Experimental investigation and multi-objective optimization of key control parameters for low-emission heavy-duty hydrogen engines

With the increasing global energy demand and the advancement of carbon neutrality goals, efficient and low-carbon power systems have attracted extensive attention in heavy-duty engine research. Hydrogen, as a clean and sustainable fuel, shows considerable potential for heavy-duty engine applications. However, optimizing key control parameters to simultaneously improve performance and reduce emissions remains a major challenge. This study combines experimental investigation with multi-objective optimization to provide a systematic assessment of the influence of key control parameters, including the excess air coefficient (λ), spark timing(ST), and throttle position opening(TPO), on the combustion, performance, and emissions of heavy-duty hydrogen engines. The results indicate that increasing λ effectively reduces nitrogen oxide (NO x ) emissions, ST significantly affects combustion characteristics, and TPO strongly influences power output. Reasonable regulation of these critical control parameters plays a vital role in enhancing the overall performance of hydrogen engines. By employing the Bayesian Optimization-Extra Trees (BO-ET) algorithm and the multi-objective whale optimization algorithm (MOWOA), this study performs multi-objective optimization of a hydrogen engine with respect to power, brake specific fuel consumption (BSFC), and NO x emissions. The optimization outcomes show that the hydrogen engine attains a power output of 96.06 kW, a BSFC of 75.31 g/kWh, and NO x emissions of 19.77 ppm. Relative to the reference operating condition from previous experiments (TPO of 40%, ST of −24.8 °CA ATDC, λ of 2.5, and peak in-cylinder pressure of 74.66 bar), the optimized condition maintained comparable power and BSFC while significantly reducing NO x emissions by 76.5%, demonstrating a pronounced optimization effect.