Assessment of Subgrid-Scale Model Effects on Large Eddy Simulation of a Back-Step Combustor

Weilin Zeng, Konstantina Vogiatzaki, Kai Luo, Salvador Navarro-Martinez

Research output: Chapter in Book/Conference proceeding with ISSN or ISBNConference contribution with ISSN or ISBNpeer-review

Abstract

Much progress has been made in large-eddy simulation (LES) of turbulent combustion in the last two decades, but a robust and cost-effective LES formulation is still lacking for turbulent combustion in practical configurations. In this paper, we present an assessment of different sgs models and the no sgs approach within the context of LES of a backward step combustor. Overall, the dynamic one equation eddy model behaves better than the WALE and one equation eddy models, in both reproducing main features and statistical quantities of the non-reactive and reactive flow fields. Increasing grid resolution does not necessarily improve the predictions. The results are largely dependent on whether the local flow is turbulence or combustion dominated. This implies that along with an adaptive grid refinement, an adaptive combustion model strategy is needed. In combustion simulation, applying only the first term in the series model is insufficient to well predict the dominating features and statistical quantities of the reacting flows. Thus, we suggest as future work the introduction of additional adaptive terms that will control the variance.
Original languageEnglish
Title of host publicationProceedings of the 8th European Combustion Meeting (ECM 2017)
Place of PublicationCroatia
Pages1-6
Number of pages6
Publication statusPublished - 20 Apr 2017
EventProceedings of the 8th European Combustion Meeting (ECM 2017) - Dubrovnik, Croatia, 18-21 April 2017
Duration: 20 Apr 2017 → …

Conference

ConferenceProceedings of the 8th European Combustion Meeting (ECM 2017)
Period20/04/17 → …

Fingerprint

Dive into the research topics of 'Assessment of Subgrid-Scale Model Effects on Large Eddy Simulation of a Back-Step Combustor'. Together they form a unique fingerprint.

Cite this