Surrogate model-driven multi-parameter optimization of four distinct combustion chambers in a four-stroke diesel engine

Abstract The reduction of emissions from compression ignition (CI) engines is a major area of research in response to increasing environmental regulations and the need for cleaner energy solutions. This study presents a comprehensive computational fluid dynamics (CFD) study on a sector combustor of a CI engine, focusing on the effects of injection parameters and combustion chamber design on emissions. Four distinct combustion chamber designs—rectangular, shallow depth combustion chamber (SCC), hemispherical combustion chamber (HCC), and toroidal combustion chamber (TCC)—were analyzed. Key parameters such as tumble ratio, swirl ratio, cross-tumble ratio, soot formation, and NOx emissions were observed and compared. This study uniquely combines a comparative analysis of these combustion chamber designs with an advanced optimization of injection parameters using a Multi-Objective Genetic Algorithm (MOGA), an approach not widely explored in previous CI engine research. Additionally, response surface methodology (RSM) and design of experiments (DOE) were employed to gain further insights into the influence of individual variables, providing a thorough understanding of their impact on emissions and combustion performance. The results show that SCC chambers are most effective at reducing soot emissions, while TCC designs are better at lowering NOx emissions. Later injection timings showed a 450% decrease in soot rate compared to earlier timings, whereas NOx has significantly benefited from earlier injections, which presented about a 1600% difference in the NOx production when compared to late injections. These findings offer valuable contributions to the design of cleaner and more efficient CI engines, with the potential to significantly impact future engine development.