A Priori Direct Numerical Simulation Assessment of Algebraic Models of Variances and Dissipation Rates in the Context of Reynolds-Averaged Navier-Stokes Simulations for Low Damkohler Number Partially Premixed Combustion

Statistically planar turbulent premixed and partially premixed flames for different initial turbulence intensity are simulated for global equivalence ratio =0.7 and 1.0 using three-dimensional simplified chemistry based Direct Numerical Simulations (DNS). For the simulations of partially premixed flames a bimodal distribution of equivalence ratio variation about the prescribed value of is introduced in the fresh reactants. The simulation parameters are chosen in such a manner that the combustion situation in all the cases represents the thin reaction zones regime with global Damkohler number smaller than unity. The DNS data has been used to analyze the statistics of the variances [image omitted], covariances [image omitted] (where Y, and c are the fuel mass fraction, mixture fraction, and reaction progress variable, respectively, and tilde and double prime represent the Favre mean and Favre fluctuation of the relevant quantities, respectively), scalar dissipation rates (i.e., [image omitted] and [image omitted]) of active scalar variances, and the cross-scalar dissipation rates ([image omitted] and [image omitted]) of the covariances of active scalar and mixture fraction fluctuations in the context of Reynolds-Averaged Navier-Stokes (RANS) simulations. The performances of different algebraic models for the variances, covariances, scalar dissipation rate of active scalars, and cross-scalar dissipation rates have been assessed with respect to the corresponding values obtained from the DNS database. It has been found that root mean square turbulence velocity fluctuation u' and global equivalence ratio have significant effects on the statistics of [image omitted]. The authors found that the maximum values of [image omitted] increase with increasing u' and values. Moreover, the modeling parameters of the algebraic models for the quantities [image omitted], [image omitted] and [image omitted] show significant u' and dependence. Based on the a priori DNS assessment, the algebraic models for [image omitted], which give rise to satisfactory agreement with the corresponding quantities obtained from DNS without any change in model parameters in response to the changes in u' and , have been identified. It has been found that none of the existing algebraic models for [image omitted] and [image omitted] are capable of predicting corresponding quantities extracted from the DNS data for the complete set of u' and values analyzed in the present study.