Modeling Of V-Shaped Flame with Flamelet Generated Manifold and Scale Adaptive Simulations Turbulence Model

Abstract

Flame stabilization on bluff bodies requires a precise balance between chemistry and flow dynamics. To correctly predict the critical aspects such as lift-off conditions, flame spreading, and shape of the flame, scale resolving turbulence models provide an accurate representation of turbulent flow, which is crucial to model the turbulence chemistry interaction in reacting flows. In this paper, the impact of turbulence on turbulent burning velocity and flame structure has been simulated by using a laboratory-scale V-flame configuration. In the V-flame configuration, a turbulent premixed CH4/air flame is stabilized on a rod, and the parameters describing this configuration are – turbulent Reynolds number of 40 and Damkohler number of 6. First, the turbulence is modeled using the Large Eddy Simulation (LES) with different grid resolutions illustrating variation in energy ratio, i.e., ratio of resolved energy to the total energy. The LES solution obtained is computationally expensive due to the grid resolution required to predict turbulent burning velocity accurately. As a result, alternate approaches have been studied for modeling turbulence-chemistry interactions namely unsteady RANS (URANS) model and hybrid RANS/LES models like Scale Adaptive Simulations (SAS) and Stress Blended Eddy Simulation (SBES). The simulations are performed using the Flamelet generated manifold (FGM) model. The manifold is generated by one-dimensional freely propagating premixed flames. Results are compared with the available experimental data to validate the flame morphology and mean velocity field. Each approach's definitive advantages and disadvantages are described based on computational accuracy and the cost of turbulence-chemistry closure. The SAS-FGM approach is promising as the results demonstrate the model's capability to resolve the scales at a similar level to LES while maintaining the computational efficiency of URANS. The results of SAS-FGM are then compared with DNS (Direct Numerical Simulation) results and experiment, and the differences are discussed. This modeling approach is then used to perform the parametric investigation, such as temperature boundary conditions on the stabilizing rod, inlet velocity changes to model lifted flame, and finally to the limits of a blow-out. The accuracy of the model is assessed by capturing the sensitivity of flame shape and spread angle to the change in input parameters leading to regime changes from the stable flame to lifted flames. The results from these laboratory-scale model simulations provide essential information on flame dynamics for flames approaching blow-out limits.