Tutorial: 2D Simulation of A 300 KW BERL Combustor Using The Magnussen Model
Tutorial: 2D Simulation of A 300 KW BERL Combustor Using The Magnussen Model
Tutorial: 2D Simulation of A 300 KW BERL Combustor Using The Magnussen Model
Introduction
The purpose of this tutorial is to provide guidelines and recommendations for setting up
and solving a natural gas combustion problem.
This tutorial demonstrates how to do the following:
Prerequisites
This tutorial is written with the assumption that you have completed Tutorial 1 from the
ANSYS FLUENT 14.5 Tutorial Guide, and that you are familiar with the ANSYS FLUENT
navigation pane and menu structure. Some steps in the setup and solution procedure will
not be shown explicitly.
If you have not used k-epsilon turbulence, P-1 radiation, and Eddy Dissipation Finite-rate
Reaction models before, it would be helpful to first refer to the ANSYS FLUENT 14.5 User’s
Guide.
Problem Description
This problem was modeled after the experiments carried out at the Burner Engineering
Research Laboratory (BERL) as part of a large project (Scaling 400 study) for combustors
ranging in size from 30 KW to 12 MW. The schematic of the problem is shown in Figure 1.
The flow under study is an unstaged natural gas flame in a 300 KW swirl-stabilized burner.
The furnace is vertically fired. It has an octagonal cross-section with a conical furnace
hood and a cylindrical exhaust duct. The furnace walls can be refractory-lined or water-
cooled. The burner features 24 radial fuel ports and a bluff centerbody. Air is introduced
through an annular inlet and movable swirl blocks are used to impart swirl. Figure 2 shows
a closeup of the burner assuming 2D axisymmetry. Appropriate area adjustments were
made to account for the 2D representation of a 3D problem. It has been ensured that the
c ANSYS, Inc. December 4, 2012 1
2D Simulation of a 300 KW BERL Combustor
cross-sectional areas of the model and real furnaces are the same. The input conditions for
this case, i.e. wall temperature, inlet boundary conditions, and profile have been derived
from this experimental data.
Preparation
2
c ANSYS, Inc. December 4, 2012
2D Simulation of a 300 KW BERL Combustor
Figure 3: Mesh
General
c ANSYS, Inc. December 4, 2012 3
2D Simulation of a 300 KW BERL Combustor
5. Enable Gravity.
Step 3: Models
Models
4
c ANSYS, Inc. December 4, 2012
2D Simulation of a 300 KW BERL Combustor
Step 4: Materials
Materials
c ANSYS, Inc. December 4, 2012 5
2D Simulation of a 300 KW BERL Combustor
(a) Click on Edit... next to Mixture Species and reorder the species as follows:
i. h2o
ii. o2
iii. fuel
iv. co2
v. n2
A transport equation is not solved for the last species in the list, instead its
concentration is determined by difference. To reduce the round-off error, the
species of the greatest quantity should be placed last in the list. In most cases,
this is n2.
(b) Click on Edit... next to Reaction and define the following reaction.
4. Enter 16.313 for Molecular Weight and -1.0629e+08 Standard State Enthalpy for fuel
under Material Type mixture.
5. Use the following TUI command to change the specific heat of the species included in
the mixture.
(set-ifrf-cp-polynomials ’mixture-template)
6. Select polynomial from the Cp drop-down list for fuel under Material Type mixture and
set the values for the coefficients as shown below:
Species 1 2 3 4 5
fuel 2005 -0.3407 2.362e-03 -1.178e-6 1.703e-10
6
c ANSYS, Inc. December 4, 2012
2D Simulation of a 300 KW BERL Combustor
Note: Global reaction mechanisms with one or two steps inevitably neglect the inter-
mediate species. In high-temperature flames, neglecting these dissociated species
may cause the temperature to be over-predicted. A more realistic temperature
field can be obtained by increasing the specific heat capacity for each species.
Above command sets the Cp polynomial coefficient to that mentioned by Peter
and Weber (1995). Fuel named as fuel in this tutorial, is not a standard species.
Therefore, Cp polynomial coefficients for fuel need to be specified manually.
Boundary Conditions
c ANSYS, Inc. December 4, 2012 7
2D Simulation of a 300 KW BERL Combustor
(a) Select Components from the Velocity Specification Method drop-down list.
(b) Select vel-prof u and vel-prof w for Axial-Velocity and Swirl-Velocity respectively.
(c) Select Intensity and Length Scale from the Specification Method drop-down list.
(d) Enter 17 % and 0.0076 m for Turbulence Intensity and Turbulence Length Scale
respectively.
(e) Click the Thermal tab and enter 312 K for Temperature.
(f) Click the Species tab and enter 0.2315 for Species Mass Fractions for o2.
(g) Click OK to close the Velocity Inlet dialog box.
8
c ANSYS, Inc. December 4, 2012
2D Simulation of a 300 KW BERL Combustor
(a) Select Components from the Velocity Specification Method drop-down list.
(b) Enter 157.25 m/s for Radial Velocity.
(c) Select Intensity and Length Scale from the Specification Method.
(d) Retain 5 % for Turbulence Intensity.
(e) Enter 0.0009 m for Turbulence Length Scale.
(f) Click theThermal tab and enter 308 K for Temperature.
(g) Click the Species tab and enter 0.97 and 0.008 for Species Mass Fractions for
fuel and co2 respectively.
(h) Click OK to close the Velocity Inlet dialog box.
c ANSYS, Inc. December 4, 2012 9
2D Simulation of a 300 KW BERL Combustor
(a) Select Select Intensity and Hydraulic Diameter from the Specification Method.
(b) Retain 5 % for Backflow Turbulent Intensity.
(c) Enter 0.6 m for Backflow Hydraulic Diameter.
(d) Click the Thermal tab and enter 1300 K for Temperature.
(e) Click the Species tab and enter 0.2315 for Species Mass Fractions for o2.
(f) Click OK to close the Pressure Outlet dialog box.
10
c ANSYS, Inc. December 4, 2012
2D Simulation of a 300 KW BERL Combustor
(a) Click the Thermal tab and select Temperature from the Thermal Conditions list.
(b) Set the following conditions:
Step 6: Solution
c ANSYS, Inc. December 4, 2012 11
2D Simulation of a 300 KW BERL Combustor
(b) Select PRESTO! from the Pressure drop-down list in the Spatial Discretization
group box.
This is often useful for buoyant flows where velocity vectors near walls may not
align with the wall due to assumption of uniform pressure in the boundary layer.
Thus, PRESTO! can only be used with quadrilateral or hexahedral meshes.
(c) Enable Pseudo Transient.
2. Deselect P1 from the Equations selection list.
Solution Controls −→ Equations...
3. Enable Set All Species URFs Together.
Solution Controls
4. Change the time scale factor for species and energy.
Solution Controls −→ Advanced
12
c ANSYS, Inc. December 4, 2012
2D Simulation of a 300 KW BERL Combustor
6. Save the initialized case and data files, berl-mag-init.cas.gz and berl-mag-init.dat.gz.
File −→ Write −→Case & Data
c ANSYS, Inc. December 4, 2012 13
2D Simulation of a 300 KW BERL Combustor
11. Save the case and data files, berl-mag-2.cas.gz and berl-mag-2.dat.gz.
File −→ Write −→Case & Data
12. Enter 1 for Time Scale Factor of Turbulent Kinetic Energy and Dissipation Rate.
Solution Controls −→ Advanced
14
c ANSYS, Inc. December 4, 2012
2D Simulation of a 300 KW BERL Combustor
16. Save the case and data files, berl-mag-3.cas.gz and berl-mag-3.dat.gz.
File −→ Write −→Case & Data
17. Compute the gas phase mass fluxes through all the boundaries.
Reports −→ Fluxes −→ Set Up...
18. Compute the gas phase energy fluxes through all the boundaries.
(a) Select Total Heat Transfer Rate from the Options list.
(b) Select all the zones from the Boundaries selection list and click Compute.
c ANSYS, Inc. December 4, 2012 15
2D Simulation of a 300 KW BERL Combustor
16
c ANSYS, Inc. December 4, 2012
2D Simulation of a 300 KW BERL Combustor
Results
Use of the DO radiation model, which is more CPU intensive, and also a second order
solution, can help to increase the accuracy of the predictions.
Summary
Inherent limitations in the available models result in inaccuracies while predicting interme-
diate species. Overall, fairly meaningful results within engineering accuracy are obtained.
References
A. A. A. Peters and R. Weber ”Mathematical Modeling of a 2.25 MWt Swirling Natural
Gas Flame. Part1: Eddy Break-up Concept for Turbulent Combustion, Probability Density
Function Approach for Nitric Oxide Formation.” Combustion Science and Technology. 110.
67-101. 1995
c ANSYS, Inc. December 4, 2012 17