| |
By Antonio Dicorato, Eliseo Covelli, Alessio Frassoldati, Tiziano Faravelli, and Eliseo Ranzi,
Dipartimento di Chimica, Materiali e Ingegneria Chimica, Politecnico di Milano, Italy
View the pdf of this Supplement
Over the years, CFD has been widely used for understanding and
optimizing the design of the combustion process in many systems,
such as gas turbines and industrial or domestic heating
devices. At Politecnico di Milano, researchers have investigated turbulent
non-premixed flames and the environmental impact of pilot burners,
with a special focus on flame stability and pollutant formation. In one
hydrogen/methane/air flame studied, the presence of hydrogen was
found to improve flame stability because higher concentrations of H, O,
and OH radicals increase several key reaction rates. The presence of H
and OH also result in reduced pollutant emissions. Lean fuel/air ratios
lead to a significant shift in flame blow-out conditions, a lower flame
temperature, and reduced NOx content. For turbulent jet diffusion
flames studied, high jet speed and increased turbulence have been
found to increase the mixing and/or ignition of the fuel/air mixture, leading
to more stable flames. Turbulence alone, however, is not enough to
ensure a stable flame. Recirculation, which can be achieved by inserting
a bluff body into an oncoming stream, can be used to increase flame stability
even more.
Comparison between flame image and temperature field calculated with FLUENT
In a recent project, a stable burner with a reduced emissions profile
has been studied. The burner is a bluff-body type, which has been presented
as a target case at TNF Workshops [1] and for which data are
freely accessible on the web [2]. The body produces a low pressure
region behind it that entrains gas in a recirculating flow pattern. Fuel is
injected into this zone, and is continually re-ignited, sustaining combustion.
Behind the recirculation zone there is a neck zone, where the velocity
is high and the flame operates close to extinction. Finally, combustion
is completed in the flame tail.
Contours of axial velocity showing the strucuture of the recirculation zone
Comparison between experimental measurements (points) and model predictions (lines) 30 mm from the inlet
The combustion process, including pollutant formation, was modeled
using the eddy dissipation concept (EDC) model [3] in FLUENT 6.1.
Combustion modeling is challenging because of the strong coupling
between the chemical reaction and fluid mixing processes. Often, the
rates at which these processes occur differ. Large spatial and temporal
variations in species composition and temperature add to the difficulty.
The EDC model addresses these issues by simulating the structure of turbulent
flames using detailed chemistry, so that the formation of pollutants
and other finite-rate effects can be accounted for. According to this
model, the chemical reactions are assumed to occur in the fine structures,
or small turbulent scales, which are treated as perfectly stirred
reactors (PSRs). By treating the reacting fine structures locally as PSRs, a
detailed chemical kinetic mechanism can be linked with turbulent combustion
modeling. According to the EDC model, chemical reactions are
quenched if the characteristic chemical times are longer than the residence
time within the fine structures.
To simulate the bluff-body flame, an unstructured, axisymmetric grid
of 35,000 cells and a RANS turbulence approach were used. The kinetic
model used for methane and hydrogen combustion included 30 species
and 110 reactions [4]. Results for temperature, velocity, and composition,
particularly CO and CO2, were found to be in very good agreement
with measurements.
References:
- www.ca.sandia.gov/tnf.
- www.aeromech.usyd.edu.au/thermofluids.
- B.F. Magnussen, 19th AIAA Aerospace Science Meeting, St. Louis, Missouri, (1981).
- T. Peeters, PhD thesis, Delft Technical University, Delft, The Netherlands, (1995).
|
|
|
