Detailed Analysis and Modeling of Stratified Combustion for Large-Eddy Simulation - DAMOSCLES
Final Report Abstract
We have successfully demonstrated that the stratification effects caused strong deviations in flame physics. The flame speed, thickness and curvature in a stratified flame were significantly different than in a comparable premixed flame. To some extent, the properties of a stratified flame can be modelled using the methods initially developed for premixed flames. However, the premixed flame models tended to fail for strongly stratified flames. We have shown that extending the premixed flame models to stratified flames was plausible, however, only through numerical analysis. It was evident that the stratification effects amplified when a detailed diffusive transport was needed, such cases could be hydrogen combustion. An important finding was that the stratification effects could be captured with reduced chemical models, which enabled complicated combustion systems to be simulated at an affordable computational cost. For the first time, a DNS of a highly stratified laboratory flame with sufficient timesteps for time-averaging was performed. The constructed data-set was validated by comparison against the experiment before trusting the data for further analysis and model development. The data-set was made available for other research groups in a structured format, where cost-effective post-processing routines were developed to analyse this (big) data set. An important merit of this data-set was that different stratification levels existed in the flame. Hence, the induced flame wrinkling due to stratification effects can be quantified. A priori analysis of this data-set showed that the RANS models could capture the direct computations for the flames with mild stratification. For the flames with strong stratification, an extension to these RANS model was needed, and a suggestion was presented in the project. A posteriori analysis of the extended models must, however, be conducted as future work. An attempt to visualize the effects of stratification on flame wrinkling in a regime diagram was not performed. We had to understand that such a regime diagram would not be a two-dimensional regime diagram and would need extension to a third-dimension. The extended dimension should include the stratification strength, such as the one given in our published work. Overall, we have demonstrated that the magnitude of the stratification effects cannot be neglected in technically relevant combustion systems, and have shown possible methods to model these effects.
Publications
-
A simple postprocessing method to correct species predictions in artificially thickened turbulent flames. Proceedings of the Combustion Institute 38, pp 2977-2984 (2020)
P. Gruhlke, E. Inanc, R. Mercier, B. Fiorina and A. M. Kempf
-
Analysis of mixture stratification effects on unstrained laminar flames, Combustion and Flame 219, pp 339-348 (2020)
E. Inanc, N. Chakraborty and A. M. Kempf
-
Effect of sub-grid wrinkling factor modelling on the large-eddy simulation of turbulent stratified combustion, Combustion, Theory and Modelling 25, pp 911-939 (2020)
E. Inanc, A. M. Kempf and N. Chakraborty
-
Detailed Simulations of the DLR Auto-Igniting Pulsed Jet Experiment, Fuel 284,118947 (2021)
E. Inanc, J.T. Lipkowicz, A.M. Kempf
-
Numerical simulation of pulsed and stratified combustion, Dissertation, Universität Duisburg-Essen (2022)
E. Inanc
-
Scalar gradient and flame propagation statistics of a flame-resolved laboratory-scale turbulent stratified burner simulation, Combustion and Flame 238, 111917 (2022)
E. Inanc, A. M. Kempf and N. Chakraborty
