An Investigation of Axisymmetric Disk Stabilized Propane-Air Flames Operating under Inlet Mixture Preheat and Stratification

ABSTRACT

Turbulent, recirculating, lean propane-air flames with inlet mixture stratification and preheat have been simulated under stable and limiting burning conditions. The modeled burner setup comprises a supply tube enclosing three sequential disks producing two consecutive premixing cavities. Fuel is injected in the first cavity and is partially premixed with primary air flowing through this cavity system, resulting in a radially stratified equivalence ratio profile at the inlet of the afterbody disk flame stabilizer. Detailed velocity, turbulence, fuel-air mixing, and imaging data have been previously reported for inlet preheats from 300 to 573 K and for a range of fuel flow rates. The simulations were carried out with a finite-volume-based method, using the dynamic Smagorinsky subgrid model coupled with two combustion methodologies, a quasi-laminar reaction rate approach and the Thickened Flame Model approach. Propane oxidation was modeled with a 22-species skeletal scheme. OH* chemiluminescence distributions were also computed by post-processing quasi-steady state derived algebraic expressions, exploiting directly simulated species thus enabling comparisons with experimental images. The simulations were evaluated against velocity, turbulence, and temperature measurements as well as chemiluminescence images. These comparisons allowed for an assessment of the methodologies’ capability to reproduce important trends such as the notable extension of stable operation to ultra-lean mixtures, the appreciable effect on the near flame aerodynamic stretch and the variations in the local flame structure.

KEYWORDS:

• Large eddy simulations
• bluff-body flames
• stratified flames
• turbulent propane flames
• preheated mixture flames

Disclosure statement

No potential conflict of interest was reported by the authors(s).

Data Availability Statement

The authors confirm that the data supporting the findings of this study are available within the article [and/or] its supplementary materials.

Radiation effects for lean flammability limit

The 1D adiabatic and radiative freely propagating laminar flame package PREMIX (“ANSYS® Academic Research, Release 2021R1,” 2021) was utilized for the laminar flame speed calculation and the Lean Flammability Limit (LFL) estimation. In order to account for the radiation effects an Optically Thin Model (OTM) was used to simulate the radiative emissions of the species CO₂, CO, H₂O and C₃H₈. The Planck Mean Absorption Coefficients necessary were obtained by Nakamura and Shindo (2019) for the CO₂, CO, and H₂O and by Wakatsuki et al. (2008) for C₃H₈. The same coefficient values were used for the parametric studies of the four kinetic schemes. To estimate the LFL value, the input Equivalence Ratio was reduced in 0.01 increments, starting from a stoichiometric mixture, until a failure to converge was encountered. Then, the Equivalence Ratio was decreased by increments of 0.001 from the last successful run, until it failed to converge. The last successful run’s equivalence ratio value was recorded as the LFL value. Additionally, to study the reactant preheat effect on the LFL, the unburnt reactant temperature and the ambient burner temperature were set in tandem to 300K, 423K, and 573K.

The adiabatic laminar flame speed (S_L,adiab) and the radiative laminar flame speed (S_{L, rad}) are presented in Figure 14 (top) as calculated using the S111 reference scheme. It can be noted that if the radiation effects are neglected, the LFL is underpredicted by 9.1%, 15.8%, and 19.7% for the three preheat values. However, as seen in Figure 14 (bottom), the ratio of radiative to adiabatic laminar flame speed quickly approaches unity as we increase the Equivalence Ratio. For equivalence ratios above 0.6, 0.5, and 0.4 (for the three preheat values) the ratio of radiative to laminar flame speed has approached 0.995.

Figure 14. Computed adiabatic (dashed lines) and radiative flame speeds (solid lines) for the three air preheat values using the S111 scheme (top); Ratio of the radiative to adiabatic flame speeds as computed above (bottom).