Wide-parameter coarse-grained state mechanism analysis in the methane combustion system

Abstract

The coarse graining idea is introduced into the analysis of detailed combustion models to determine the key reaction steps that drive macroscopic combustion characteristics such as ignition and extinction as well as the overall reactivity of the chemical system. Based on concepts elaborated in the fields of artificial intelligence and big data analytics, a novel method that combines network community detection and parametric bifurcation techniques is proposed. The effectiveness of the method is demonstrated with simulation results of a perfectly stirred reactor with inlet temperature with in 1000 K to 2500 K and pressure ranging from 1 × 10⁴ Pa to 2 × 10⁶ Pa, and adopting the GRI-mech 3.0. It is revealed that the evolution process represented by the coarse-grained states not only can distinctively separate the process behaviour before and after ignition but also is capable of effectively identifying commonalities of the states over a wide range of parameters. In addition, analysis of the highly persistent species communities indicates that the hydrogen oxidation sub-mechanism and the secondary reactions involving formaldehyde, methanol, and carbon monoxide are the core processes for methane ignition. The results obtained agree well with previous related studies, which validate the proposed method. Thus, the coarse-grained state-based mechanism analysis method is suitable for the semi-automatic identification of the important reactions of a detailed mechanism acting under different conditions.