Determination of Flame Plume Characteristics utilising CFD and
Experimental Approaches
NEIHAD HUSSEN AL-KHALIDY
CFD, Wind and Energy Technical Discipline,
SLR Consulting,
202 Submarine School, Sub Base Platypus, North Sydney,
AUSTRALIA
Abstract: - The potential for plumes to affect the safety of aircraft operations is often predicted using MITRE
EPA Models. For many projects, key input parameters to MITER EPA are not available and conservative
assumptions or models such as OHIO model are used to characterize the combustion and approximate the key
input parameters to EPA plume rise model. These assumptions and conservative models lead to inaccurate
results of simulations. The current study provides a novel approach to use a combination of Computational
Fluid Dynamics (CFD) tool and EPA Models to reliably predict the risk of turbulence and upset being
encountered by a range of aircraft types that operate through a rising plume.
The main objective of the current study is to develop a Computational Fluid Dynamics (CFD) combustion
model and a procedure to determine an improved set of flare inputs for the Air Quality (AQ) and MITER EPA
models. A CFD model has been developed to determine flame plume characteristics (Effective Height,
Effective Diameter, Temperature and Velocity) from two flare stacks which are part of a trailer-mounted
Mobile Purge Burner (MPB) system.
A subsequent experimental test of a similar trailer-mounted MPB system has validated the CFD results.
Plume temperatures within the combustion zone of the flares were very much in line with the temperatures
predicted by the CFD Simulation study. Plume temperatures above the MPB System appear to drop very
quickly, such that the plume temperature fell from just under 500°C at 5 m above ground level to around 15-
16°C (and close to the ambient temperature) at 22 m above ground. Again, this is consistent with the CFD
Study results.
The CFD simulations in the current study accounted for the turbulent flow with chemical species mixing
and reaction and utilised an advanced radiation model to solve participating radiation in the combusted zones.
This study assesses all the parameters that have impact on the accuracy of the numerical model including
computational domain, mesh distribution, numerical scheme and flame plume characteristics including ambient
conditions (wind speed and temperature) and combustion under various air to fuel ratio scenarios.
Key-Words: - CFD, Combustion, Flame Plume Characteristics, Complex Radiation, Modelling Inputs to
MITER EPA
Received: December 26, 2022. Revised: April 14, 2023. Accepted: May 15, 2023. Published: June 6, 2023.
1 Introduction
Exhaust plumes that originate from power stations,
cooling towers, industrial facilities such as smelters,
and flaring associated with the de-pressurisation of
gas systems, can result in a potential hazard to
aircraft operations because of the velocity,
turbulence, and/or location of the associated gas
plume. Plumes can affect the handling characteristics
of aircraft, create the potential for aircraft stalling or
rolling and, in extreme circumstances, cause airframe
damage.
Observations of plume trajectories from a
500 MW power station have been used to test
the ability of a simple parameterization of
entrainment to describe the rising plume in [1].
As the government body that regulates Civil
Australian Aviation Safety (CASA) and the
operation of Australian aircraft overseas, CASA
is required by legislation to assess the potential
for plumes to affect the safety of aircraft
operations pursuant to CASA 139.370, [2]. The
Advisory Circular sets out the procedure for
conducting the assessment of a proposal that
will create a plume rise. This methodology
relies on a plume rise model originally
developed by Commonwealth Scientific
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
533
Volume 19, 2023
Industrial Research Organisation (CSIRO) to
model air pollution called TAPM (The Air
Pollution Model). TAPM is a PC-based,
nestable, prognostic meteorological and air
pollution model (with photochemistry) driven
by a Graphical User Interface and is a viable
tool for year-long simulations, [3], [4].
In 2012, the United States Federal Aviation
Administration (FAA) commissioned the
MITRE Corporation, [5], to develop an exhaust
plume analyses model that would predict the
risk of turbulence and upset being encountered
by a range of aircraft types that operate through
a rising plume. The outcome of this work was
the MITRE Exhaust Plume Analyzer (EPA)
model. In September 2015, the US FAA
formally adopted the MITRE EPA model as the
assessment tool for assessing the impact of
plume rise sources on aircraft safety.
The MITRE EPA model, [5], uses a
combination of models to simulate the plume
and aircraft behaviour; including a convective
flow model describing the mean flow of the
plume, a turbulence model computing the
probability of experiencing a gust capable of
causing severe turbulence or aircraft upset and
aircraft response models judging the required
vertical gust to achieve severe turbulence or
aircraft upset. The key input parameters for flare
plume modelling are:
Effective Height of the Plume at the
Equivalent Exhaust Emission Point Heff (m)
Effective Diameter of the Plume at the
Equivalent Exhaust Emission Point Diaeff (m)
Velocity of the Plume at the Equivalent
Exhaust Emission Point Vp (m/s)
Temperature of the Plume at the Equivalent
Exhaust Emission Point Tp (oC)
For many projects, measurement data for the
above key input parameters are not available and
conservative assumptions or models (e.g., OHIO
EPA model) are used to characterize the
combustion source in terms of an “equivalent”
stack and approximate the above parameters.
The current study presents Computational
Fluid Dynamics (CFD) as a tool to determine an
improved set of flare inputs for the EPA models in
lieu of using conservative and unrealistic
assumptions and models. The main elements of this
study are:
Assess the fuel combustion. The gas
composition used in this study consists of
Methane 88.8%, CO2 1.9%, N2 1.5%, Ethane
7.8% and Propane 0.2%.
Determine the fuel rate in the system.
Model the combustion process to estimate the
chemical reacting species and associated energy
due to the chemical reaction, mixture speed in
three directions, pressure profile, temperature,
and turbulence parameters.
Predict the participating radiation during the
combustion process.
Predict flare characteristics for various
atmospheric conditions (eg. near calm wind,
light breeze, etc.).
Assess the impact of the Air to Fuel ratio on
the flame plume.
Determine the flame plume characteristics
(Effective Height, Diameter, Temperature and
Velocity).
2 Problem Description
The use of Mobile Purge Burner (MPB) Trailer to
burn off residual gas to depressurise redundant main
and support commissioning operations is necessary
to avoid release of natural gas to atmosphere
(environmental regulations) and prevent a natural
gas plume gathering in airspace.
The assessed MPB Trailer in the current
study consists of two stacks (Refer Figure 1).
The trailer system is designed for 1050 kPag
with a step-down regulator set between 480
kPag to 1030 kPag.
A trailer inlet pressure of 1030 kPa is
adopted for the flow rate calculation as this will
result in a higher flow rate.
Fig. 1: Mobile Purge Unit
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
534
Volume 19, 2023
3 Problem Formulation
The CFD model solves the continuity and
momentum, energy and chemical species equations.
The equations for a steady state case can be written
as follows, [6]:
Where ρ is the density of air, u is the air velocity, p
is the static pressure, ρg and F are the gravitational
body and external body forces, Ʈij is the stress
tensor, h is the enthalpy, keffective is the effective
thermal conductivity and S is the volumetric heat
source. Ji is diffusion flux of specie i, Ri is the net
rate of production of species i by chemical reaction
and Si is the rate of creation reaction by addition
from the dispersed phase plus any user-defined
sources.
The local mass fraction of each species Yi is
predicted through the solution of a convection-
diffusion equation for the ith species, [6].
4 CFD Simulation Methodology
4.1 Geometry for CFD Modelling
A detailed 3D model of the MPB trailer was created
based on the provided engineering drawings. The
overall model is shown in Figure 2.
The CFD model includes all the flare features
as per the given engineering drawings.
o The main pipe is 50 NB x 603
Schedule 45 Pipe.
o The flare tip diameter is 165 mm.
o The flare height is 3.1 m above
local ground level.
Air is introduced through an annular inlet. The
air disc intake is modelled as per the provided
layout drawings.
The CFD model also accounts for the impact
of the approaching winds on the flare
characteristics.
4.2 Boundary Conditions and Properties
4.2.1 Fuel Composition and Operating
Conditions
The gas composition used in the MPB consists of
Methane 88.8%, CO2 1.9%, N2 1.5%, Ethane 7.8%
and Propane 0.2%.
For the current study it is conservatively
assumed that the fuel consists of 100% methane.
Methane is more flammable than ethane.
The fuel rate in the system has been predicted
in a previous study using HYSYS software. The
study found the following:
The total flare flow rate (from both stacks)
will be limited by the pressure regulator located
on the trailer.
The regulator is a DN25 Fisher 627 with
1/8"orifice and downstream pressure setpoint of
210 kPag. The minimum gas temperature of
10oC was adopted as this gives the worst case
(i.e., largest flowrate through the Fisher 627
regulator).
The trailer system is designed for 1050 kPag
with a step-down regulator set between 480
kPag to 1030 kPag. A trailer inlet pressure of
1030 kPag was adopted for the flow rate
calculation as this will result in a higher flow
rate.
The fuel mass flow rate in the system is
0.0159 kg/s. The flow is equally split through
the two stacks, i.e., 0.00799 kg/s through each
flare stack.
The fuel temperature is 5oC pre-combustion.
4.2.2 Air Intake
The air disc (Refer to Figure 2) can be adjusted. In
the current study, two scenarios were analysed:
Scenario AFI-1: Limited air flow intake to
complete the chemical reaction.
Scenario AFI-2: Increased air flow intake to
accelerate the combustion process.
4.2.3 Wind and Ambient Temperature
Conditions
The CFD results are presented for the following
worst-case scenarios:
Scenario 1: Near Calm Wind Condition (Wind
Speed = 0.1 m/s)
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
535
Volume 19, 2023
Scenario 2: Very light Breeze with an Average
Wind Speed of 0.28 m/s (1 km/hr)
The impact of the atmospheric air temperature
has also been assessed in the current study. The
computational domain for CFD modelling is shown
in Figure 3.
Fig. 2: 3D Geometry for CFD Modelling
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
536
Volume 19, 2023
Fig. 3: Computational Domain for CFD Modelling
4.3 Combustion Modelling
Natural gas combustion is a complex
thermodynamic process and involves a large
number of interacting processes, including
turbulent flow, gas phase chemical reactions, and
heat transfer.
A generalized Eddy-Dissipation Model
(EDM), [6], was used to analyse the methane-air
combustion system. The EDM model is suitable for
chemical reactions where fuels burn quickly, and
the overall rate of reaction is controlled by
turbulent mixing.
The combustion is modelled using a global
chemical reaction mechanism, assuming complete
conversion of the CH4 to H2O and CO2 plus heat.
This reaction is defined in terms of
stoichiometric coefficients, formation enthalpies,
and parameters that control the reaction rate. The
rate at which the reaction proceeds is directly
proportional to the concentrations of the two
reactant species.
The EDM models have been implemented in
major commercial CFD software such as ANSYS-
Fluent, [6], and the model has been used in the
CFD simulation of a variety of different
combustion systems, [7], [8], [9], [10], [11], and,
importantly, validated against experimental data.
The air-fuel mixing and transport of chemical
species are modelled by solving conservation
equations describing convection, diffusion, and
reaction sources for each component species with
reactions occurring in the fluid phase (volumetric
reactions).
The effects of radiation are also examined in
the current study using the Discrete Ordinates (DO)
Radiation Model, [12]. The DO model provides the
ability to solve problems ranging from surface-to-
surface radiation to participating radiation in
combustion problems.
4.4 Discretization
The software package used in the current CFD
analysis is the commercially available code Fluent.
The CFD model solves continuity, momentum,
energy and transport species equations to predict
the combustion and fluid flow characteristics in the
computational domain.
The quality of the mesh is a critical aspect of
the overall numerical simulation, and it has a
significant impact on the accuracy of the results
and solver run time. A mesh sensitivity
assessment has been carried out for the current
study. A procedure was developed to adopt a
very fine mesh at the areas of interest, i.e., in
the immediate area around the two flares.
Initial runs were conducted using
approximately 10 million mixed cells. The
mesh was then optimized using polyhedral
elements.
o For the current analysis, polyhedral
elements with a total 10,915,033 nodes
have been used to cover the
computational domain.
o Polyhedral cells are especially beneficial
for handling complex flows and used to
provide even more accurate results than
with a hexahedral mesh. For a
hexahedral cell, there are three optimal
flow directions which lead to the
maximum accuracy while for a
polyhedron with twelve faces there are
six optimal directions which, together
with the larger number of neighbours,
lead to a more accurate solution with a
lower cell count, [13], [14].
o In general, very fine meshes are used for
the` entire computational domain and the
maximum Y+ in the computational
domain is 52. An example of the final
mesh scheme is shown in Figure 4.
The Eddy-Dissipation Model (EDM) was used
to model the combustion process and analyze
the methane-air combustion system.
A Realizable k-ɛ turbulence model was
adopted in the current study.
The Discrete Ordinates (DO) Radiation Model
was used to predict the participating radiation
during the combustion process.
An iterative procedure was used to estimate
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
537
Volume 19, 2023
the chemical reacting species and associated
energy due to the chemical reaction, mixture
speed in three directions, pressure profile,
temperature and turbulence parameters.
For the pressure velocity coupling, a global
solver based on the COUPLE algorithm was
employed.
Fig. 4: Mesh Density for the Area of Interest
A second order numerical scheme was used for
discretization of pressure and momentum to
obtain more accurate results.
All of the above represent state-of-the-art CFD
modelling techniques for combustion simulations.
The normalised residuals of continuity x-, y-,
and z-velocity, reacting species, energy, DO
intensity, k and epsilon was reduced between five
and seven orders of magnitude demonstrating a
valid numerical solution. Figure 5 shows that the
normalised residuals for all variables.
Fig. 5: Scaled Residual History
5 CFD Results and Discussion
5.1 Near Calm Wind Condition
(Approaching Wind = 0.1 m/s)
5.1.1 Scenario AFI-1 – Air to Fuel (Methane)
Ratio = 1.5
The objective of this scenario is to predict the
characteristic of plum during poor combustion
process. Limited air flow intake (0.015 kg/s)
(Scenario “AFI-1”) with an ambient temperature of
15oC was initially used to complete the combustion
process.
The mass fraction of reactants is shown in
Figure 6. The following conclusions can be reached
from Figure 6:
Methane and air are introduced at the base of
the combustion chamber. Refer Figure 6A
The mass fraction of the methane is reduced
gradually due to the methane oxidization
reaction process.
Initial mass fraction of the air is 23% (Figure
6B) as per the given boundary condition.
As the reaction initiates, the oxygen in the air is
consumed inside the chamber and above the tip
of the flare. Refer to Figure 6B
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
538
Volume 19, 2023
Fig. 6: Mass Reaction of Reacting Species
Figure 6C shows the mass fraction variation of
CO2 products. The formation of CO2 continues
to increase as the methane consumption
increases.
Figure 6D shows the mass fraction variation of
H2O products. As the reaction proceeds, the
H2O formation increases and the final mass
fraction of H2O in the mixture is 12.8%.
The predicted mass fractions of H2O and CO2
are in a very good agreement with the data in the
open literature (e.g., [15]).
Figure 7 shows the mean velocity results at a
2D Section through the purge burner. Dark blue
represents still conditions at 0 m/s, red represents
26 m/s.
The methane velocity is increased at the inlet
nozzle to approximately 25 m/s at the base of
the combustion chamber.
The average velocity at the tip of the flare is
reduced to 4.8 m/s due to the combustion
process and venturi shape of the flare.
The maximum velocity at 7.1 m above ground
(area of interest for the Air Quality assessment)
is 0.43 m/s.
Fig. 7: Velocity Magnitude (m/s) - Near Calm
Wind Condition, Approaching Wind =0.1 m/s
Figure 8 with contours of static temperature
indicates the following:
The predicted maximum flame temperature is
1700oC, occurring at approximately 1 m above
the flare tip position.
The mass-weighted average temperature at the
tip of the flare is 910oC. The mass-weighted
average is computed by dividing the summation
of the value of the temperature multiplied by
the absolute value of the dot product of the
facet area and momentum vectors by the
summation of the absolute value of the dot
product of the facet area and momentum
vectors:
A: Mass Fraction of Methane
B: Mass Fraction of Oxygen
C: Mass Fraction of CO2
D: Mass Fraction of H2O
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
539
Volume 19, 2023
This parameter is introduced to assess the
impact of various operating conditions on the
temperature profile at the tip of the flare.
The predicted peak temperature at 7.1 m above
ground is 480oC. Refer to Figure 8B
The temperatures above the MPB System
appear to drop very quickly
A plume iso-surface results (Temperature and
Velocity) indicating the approximate three-
dimensional spread of plume with an upstream
ambient wind speed of 0.1 m/s is shown in Figure
9.
5.1.1.1 Impact of Atmospheric (Ambient) Air
Temperature
The impact of the atmospheric air temperature is
shown in Figure 10 and Figure 11. The following
conclusions can be reached from the above figures:
When the outside air temperature was reduced
to 5oC (with no changes to other boundary
conditions), the mass weighted average
temperature at the tip of the flare increased
slightly to 928oC. The predicted plume
temperature at 7.1 m above ground also
increased minimally to 481oC (Refer Figure
11B).
When the outside air temperature was increased
to 35oC (with no changes to other boundary
conditions), the mass weighted average
temperature at the tip of the flare was 880oC.
The predicted plume temperature at 7.1 m
above ground was slightly lower at 470oC
(Refer Figure 12B).
From the above, a lower atmospheric ambient
temperature leads to higher static temperature at the
tip of the flare and slightly higher plume
temperature at the area of interest (7.1 m above
ground).
5.1.1.2 Scenario AFI-2 – Air to Fuel (Methane)
Ratio = 6
For the AFI-2 scenario, the amount of entrained air
in the system was increased by 400% without
increasing the fuel mass flow rate. The results of
the simulations are presented in Figure 12 and
Figure 13. The following major conclusions can be
reached from the above figures:
The predicted peak velocity at 7.1 m above
ground increases to 1.4 m/s. Refer Figure 13A
The predicted peak static temperature at 7.1 m
above ground increases to 620oC. Refer to
Figure 13B
The predicted plume equivalent diameter for
the high temperature zone is approximately
1 m.
The following set of input parameters is
obtained from the above CFD simulation and used
for MITRE EPA simulations:
Heff = 7.1 m
Diaeff = 1.0 m
Vp = 1.4 m/s
Tp = 620°C
The MITRE EPA simulations are not subject to
the current study. The following high-level
comments are made based on the simulation:
Based on the CFD results, the MITRE EPA
results showed that the proposed flaring would
comply with the MITRE EPA aviation safety
thresholds (for critical vertical velocity and
critical turbulence respectively) for all category
aircraft, including Light GA and Light Sport
(which covers helicopter operations).
Based on the OHIA model. The MITRE EPA
results showed that the proposed flaring would
not comply with the MITRE EPA aviation
safety thresholds. The OHIO EPA Model is
standard industry practice in the US Gas
Industry for the characterisation of a
combustion source in terms of an “equivalent”
stack; it is also commonly used globally.
The static temperature at the area of interest is
reduced when the approaching wind is increased to
0.28 m/s (1 km/hr) due to wind interaction with the
flame. Refer Figure 14.
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
540
Volume 19, 2023
Fig. 8: Static Temperature (oC) - Atmospheric Temperature = 15oC (Near Calm Wind Condition, Approaching
Wind =0.1 m/s)
T=1,700oC
7.1 m above ground
T=480oC
Temperature > 550oC in the scoured
region
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
541
Volume 19, 2023
Fig. 9: Iso Surface of Velocity Magnitude (m/s) and Static Temperature (oC) - Near Calm Wind Condition,
Approaching Wind =0.1 m/s
A: Static Temperature (oC)
B: Velocity Magnitudes (m/s)
1000oC
480oC
1 m/s
0.45 m/s
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
542
Volume 19, 2023
Fig. 10: Static Temperature (oC) - Atmospheric Temperature = 15oC (Near Calm Wind Condition, Approaching
Wind =0.1 m/s)
A: On 0 – 1 m/s Scale
B: Static Temperature on 0 – 550oC Scale
T=481oC
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
543
Volume 19, 2023
Fig. 11: Static Temperature (oC) - Atmospheric Temperature = 35oC (Near Calm Wind Condition, Approaching
Wind =0.1 m/s)
A: Velocity Magnitudes on 0 – 1 m/s Scale
B: Static Temperature on 0 – 550oC Scale
V=0.40 m/s
7.1 m above ground
T=470oC
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
544
Volume 19, 2023
Fig. 12: Contours of Plume Velocity (m/s) and Temperature (°C) – Atmospheric Temperature = 5°C (Near
Calm Wind Condition, Approaching Wind =0.1 m/s)
A: Velocity Magnitudes on 0 – 1.4 m/s Scale
B: Static Temperature on 0 – 625oC Scale
V=1.4 m/s
7.1 m above ground
T=620oC
On 0 – 1500oC
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
545
Volume 19, 2023
Fig. 13: Contours of Velocity Magnitude (m/s) and Plume Temperature (oC) at 7.1 m above Ground (Near Calm
Wind Condition, Approaching Wind =0.1 m/s)
A: Velocity Magnitudes on 0 – 1.4 m/s Scale
B: Static Temperature on 0 – 625oC Scale
1.4 m/s
Equivalent diameter of
the high temperature
zone (620oC) is ~ 1 m
Projection of the
Flare Tips
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
546
Volume 19, 2023
Fig. 14: Contours of Plume Velocity (m/s) and Temperature (°C) – Atmospheric Temperature = 15°C (Very
Light Breeze Wind Condition, Approaching Wind =0.28 m/s)
A: Velocity Magnitude (oC)
B: Velocity Magnitudes (m/s)
2m/s
550-639oC
15oC
7.1 m above ground
5 m above ground
7.1 m above ground
5 m above ground
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
547
Volume 19, 2023
6 In-Situ Measurements and
Validation of CFD Results
Subsequently , a test was carried out using a
similar Mobile Purge Burner (MPB) unit. The
distance between the flare centrelines in the tested
unit was 700 mm higher than that in the CFD
model.
The test environment cannot be controlled
precisely due to change to the mean and gust wind
speeds.
6.1 Test Conditions
The test was undertaken under the following test
conditions:
Test Date: Wednesday, 16 February 2022
Test Time & Duration: 9:30 am / ~15 minutes.
Ambient Temperature: 21.6°C (recorded
at the nearby Bureau of Met Weather Station)
Ambient Mean Velocity: 4 km/h (recorded
at the nearby Bureau of Met Weather Station).
Lower wind speeds are anticipated at the test
location due to shielding from suburban
development.
Rain: NIL
6.2 Measurement Instrument
Infrared Vido Camera was used to measure plume
details at the areas of interest. The Video Camera
was placed at a distance of 50 m from the MPB
system. Spot Readings were taken at a number of
locations above the MPB flared tubes, ranging
from 5 m above ground to 22 m above ground.
6.3 Key Output
A key output test from the above trial has been
reproduced in Figure 15.
A - Snapshot from Video Recording 3
B – Time Recording of Peak Flame Temperatures
The chart below shows the mean peak readings (of two flares) recorded with the captured footage clip (~1min).
Mean readings typically range between 440°C and 500°C.
Extremes range from below 400°C up to just below 600°C
SP9
SP10
SP10
SP6
SP7
SP8
SP3
SP4
SP5
SP2
SP1
Fig. 15: Video Output 3 (camera viewed sideways)
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
548
Volume 19, 2023
Key results include the following:
Height = 5 m above ground:
- Temperatures at SP1 & SP2 = 491.0°C
& 353.9°C
Height = 22 m above ground:
- Temperatures at SP9, SP10 & SP11 =
16.2°C, 14.8°C & 15.7°C
Based on the video output from the Field Trial,
the following conclusions were reached:
Plume temperatures within the combustion
zone of the flares were very much in line with
the temperatures predicted by the CFD
Simulation study.
The CFD Plume Rise Model, which adopts
higher plume temperatures and velocities due to
unchanged boundary conditions (e.g., mean
wind speed) than indicated in the test study,
would therefore appear to be moderately
conservative.
Plume temperatures above the MPB System
appear to drop very quickly, such that the
plume temperature fell from just under 500°C
at 5 m above ground level to around 15-16°C
(and close to the ambient temperature) at 22 m
above ground. Again, this is consistent with the
CFD Study results.
4 Conclusion
The risk of turbulence and upset being encountered
by a range of aircraft types that operate through a
rising plume is typically assessed using MITRE EPA
simulations. Key input parameters to MITER EPA
include Effective Height of the Plume, Equivalent
Exhaust Emission Point, Velocity of the Plume at the
Equivalent Exhaust Emission and the Equivalent
Exhaust Emission Point Tp (oC).
It is a common practice to use series of “best
estimate” Scenarios (e.g., OHIO Model) to predict
key input parameters to the MITRE EPA
simulations. The best estimate scenarios may
generate a very conservative input and are not
suitable for all flare systems due to changes to
geometry, number of stacks, combustion
parameters, etc.
The current study presents a CFD as a reliable
tool to determine the flame plume characteristics
(Effective Height, Effective Diameter, Temperature
and Velocity) on the example of trailer-mounted
Mobile Purge Burner (MPB) system for a project
site in Sydney Australia. The assessed MPB
consists of two-flare stacks.
The CFD simulations accounted for the
turbulent flow with chemical species mixing and
reaction and utilised an advanced radiation model to
solve participating radiation in the combusted
zones.
A subsequent experimental test of a similar
trailer mounted MPB system has validated the CFD
results. Plume temperatures within the combustion
zone of the flares were very much in line with the
temperatures predicted by the CFD Simulation
study. Plume temperatures above the MPB System
appear to drop very quickly, such that the plume
temperature fell from just under 500°C at 5 m above
ground level to around 15-16°C (and close to the
ambient temperature) at 22 m above ground. Again,
this is consistent with the CFD Study results.
This study assesses all the parameters that have
impact on the accuracy of the numerical model
including, computational domain, mesh distribution,
numerical scheme and flame plume characteristics
including ambient conditions (wind speed and
temperature) and combustion under various air to
fuel ratio.
The accuracy of the results has been improved
by optimizing the mesh size in the computational
domain via an initial mesh sensitivity analysis, the
use of second-order numerical schemes for the
discretization of pressure and momentum equations,
and the use of a staged approach and powerful
hardware to enable modelling chemical reaction,
plume dispersion, turbulence, and radiation in the
computational domain.
References:
[1] K. Spillane, Observations of Plume
Trajectories in the Initial Momentum
Influenced Phase and Parameterization of
Entrainment, Atmospheric Environment,
Vol.11, No7, pp 1207-1214, 1983.
[2] Australian Government, Civil Aviation
Safety Authority, Plume Rise Assessment”,
Advisory Circular, AC 139.E-0.2 V1.0,
D19/177237, May 2022.
[3] TAPM (The Air Pollution Model), SCIRO,
http://www.csiro.au/Outcomes/Environment
/Population-Sustainability/TAPM.aspx
[4] P. Hurley, P. Manins, et al, Year-long,
High-Resolution Urban Airshed Modelling:
Verification of TAPM Predictions of Smog
and Particles in Melbourne, Australia,
Atmospheric Environment, Vol.37, no.14,
2003, pp 1899-1910.
[5] https://www.mitre.org
[6] Ansys Fluent Theory Manual, ANSYS,
USA, 2022.
[7] D. Gouldy, J. Hopper, Determining the Risk
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
549
Volume 19, 2023
of Experiencing Severe Turbulence when
Flying through and Exhaust Plume,
Environmental Science, Corpus ID
204831699, 2014.
[8] B. Magnussen and B. Hjertager, On
Mathematical Models of Turbulent
Combustion with Special Emphasis on Soot
Formation and Combustion, In 16th Symp.
(Int’l.) on Combustion, The Combustion
Institute, 1976.
[9] H. Gabler, R. Yetter, and I. Glassman,
Symmetric Whirl Combustion: A New
Approach for Non-premixed Low NOx Gas
Turbine Combustor Design. Proceedings of
the 34th AIAA/ASME/SAE/ASEE Joint
Propulsion Conference, Cleveland, USA,
1998.
[10] S. Kim, K. Ito, D. Yoshihara and T.
Wakisaka, Application of a Genetic
Algorithm to the Optimization of Rate
Constants in Chemical Kinetic Models for
Combustion Simulation of HCCI engines,
JSME Int. J. Ser. B, Vol.48, No.717, 2005.
[11] P. Wang, The Model Constant A of the
Eddy Dissipation Model, Progress in
Computational Fluid Dynamics
International Journal, Vol.16, No.2, 2016.
[12] E. hui and G. Raithby. Computation of
Radiant Heat Transfer on a Non-Orthogonal
Mesh Using the Finite-Volume Method,
Numerical Heat Transfer, Part B. Vol.23,
1993, . pp. 269–288.
[13] N. H. Al-Khalidy, Integrated Approach to
Predict Wind Resource Energy from an
Urban Wind Turbine in a complex Built
Environment, Fluid Mechanic, Vol.7, No.2,
2021, pp. 17-28.
[14] N. H. Al-Khalidy, Building Generated Wind
Shear and Turbulence Prediction Utilising
Computational Fluid Dynamics, WSEAS
Transactions On Fluid Mechanics, Vol.13,
2018, pp.126-135.
[15] N. Bhondwe, C.D. Koshti “Analysis of
Premixed Methane-Air Combustion” using
Laser Ignition” International Engineering
Research Journal Special Edition PGCON-
MECV H-2017, 2017.
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
The author contributed in the present research, at all
stages from the formulation of the problem to the
final findings and solution.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
No funding was received for conducting this study.
Conflict of Interest
The authors have no conflict of interest to declare.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2023.19.52
Neihad Hussen Al-Khalidy
E-ISSN: 2224-3496
550
Volume 19, 2023