, the first step of this thesis is the characterization of the fluid dynamics of high-pressure hydrogen direct injection into the combustion chamber. The study analyzes the fluid-dynamic structures characterizing these flows, with particular attention to the numerical setup required to accurately resolve them. With respect to the state of art, the analysis focuses on the effect of the typical combustion chamber conditions on the jet phenomenology. More specifically, an analysis is proposed by varying the cylinder pressure over time to determine the possible presence of dynamic phenomena that could alter the flow configuration compared to what is observed under constant pressure ratios. As a second step, focus has been posed on the influence of the injector exit section geometry, to determine whether specific configurations are more suitable for use in internal combustion engines. Although the injector outlet sections are primarily circular or annular, new geometries (triangular, star-shaped, rectangular, and elliptical) have been studied to understand their effect on the mixing between the jet and the surrounding environment. While some of these geometric configurations have already been analyzed in the aerospace field, their applicability in the automotive sector has been tested in this thesis work. Significant differences have been observed in the shock and expansion system characterizing these flows, for which a detailed explanation has been provided. Subsequently, the analysis has been extended to the design space of key variables affecting the mixing process, such as the injection pressure, timing, and nozzle geometry. The goal has been to extend the engine operability under increasingly lean mixtures compared to the current state of the art. More specifically, focus has been placed on mixtures featured by an equivalence ratio of 0.25, which help to reduce wall heat losses while limiting nitrogen oxide emissions. It has been demonstrated that the mixture stratification under these conditions is the key to increasing the combustion process speed. In particular, multi-injection strategies have been tested, showing the potential to optimize the combustion speed and NOx emissions. Finally, in an effort to improve internal combustion engine modeling, a procedure has been developed to account for the effects of thermodiffusive instabilities in hydrogen flames within the framework of the G-equation combustion model. Preliminary results show the potential to effectively capture pressure trends with a single preliminary calibration process regarding the interaction between the flame front and the in-cylinder turbulence.