Supersonic Combustion and Detonation

Stochastic Spectra Contents of Detonation Initiated by Compressible Turbulence Thermodynamic Fluctuations
Ladeinde, F. and Oh, H., Physics of Fluids 33, 045111 (2021);
https://doi.org/10.1063/5.0045293

Abstract
A canonical system that contains the basic elements of detonation initiated by compressible turbulence thermodynamic fluctuations was proposed by Towery et al. [“Detonation initiation by compressible turbulence thermodynamic fluctuations,” Combust. Flame 213, 172–183 (2020)] and successfully tested using direct numerical simulation (DNS). In the present study, the DNS dataset is used to assess the applicability of a few compressible turbulence theories to the problem of detonation and to investigate some statistical aspects of the problem and the spectra of the turbulence kinetic energy (TKE) and reactive scalar fields within the context of detonating and deflagrating compressible flows. It has been found that scaling of the dilatational and solenoidal components of compressible turbulence as reported in the literature does not apply to the case of detonation or deflagration. Also, the detonating cases were found to have significantly lower Karlovitz numbers compared to the nondetonating ones and to conform to the Bray–Moss–Libby premixed combustion (bimodal probability density function) model for intense turbulence and large Damkhöler number. The nondetonating cases show distributed pdfs. The stronger combustion in the detonating cases shows effects on the spectra of the TKE and reactive scalars, particularly when the wavenumber is scaled with the Kolmogorov length and the Schmidt number.

ladeinde1
Fig. 1. Snapshots of the temperature field (peakHRR) (a) Z1, (b) Z2a, (c) Z2b, (d) Z3a, (e) Z3b

The Effects of Pressure Treatment on the Flamelet Modeling of Supersonic Combustion
Ladeinde, F., Lou, Z. and Li, W., Combustion and Flame 204, pp. 414-429 (2019);
https://doi.org/10.1016/j.combustflame.2019.03.030

Abstract
The flamelet method has been used extensively as an affordable turbulence-combustion interaction model that obviates the need to solve the evolution equations for the species mass fractions during a large-eddy or Reynolds-averaged Navier–Stokes calculation of a reactive flow field, leading to substantial savings in the simulation time and enabling modeling with relatively complex kinetic mechanisms. The canonical problem analyzed and stored in a look-up array in the flamelet procedure usually assumes some baseline fields; in particular, the pressure is often specified at a fixed value that is characteristic of the examined configuration. However, pressure in supersonic combustion has significant dynamical roles, unlike in low-Mach number or incompressible flows, and a constant pressure field will not be adequate for the former. To remedy this problem, reaction rate in the combustor is often assumed to scale squarely with pressure. This approach, which is probably acceptable for low-speed, high pressure combustors, is not suitable for dealing with the variable pressure conditions in supersonic combustion. This paper focuses on the assessment of the aforementioned scaling, in absolute sense, and also relative to an approach where pressure is added as a control parameter in the flamelet library. To achieve this, three classes of reactive systems with different levels of modeling complexities are investigated to show that representative chemical variables do not scale squarely with pressure. For the case of supersonic combustion, the scaling treatment in general leads to over-prediction of pressure and combustion and also tends to stabilize the flame. To the knowledge of the authors, no previous studies have reported on the issues addressed in the present paper.

02-2
Fig. 9. (a) Contour maps (top to bottom): mixture fraction, pressure, temperature, water mass fraction, hydroxyl radical mass fractions and turbulent viscosity for SLP at 6FT. (b) Contour maps (top to bottom): mixture fraction, presssure, temperature, water mass fraction, hydroxyl radical mass fractions and turbulent viscosity for SPS at 6FT.