Mixing Modelling Framework Based on Multiple Mapping Conditioning for the Prediction of Turbulent Flame Extinction

Abstract

This is the accepted version of the paper.This version of the publication may differ from the final published version. Abstract A stochastic implementation of the Multiple Mapping Conditioning (MMC) 8 approach has been applied to a turbulent piloted jet diffusion flame (Sandia flame F) that 9 is close to extinction. Two classic mixing models (Curl's and IEM) are introduced in the 10 MMC context to model the turbulent mixing. The suggested model involves the use of a 11 reference space (that is mapped to mixture fraction space) in order to define particle prox-12 imity. The addition of the MMC ideas to the IEM and Curl's models, that is suggested in 13 the current work, aspires to combine the simplicity of these two models with the enforced 14 compositional locality without violating the linearity and independence principles. The for-15 mulation of the approach is discussed in detail and results are presented for the mixing field 16 and reactive species. The predictions are compared with joint-scalar PDF simulations using 17 the same mixing models and experimental data. Moreover, the sensitivity of the model to 18 the particle number is examined. It is shown that MMC is less sensitive to the number of 19 particles and can generally produce improved predictions of major and minor chemically 20 reacting species with a lower number of particles. Permanent repository link 21K. Vogiatzaki

A multiple mapping conditioning mixing model with a mixture-fraction like reference variable. Part 2: RANS implementation and validation against a turbulent jet flame

“…By using independent reference variables to enforce this localness, MMC does not violate the requirements of linearity or independence. MMC mixing models in both their original [21,5] and generalised [4] forms have been successfully validated for RANS simulations of the Sandia D-F flames.…”

A multiple mapping conditioning mixing model with a mixture-fraction like reference variable. Part 1: Model derivation and ideal flow test cases

“…Vogiatzaki, Kronenburg and co-workers have developed the stochastic version of MMC-RANS in recent years [29,30,31]. They demonstrate how the dispersion between the reference variables, ξ, having a stationary Guassian distribution, and the scalar fields, φ, can be used to independently model the conditional fluctuations while maintaining the correct rate of decay of the macroscale unconditional fluctuations.…”

Developments in multiple mapping conditioning for turbulent non-premixed and premixed combustion simulation

“…Initially, when ∆ p ∆ g , the reduction of the number of particle per cell does not change the filtering scales ∆ L ≈ ∆ E ; the main effect of the reduction is the increase of stochastic errors in evaluation of average properties within each cell, which, at the leading order, does not change the model. (Strictly speaking mixing models are not fully invariant with respect to the number of particles and reduction of particles may lead to loss of stochastic independence of the particles or to increased probability of extinction events [118]; simulations often indicate existence of a degree of dependence of the results on the particle numbers [119]). …”