Multi-objective optimization of combustion and emission characteristics in an ammonia‑hydrogen-diesel tri-fuel engine using ELM-MOPSO

Ammonia/diesel blended fuel significantly reduces carbon emissions, but it has issues such as high unburned ammonia emissions and low combustion efficiency. To achieve the goals of carbon peak and carbon neutrality, this study developed a combustion and emission model for an ammonia‑hydrogen-diesel tri-fuel engine based on numerical simulations, and the model was validated using experimental data available in the literature. On this basis, the effects of hydrogen energy substitution ratio, injection timing, and nozzle diameter on combustion characteristics and emission performance were systematically investigated. Furthermore, a collaborative optimization of injection parameters was conducted by integrating the Extreme Learning Machine with a multi-objective Particle Swarm Optimization method. The results show that hydrogen addition increases nitrogen oxides, and significantly reduces unburned ammonia emissions. At an ammonia energy substitution ratio of 40%, delaying the injection timing increases unburned ammonia and carbon soot emissions, while reducing nitrogen oxide emissions by approximately 22%. The optimal parameter combination obtained through optimization is 12 °CA BTDC and a nozzle diameter of 0.16 mm, which reduces unburned ammonia emissions by 58.7% and increase in-cylinder pressure by 8.5%, achieving a coordinated optimization of emission control and combustion efficiency. This study provides a theoretical basis for the optimization of ammonia/hydrogen/diesel blended-fuel engines.