Computational Fluid Dynamics modelling of radiative heat transfer of an in-house helium stabilized, laminar premixed flat flame

Abstract

The current work describes a multi-dimensional Computational Fluid Dynamics study of an in-house helium stabilized, laminar premixed flat flame, with an emphasis on the radiation modelling. The experimental work first deals with the post flame instability induced by the difference of gas densities at the downstream and upstream of the flat flame. A non-intrusive method which is addressed as helium stabilization is developed, where helium enters the chamber as a co-flow jet to dilute the combustion products and hence minimize the difference of gas densities between downstream and upstream of the flame. A thermometry method based on infrared emission and absorption by carbon dioxide between 2100 and 2400 cm-1 is then applied for temperature measurement. In the numerical efforts, a sooting flame with an equivalence ratio value of 2.15 is simulated. The associated experimental temperature and soot volume fraction profiles at different heights along the axial direction are used to validate a local time stepping (LTS) based solver. Radiation of the both gas and soot is taken into account. For the gas phase species, a Wide Band Box model is incorporated into the solver to account for thermal radiation from CO2 and H2O and the model parameters are calculated using HITRAN2012 radiation database. On the other hand, the soot absorption coefficient is calculated using the relation based on the associated soot volume fraction and local cell temperature. Results generated using the Weighted Sum of Gray Gases (WSGG) model are also included for comparison purposes. The temperature profile calculated using the Wide Band Box model is found closer to the measurements, as compared to those predicted using the WSGG model. The soot volume fraction calculated by the former are also closer to the measurement for the first 30 mm from the burner exit but both models overestimate the soot volume fraction at a further downstream location. The current results showed that the LTS solver predict the temperature and soot volume fraction of the stabilized, laminar premixed flames reasonably well. The solver can be used to examine more comprehensive yet computational expensive radiation submodels.