Numerical optimization of methane-based fuel blends under engine-relevant conditions using a multi-objective genetic algorithm

The objective of this work is to examine in a systematic way, how conflicting requirements such as maximum ignition delay time and laminar flame speed can be met by adding gaseous components to methane in order to obtain the optimal fuel blend under engine-relevant conditions. Low-dimensional models are coupled with a multi-objective optimization algorithm in order to compute optimal methane/hydrogen, methane/syngas and methane/propane/syngas blend compositions that maximize simultaneously the ignition delay time, the laminar flame speed and the Wobbe number. The non-dominated sorting genetic algorithm (NSGA-II) is used to generate a set of Pareto solutions, and the best compromise solutions are then determined by the technique for order preference by similarity to ideal solution (TOPSIS). It was found that the GRI-Mech 3.0 mechanism could not accurately predict ignition properties of methane-based fuel blends under engine-relevant conditions. The optimization results revealed that initial conditions have a significant effect on the optimal fuel blend composition. For methane/hydrogen and methane/syngas blends, pure methane was the optimal fuel at high temperatures and low equivalence ratios, while high hydrogen contents were beneficial at lower temperatures. When the ignition delay time is of higher importance, the optimal composition shifted towards higher carbon monoxide contents. Blends with higher hydrogen and syngas contents resulted in reduced ignition delay times and higher laminar flame speeds. Regarding the methane/propane/syngas blend, the presence of propane in the optimal blend was found to be more favorable than hydrogen and carbon monoxide to satisfy the objectives.