Computational Fluid Dynamics Study of a Nonpremixed Turbulent Flame Using openfoam: Effect of Chemical Mechanisms and Turbulence Models

Abstract

This article presents a study of the influence of chemical mechanisms and turbulence models on Reynolds-averaged Navier–Stokes (RANS) simulations of the CH4/H2/N2-air turbulent diffusion flame, i.e., the so-called DLR-A flame. The first part of this study is focused on the assessment of the influence of four chemical models on predicted profiles of the DLR-A flame. The chemical mechanisms considered are as follows: (i) a C2 compact skeletal mechanism, which is derived from the GRI3.0 mechanism using an improved multistage reduction method, (ii) a C1 skeletal mechanism containing 41 elementary reactions amongst 16 species, (iii) the global mechanism by Jones and Lindstedt, (iv) and a global scheme consisting of the overall reactions of methane and dihydrogen. RANS numerical results (e.g., velocities, temperature, species, or the heat production rate profiles) obtained running the reactingFOAM solver with the four chemical mechanisms as well as the standard k − ɛ model, the partially stirred reactor (PaSR) combustion model, and the P − 1 radiation model indicate that the C2 skeletal mechanism yields the best agreement with measurements. In the second part of this study, four turbulence models, namely, the standard k − ɛ model, the renormalization group (RNG) k − ɛ model, realizable k − ɛ model, and the k − ω shear stress transport (SST) model, are considered to evaluate their effects on the DLR-A flame simulation results obtained with the C2 skeletal mechanism. Results reveal that the predictions obtained with the standard k − ɛ and the RNG k − ɛ models are in very good agreement with the experimental data. Hence, for simple jet flame with moderately high Reynolds number such as the DLR-A flame, the standard k-epsilon can model the turbulence with a very good accuracy.