A reduced combustion mechanism of ammonia/diesel optimized with multi-objective genetic algorithm

For the deep understanding on combustion of ammonia/diesel, this study develops a reduced mechanism of ammonia/diesel with 227 species and 937 reactions. The sub-mechanism on ammonia/interactions of N-based and C-based species (N–C)/NOx is optimized using the Non-dominated Sorting Genetic Algorithm II (NSGA-II) with 200 generations. The optimized mechanism (named as 937b) is validated against combustion characteristics of ammonia/methane (which is used to examine the accuracy of N–C interactions) and ammonia/diesel blends. The ignition delay times (IDTs), the laminar flame speeds and most of key intermediate species during the combustion of ammonia/methane blends can be accurately simulated by 937b under a wide range of conditions. As for ammonia/diesel blends with various diesel energy fractions, reasonable predictions on the IDTs under pressures from 1.0 MPa to 5.0 MPa as well as the laminar flame speeds are also achieved by 937b. In particular, with regard to the IDT simulations of ammonia/diesel blends, 937b makes progress in both aspects of overall accuracy and computational efficiency, compared to a detailed ammonia/diesel mechanism. Further kinetic analysis reveals that the reaction pathway of ammonia during the combustion of ammonia/diesel blend mainly differs in the tendencies of oxygen additions to NH2 and NH with different equivalence ratios.