0D/1D simulation of a heavy-duty engine fuelled with methane-hydrogen mixtures

Climate change is an issue that can no longer be overlooked, representing one of the most critical emergencies of our time, due to the higher amount of greenhouse gas (GHG) emissions. The transportation sector has a significant role in the production of GHG. New legislations require a reduction of greenhouse gas and pollutant emissions. Therefore, internal combustion engines must face challenges but also opportunities for innovation, especially through the adoption of alternative low-carbon fuels. Methane is considered one of the most promising fuels for heavy-duty engines, due to its high octane number and higher thermal efficiency compared to gasoline. However, the sole adoption of methane is not sufficient to be compliant with the more stringent regulations. Hydrogen, on the other hand, is a carbon-free energy carrier capable of reducing GHG emissions. Blending hydrogen with methane is considered a promising transitional strategy. However, hydrogen enrichment introduces challenges related to abnormal combustion phenomena and higher NOx formation. This trade-off highlights the importance of an accurate simulation capable of reproducing the effect of the blend on engine performance, emissions, and combustion stability. This thesis aims to analyse the potential benefits of adding hydrogen to a six-cylinder heavy-duty compressed natural gas SI engine. Two hydrogen blending ratios are evaluated, specifically 15% and 25%. 0D–1D numerical models were validated and optimised within the GT-Suite simulation environment. The activity has been carried out in collaboration with the Institute of Science and Technologies for Sustainable Energy and Mobility (STEMS) of the Italian National Research Council in Naples. The study is structured into two main parts: steady-state analysis and transient analysis. In the first part, four representative engine operating points from the engine workplan were analysed. Starting from a complete six-cylinder engine model, an equivalent mono-cylinder model was created to simplify the analysis without losing information. A calibration and validation process was carried out to ensure validity. Subsequently, two approaches to handling the combustion were investigated: the Three Pressure Analysis (TPA) and the SI-TURB model. The TPA method estimates the apparent burn rate from intake, in-cylinder, and exhaust pressure data. The SI-TURB model, on the other hand, is a predictive approach based on turbulence and flame propagation that requires an extensive calibration through sensitivity analyses and optimisation campaigns. After validation with methane, hydrogen was inserted into the model. For this purpose, the injector system was modelled to account for both methane and hydrogen injection. Knock tendency was also investigated through a dedicated analysis aimed at identifying the conditions under which this abnormal combustion phenomenon occurs. Specifically, variations in compression ratio, intake pressure, and spark advance were analysed. In transient conditions, the model was run in a specified portion of the World Harmonised Transient Cycle (WHTC) to validate it under realistic driving conditions. Two modelling strategies were compared. The first employed a single representative operating point as the combustion object for the entire engine model, corresponding to the most frequent condition in that cycle segment. The second approach was based on a Look-Up Table (LUT) that included the four operating points tested in the steady-state analysis.