Repository logo

Prediction of Global Extinction Conditions and Dynamics in Swirling Non-premixed Flames Using LES/CMC Modelling

Change log


Zhang, H 
Mastorakos, Epaminondas  ORCID logo


The Large Eddy Simulation (LES)/three-dimensional Conditional Moment Closure (CMC) model with detailed chemistry is applied to predict the operating condition and dynamics of complete extinction (blow-off) in swirling non-premixed methane flames. Using model constants previously selected to provide relatively accurate predictions of the degree of local extinction in the piloted jet flames Sandia D−F, the error in the blow-off air velocity predicted by LES/3D-CMC in short, recirculating flames with strong swirl for a range of fuel flow rates is within 25 % of the experimental value, which is considered a new and promising result for combustion LES that has not been applied before for the prediction of the whole blow-off curve in complex geometries. The results also show that during the blow-off transient, the total heat release gradually decreases over a duration that agrees well with experiment. The evolution of localized extinction, reactive scalars and scalar dissipation rate is analyzed. It has been observed that a consistent symptom for flames approaching blow-off is the appearance of high-frequency and high-magnitude fluctuations of the conditionally filtered stoichiometric scalar dissipation rate, resulting in an increased fraction of local extinction over the stoichiometric mixture fraction iso-surfaces. It is also shown that the blow-off time changes with the different blow-off conditions.



Global extinction, Blow-off curve, Swirling non-premixed flames, Large eddy simulation, Conditional moment closure

Journal Title

Flow, Turbulence and Combustion

Conference Name

Journal ISSN


Volume Title


Springer Science and Business Media LLC
Engineering and Physical Sciences Research Council (EP/K025791/1)
This work was financially supported by Engineering and Physical Sciences Research Council (EPSRC) and Rolls-Royce through a Dorothy Hodgkin Postgraduate Award. This work used the computational resources from ARCHER cluster of the UK National Supercomputing Service (http://www. under the project of United Kingdom Consortium on Turbulent Reacting Flows (UKCTRF) and the Darwin Supercomputer of the University of Cambridge High Performance Computing Service (http://