湍流甲烷OXY型有氧燃烧的动力学，稳定性和比例效应

Carbon capture and storage (CCS) is an important strategy for reducing CO2 emissions,
with oxy-fuel combustion being one of the most promising technologies because
of it is high efficiency and low cost. In oxy-combustion, CH4/0 2/CO 2 mixtures burn
at low temperatures (~1700 K), high pressures (~40 bar), where laminar burning velocities
are about 7 times lower than in traditional CH4 /Air mixtures. Thus oxy-fuel
combustors are more prone to blowoff and dynamic instabilities. In this thesis we
examine turbulent oxy-combustion flame stabilization physics at the large and small
scales using experimental studies and numerical simulations.
Experimental measurements are used to establish the stability characteristics of
flame macrostructures in a swirl stabilized combustor. We show that the transition
in the flame macrostructure to a flame stabilized along both the inner and outer
shear layers (Flame IV), scales according to the extinction strain rate, similar to air
flames. To achieve accurate scaling, extinction strain rates must be computed at the
thermal conditions of the outer shear layer, emphasizing the role of heat interactions
with the wall boundary layer. Care must be exercised while modeling the chemical
structure of oxy-flames. We show that the kinetics of CO2 (used as a diluent in oxycombustion)
is important in determining the consumption speed and flame extinction
strain rate. Specifically, the extinction strain rate was found to be heavily impacted
by the reaction C02+ H - CO + OH.
Large Eddy Simulations (LES) models, first validated for various combustor geometries,
fuels and oxidizers, are used to examine the stabilization mechanisms of
these flames. First, we demonstrate the importance of choosing the correct global
chemical kinetics mechanism in predicting the flow structures in multi-dimensional
simulations and develop a priori criterion of selecting a reduced mechanism based on
the extinction strain rate. Besides flame macrostructures, recirculation zone lengths
are found to linearly scale with extinction strain rates. This scaling holds regardless
of fuel or oxidizer type, Reynold's number, inlet temperature, or combustor geometry.
It is thus very important that a chemical mechanism is able to correctly predict extinction
strain rates if it is to be used in CFD simulations. We use the validated LES

framework to model the transition to Flame IV in the swirl combustor for methane
oxy-combustion mixtures. The 3D turbulent flame structure strongly resembles a
ID strained adiabatic laminar flame structure in the combustor interior, and nonadiabatic
flames near the combustor wall. The results support the earlier conclusions
regarding the use of the extinction strain rate and the wall thermal boundary condition
in scaling and modeling turbulent combustion dynamics.