In-cylinder pressure provides critical insights for analyzing and optimizing combustion in internal combustion engines, yet its acquisition across the full operating space requires extensive testing, while physics-based models are computationally demanding. Machine learning (ML) offers an alternative, but its application to direct reconstruction of full pressure traces remains limited. This study evaluates three strategies for reconstructing cylinder pressure under unmeasured operating conditions, establishing a machine learning-assisted framework that generates the complete pressure–crank angle (P–CA) trace. The framework treats crank angle and operating conditions as inputs and predicts either pressure directly or apparent heat release rate (HRR) as an intermediate variable, which is then integrated to reconstruct pressure. In all approaches, discrete pointwise predictions are combined to form the full P–CA curve. Direct pressure prediction achieves high accuracy for overall traces but underestimates HRR-related combustion features. Training on HRR improves combustion representation but introduces baseline shifts in reconstructed pressure. A hybrid approach, combining non-combustion pressure prediction with combustion-phase HRR-based reconstruction delivers the most robust and physically consistent results. These findings demonstrate that ML can efficiently reconstruct in-cylinder pressure at unmeasured conditions, reducing experimental requirements while supporting combustion diagnostics, calibration, and digital twin applications.

The indicated mean effective pressure (IMEP) is a key parameter for measuring the power output of an internal combustion engine (ICE). This indicator can be used to locate the high efficiency regions of engines. Therefore, it makes sense to predict the IMEP based on the machine learning (ML) approaches. However, different ML models are applicable to different scenarios, so it is important to choose the right model for prediction. The objective of this paper was to compare three ML models’ (ANN, SVR, RF) predictive performance in forecasting IMEP indicator with the input parameters spark timing (ST), speed and load. A validated one-dimensional (1D) computational fluid dynamics (CFD) model was employed to provide 756 sets of data for the training, validation, and testing of the model. The results indicated that the random forest (RF) model had the worst prediction performance, and support vector regression (SVR) had a slightly better prediction performance than the artificial neural network (ANN), at least for the investigations in this study. Overall, the ANN and SVR models showed good predictive performance for IMEP, as the coefficient of determination (R²) was close to unity, and the root mean squared error (RMSE) was close to zero. Whereas the overall prediction results of the RF model are acceptable, the RF model does not learn well for some internal engine laws.

Due to the pressing issue of global warming, there has been a significant focus on zero- and low-carbon fuels globally. Among hydrocarbon fuels, methane is widely used in spark ignition engines due to its abundance and relatively low carbon footprint. However, to further reduce carbon emissions, interest is growing in the use of ammonia, a zero-carbon fuel, as a partial replacement for methane. Consequently, it is crucial to investigate the impact of ammonia addition on the performance of natural gas spark ignition engines. A key challenge in studying ammonia-methane engines is that the introduction of ammonia alters the formation mechanisms of nitrogen-based pollutants, resulting in the coupling of fuel-borne and air-borne nitrogen pollutants. As a result, research on the nitrogen-based emissions of ammonia-methane engines has been limited. This study addresses this issue by differentiating between atmospheric nitrogen and fuel nitrogen elements, effectively decoupling fuel-borne and air-borne nitrogen pollutants. This approach provides valuable insights into the effects of ammonia addition on the nitrogen-based pollutant characteristics of natural gas engines. The results indicate that ammonia addition introduces N₂O, a species absent in pure methane engines. The N₂O primarily originates from cold wall regions and the partial oxidation of ammonia released from engine crevices during the late oxidation process. Although NO remains the dominant nitrogen-based pollutant and the amount of N₂O is small, the significant greenhouse gas potential of N₂O warrants further attention. Furthermore, while ammonia addition increases the NO concentration in the burning zone, it slightly reduces the NO concentration at chemical equilibrium under stoichiometric conditions. As a result, engines operating with an ammonia energy substitution ratio of 0.4 exhibit lower NO_x emissions compared to those fueled solely by methane. These findings underscore the need for further research into the combustion and emission characteristics of ammonia-methane spark ignition engines.

Amid climate change, reducing reliance on fossil fuels and transitioning to renewable energy sources is crucial. Ammonia, a carbon-free and renewable fuel, shows significant potential as an alternative energy source. By incorporating hydrogen as an additive, its flammability can be enhanced to suit existing spark ignition engines. However, understanding the characteristics of nitrogen pollutant emissions (i.e., NO_x, which includes NO and NO₂, and N₂O) from ammonia-hydrogen combustion is challenging due to contributions from both fuel-borne and air-borne nitrogen. However, understanding the characteristics of nitrogen pollutant emissions from ammonia-hydrogen combustion is challenging due to contributions from both fuel-borne and air-borne nitrogen. Therefore, a comprehensive understanding of fuel-borne nitrogen pollutants during ammonia-hydrogen combustion is essential. This study focuses on investigating fuel-borne nitrogen pollutants in an argon-oxygen atmosphere, thereby eliminating nitrogen from the oxidizer and its role in thermal NO_x formation. The research examines the formation and evolution of fuel-borne nitrogen pollutants during ammonia-hydrogen combustion under engine-like conditions. Results indicate that fuel-borne nitrogen pollutants act as intermediates, potentially originating from chemical equilibrium. While fuel NO predominantly forms in the burning zone, it undergoes reduction in the burned zone. N₂O, absent in thermal NO_x mechanisms, shows significant concentrations in the burning zone and is mostly converted to N₂, leading to limited N₂O in the final fuel-borne nitrogen pollutant concentration. Lean burn conditions, hydrogen addition, and oxyfuel combustion promote fuel NO_x formation. Additionally, the equivalence ratio affects the ammonia-hydrogen premixed flame structure due to the de-NO_x effect of ammonia. Overall, these findings highlight that fuel-borne nitrogen pollutant mechanisms differ from thermal NO_x mechanisms, necessitating specially designed reduction technologies for clean spark ignition engines.