LES of a laboratory-scale turbulent premixed bunsen flame using FSD, PCM-FPI and thickened flame models

F.E. Hernandez Perez, F.T.C. Yuen, C.P.T. Groth, O.L. Gülder

    Research output: Contribution to journalArticleAcademicpeer-review

    70 Citations (Scopus)
    4 Downloads (Pure)


    Large-eddy simulations (LES) of a turbulent premixed Bunsen flame were carried out with three subfilter-scale (SFS) modelling approaches for turbulent premixed combustion. One approach is based on the artificially thickened flame and power-law flame wrinkling models, the second approach is based on the presumed conditional moment (PCM) with flame prolongation of intrinsic low-dimensional manifolds (FPI) tabulated chemistry, and the third approach is based on a transport equation for the flame surface density (FSD). A lean methane–air flame at equivalence ratio ¿=0.7, which was studied experimentally by Yuen and Gülder, was considered. The predicted LES solutions were compared to the experimental data. The resolved instantaneous three-dimensional structure of the predicted flames compares well with that of the experiment. Flame heights and resolvable flame surface density and curvature were also examined. In general, the average flame height was well predicted. Furthermore, the flame surface data extracted from the simulations showed remarkably good qualitative agreement with the experimental results. The probability density functions of predicted flame curvature displayed a Gaussian-like shape centred around zero as also observed in the experimental flame, although the experimental data showed a slightly wider profile. The results of the comparisons highlight the weaknesses and the strengths of SFS modelling approaches commonly used in LES of turbulent premixed flames
    Original languageEnglish
    Pages (from-to)1365-1371
    JournalProceedings of the Combustion Institute
    Issue number1
    Publication statusPublished - 2011


    Dive into the research topics of 'LES of a laboratory-scale turbulent premixed bunsen flame using FSD, PCM-FPI and thickened flame models'. Together they form a unique fingerprint.

    Cite this