Numerical Study of Premixed Flames Parametric Acoustic Instability

Abstract

Numerical simulations of flames unstable, due to a parametric acoustic forcing, are performed in this paper. Results concerning the threshold of this instability are compared to quasi-steady theoretical predictions. Important disagreements are found at large frequencies. suggesting that the Markstein lengths relative to curvature and stretch are frequency dependent. Results are discussed in connection to experiments and to new theoretical ideas.

Forcing of a flame by a periodic flow in a Hele-Shaw burner

“…Their measurements seem to corroborate Joulin's theory: at high frequency, all the flames seem to respond in the same way, independently of their chemical composition. Similar observations were obtained in [14] with threshold measurements obtained from DNS.…”

Spray-Flame Dynamics in a Rich Droplet Array

“…Three snapshots of the centered 2D-lattice soon after ignition at the right-hand side, a) Fuel mass fraction, a) Oxygen mass fraction c) Field of production term (in color) [flame propagates from right to left].system is coupled with the Navier-Stokes equations. The overall scheme that computes the Navier-Stokes equations has previously been described in Denet and Haldenwang[13].3.1 Results: Spray-Flame spreadingThe numerical simulations have been carried out using the following parameters: the Lewis numbers of fuel, oxygen and nitrogen are respectively Le F = 1.8, Le O = 0.9, Le N = 1. Let us recall that the overall equivalence ratio ϕ T is varied in the range [0.9, 2] .The high-level of the non-homogeneity in a spray-flame and the complexity of the spreading are illustrated in figures 1 , 2 , 3 and 4, which will be commented below.The numerical simulations concern 2D sprays where the droplets are conceived as very dense pockets of pure fuel, initially located at the node of a face-centered lat-tice.…”

Low Vorticity and Small Gas Expansion in Premixed Flames

“…(28) is a well-known DL dispersion relation. Let us note, however, that including curvature effects in this type of equation may be nontrivial, because the Markstein lengths are actually frequency dependent (see Clavin and Joulin, 1997;Denet and Toma, 1995;Joulin, 1994). The unknown coefficient C 1 may be adjusted by using direct numerical simulations of curved stationary flames.…”