Comprehensive comparison of global mechanisms for methane combustion optimized using artificial neural network

Résumé fourni par la source

Global mechanisms significantly reduce computational costs in methane combustion simulations but their predictive accuracy and applicability are often limited due to the lack of validation across diverse combustion conditions. This study systematically evaluates various global mechanisms by comparing them to the detailed mechanism GRI-Mech 3.0 (abbreviated as GRI-3.0) under a range of temperature (1200 K−2500 K), oxygen concentrations (0.1–1), and mixing modes. All global mechanisms are first optimized using an artificial-neural-network (ANN)-based method to closely match GRI-3.0 in the perfectly stirred reactor (PSR). Their performances are tnen assessed in computational fluid dynamics (CFD) simulations of various combustion systems, including the open jet-in-hot-coflow (JHC) and in-furnace systems, under both non-premixed and premixed conditions. Results indicate that under non-premixed conditions, the global mechanism proposed by Si et al. (Energy & Fuels, 2021, 35(18), 14941-14953.) performs best (RECO,averaged = 0.179, RET,averaged = 0.137), and the reaction CH4+H2O→CO+3H2 leads to an overestimation of XCO. Under premixed conditions, the mechanism proposed by Jones & Lindstedt (Combust Flame, 1988, 73(1), 233–249.) more accurately predicts CO distribution (RECO,averaged = 0.354), highlighting the critical role of the reaction CH4+H2O→CO+3H2 in converting intermediate CO and the reaction CO+0.5O2↔CO2 leads to inaccurate CO conversion.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.