Development of reduced and optimized mechanism for ammonia/ hydrogen mixture based on genetic algorithm

Abstract

As carbon-free and renewable fuels, ammonia and hydrogen are promising fuels to be utilized in combustion devices. For this, the genetic algorithm was employed to reduce and optimize the mechanism for ammonia/hydrogen mixture, which was aimed to develop a compact and reliable skeletal mechanism for real engine simulation with ammonia as fuel. During the progress of GA reduction, the extracted sub-mechanisms was constrained high fidelity with original mechanism on predicting the combustion characteristics through the definition of penalty functions. Besides, the number of reactions and computer calculation time of the sub-mechanism were simultaneously tracked to reduce the chemical stiffness. The resulting mechanism contained 29 species and 63 reactions and was streamlined to be coupled into complex CFD calculation. The reduced mechanism was adjusted to better capture fundamental combustion data and the mechanism optimization was finished automatically through GA operation. To verify the reliability of the optimized mechanism, the experimental data including ignition delay times, laminar flame speed, JSR species concentration, premixed flame structure and HCCI engine in-cylinder pressure at various conditions was utilized to evaluate the performance of the models. As a result, the optimized mechanism was capable to reproduce those measurements and was credible for engine combustion simulation.