Research Article A Comparative CFD Study on Simulating...

Research ArticleA Comparative CFD Study on Simulating FlamelessOxy-Fuel Combustion in a Pilot-Scale Furnace

Mersedeh Ghadamgahi12 Patrik Oumllund1 Tomas Ekman3

Nils Andersson2 and Paumlr Joumlnsson2

1Ovako Sweden AB Hofors Sweden2Department of Materials Science and Engineering KTH Royal Institute of Technology Stockholm Sweden3AGA AB The Linde Group Stockholm Sweden

Correspondence should be addressed to Mersedeh Ghadamgahi mersedehkthse

Received 3 June 2016 Revised 17 August 2016 Accepted 19 September 2016

Academic Editor Richard Saurel

Copyright copy 2016 Mersedeh Ghadamgahi et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

The current study presents a method to model the flameless oxy-fuel system with a comparative approach as well as validation ofthe predictions The validation has been done by comparing the predicted results with previously published experimental resultsfrom a 200 kW pilot furnace A suction pyrometer has been used to measure the local temperature and concentrations of CO CO2and O2 at 24 different locations A three-dimensional CFD model was developed and the validity of using different submodelsdescribing turbulence and chemical reactions was evaluated The standard 119896-120576 model was compared with the realizable 119896-120576 modelfor turbulence while Probability Density Function (PDF) with either chemical equilibrium or the Steady Laminar Flamelet Model(SLFM) was evaluated for combustion Radiation was described using a Discrete Ordinates Model (DOM) with weighted-sum-of-grey-gases model (WSGGM) The smallest deviation between predictions and experiments for temperature (12) was foundusing the realizable 119896-120576 model and the SLFMThis improvement affects the prediction of gaseous species as well since the deviationbetween predictions and experiments for CO2 volume percentages decreased from 6 to 15 This provides a recommendationfor model selections in further studies on flameless oxy-fuel combustion

1 Introduction

As the challenges of industrial pollution and energy con-sumption grow into one of the biggest issues of our timethe modernization of industrial heating systems becomes thecenter of attention formany academic researchers One of theknownmethods tomeet this progressive problem in industryis to modify the combustion system by using pure oxygen oroxygen-enriched air as the primary oxidizer These methodswhere the former is called oxy-fuel combustion have beenextensively investigated and also occasionally employed inindustry [1]

The technology of oxy-fuel combustion combined withflue gas recirculation was introduced by Horn and Steinberg[2] and Abraham et al [3] in the early eighties SpecificallyAbraham et al mainly focused on oxy-fuel combustion for

improving the Enhanced Oil Recovery (EOR) and Horn andSteinberg addressed the environmental impacts in energygeneration systems Afterwards this technology receivedeven more attention in research due to the increased incen-tives to decrease the CO2 levels in industrial applications [4]

In 2006 Golchert et al [5] performed a numerical CFDsimulation on an aluminium melting furnace with varyingconcentrations of oxidants They concluded that NO119909 emis-sions increase with an increased temperature andor nitrogenconcentrations in the oxidizer Later Kim et al [6] conductedan experimental investigation on NO119909 emissions from oxy-fuel combustion equipped with a flue gas recirculation (FGR)system with 003MW and 02MW burner capacities Theirresearch indicates the effectiveness of the FGR technology toreduce the NO119909 emissions during oxy-fuel combustion In

Hindawi Publishing CorporationJournal of CombustionVolume 2016 Article ID 6735971 11 pageshttpdxdoiorg10115520166735971

2 Journal of Combustion

2008 Andersson et al [7] experimentally studied the com-bustion chemistry in different combustion processes withrespect to theNO119909 formation in an oxy-fuel and oxy-air com-bustion They showed that the NO119909 emission is considerablyreduced (up to 30) in oxy-fuel combustion compared to itsidentical air-firing system A year later in 2009 Normann etal [8] made another study on the available techniques forNO119909 emission control in oxy-fuel processesTheir study indi-cated that even an implementation of the conventional NO119909control system in a first generation of oxy-fuel power plants isadequate to meet the emission regulation at the time Later in2010 Toftegaarda et al [9] made an extensive literature studyon oxy-fuel combustion with a special focus to investigatethe combustion fundamentals such as flame temperature andemissions They highlighted the deficiency of research inmany areas related to oxy-fuel with both pilot and planttesting such as investigating the heat transfer profiles emis-sion levels and optimal oxygen excess levels at the inlet andvalidating models to predict NO119909 and the formation of SO3

In the recent years improvements in CFD modeling andthe increasing usage of super computers made the CFDtechnique more accessible These studies are mainly forinvestigating the possible design or optimizing methods forburners and furnaces As a result many researchers [10ndash12] started to validate the CFD models for modeling thecomplicated case of oxy-fuel combustion

In 2010 Johansson et al [10] investigated the influence ofdifferent radiation models to simulate gas radiation in oxy-fired boilers Specifically the following radiationmodels weretested the weighted-sum-of-gray-gases model (WSGGM)the spectral line-based WSGGM the two gray-gas approxi-mations and the Statistical Narrow Band (SNB) models Thecomparisons between all the CFD results and experimentaldata show that the WSGGM is the most suitable choice ofall models This shows that this can be used with a lowcomputational cost The authors also discussed that this iseven more important in complicated cases where both wallradiative fluxes and radiative source term are computedLater in 2010 Yin et al [13] derived a modified WSGGMwhich is suitable to use for both oxy-fuel and air-fuel CFDsimulationsThey compared their results with thewidely usedmodel predictions in the literature and derived a set of usefulguidelines for how to model oxy-fuel combustion

As a complement to the former works Hjartstam et al[12] examined the gray and nongrayCFDmodels for gas radi-ation This was done for both oxy-propane and air-propanecombustion systems By comparing the predictions to theexperimental data it was shown that the nongray approachgives an accurate prediction of the source term in both com-bustion systems Furthermore it was found that the simpli-fied gray model fails to predict the source term accurately In2014Wang et al [14] presented research results with a similarfocus Specifically they ran several CFD cases to predict thegas heat transfer by using the gray gas model while they con-figured the accuracy by using the SNB model They showedthat the gray gas model can accurately predict the peak heattransfer flux location but that it overpredicts the heat flux andheat transfer rate by almost 23 for the examined cases

Porter et al [15] also investigated the radiative propertiesof the combustion gases using the P1 model and the DiscreteOrdinate Model (DOM) where both models used a nongrayand a gray method They also concluded that the use of thegray gas model to simulate oxy-fuel combustion might causelarge errors Later Wang et al [16] used many results of theliterature at the time and ran a parametric numerical studyon a propane furnace Specifically they used the 119896-120576 modelincluding a wall function to simulate turbulence DOM tosimulate radiation and the Probability Density Function(PDF)model to calculate the combustion Similarly Yin et al[13] made a parametric study on the use ofWSGGM to fit theparameters for both the air and oxy-fuel combustion systems

By summing up the results from all the former numericalinvestigations and many experimental studies the followingaspects have been found to characterize the oxy-fuel combus-tion

(1) Adrastic decrease in the flue gas volume and thereforean increase in the furnace efficiency and a reductionof the fuel consumption

(2) A reduction of the CO2 emission(3) A higher flame temperature compared to conven-

tional combustion systems(4) An increment in heat transfer due to better radiative

properties of the final products (this will result inincreased productivity)

(5) Improved flame properties with respect to flamestability and ignition characteristics [17]

Parallel to all the proven advantages of this technologymany challenges were also observed This was specificallyreported in the experimental investigations and the industrialimplementations of this technology in existing combustionsystems One of the biggest challenges was seen to be anundesirable increment in the formation of thermal NO119909 incase of air infiltration This is partly due to a very high localtemperature (up to 2000∘C) in the flame region Moreoveran undesirable temperature distribution and a nonuniformheat transfer ratio were also observed in many studies thatemployed this technology [18] Overall extensive efforts andresearch were carried out to find a suitable modification foroxy-fuel combustion in order to get the advantages but avoidthe undesirable effects In this regard the ldquoflameless oxy-fuelrdquo technology was introduced in 1980 as one of the mostpromising solutions to decrease the environmental impactfrom combustion systems

The special burner design in flameless oxy-fuel combus-tion systems surcharges a high internal flue gas recirculation(IFGR) in combination with the other characteristics ofoxy-fuel burners The enhancement of the mixing degreein this mode of the burner is achieved by obtaining anasymmetric oxygen injection with a high (near supersonic)velocity In this configuration the combustion is utilizingthe advantage of a lower exhaust gas volume by using pureoxygen as the oxidant Therefore these types of flames arealmost invisible and they are volumetrically spread alongsidethe chamber The combination of these aspects promises

Journal of Combustion 3

Table 1 The detail of the three developed CFD models

Turbulent model Combustion model Radiationmodel

Model 1 Standard 119896-120576 PDF in chemicalequilibrium

DOM withWSGGM

Model 2 Realizable 119896-120576 PDF in chemicalequilibrium

DOM withWSGGM

Model 3 Realizable 119896-120576 PDF with SLFM DOM withWSGGM

a higher efficiency a much lower NO119909 production and amore uniform temperature profile along the heating zonecompared to standard combustion systems [1 19]

In 2006 Narayanan et al [19] carried out experimentaltrials using a pilot furnace with different burner configu-rations including the flameless oxy-propane Some of theresults from this trial which were done at Linde AGA inStockholm are used in the study Specifically they measuredthe local temperature and gaseous species for different burnertypes including a flameless oxy-propane with the focuson characterizing and comparing the NO119909 formation Theypointed out several advantages of flameless oxy-flame burn-ers in comparison to 4 other types of burner such as a lowerNO119909 formation ratio and a better temperature uniformity[19 20]

Even though using a flameless oxy-fuel combustion sys-tem has been proven to be a good solution to many industrialapplications there is a big scarcity of studies in this areaIn particular a combination of numerical and experimentalstudies is rare Thus this study will target this gap by provid-ing predicted numerical as well as experimental results Ini-tially a basic CFDmodel is developed evaluating the choice ofturbulence model and successively investigating the implica-tions of considering a combustionmodel at either infinite rateor the more computationally expensive steady laminar fla-melet model (SLFM) These combinations of submodels aresummarized in Table 1 Thereafter the results are comparedwith experimental data taken from the experiment trial byLinde AGA in 2005 The authors believe the current resultswill highlight the applicability of adjusting and employingthis new combustion technology in industrial applicationsfor the benefit of the environment

2 Material and Methods

21 Experimental Procedure and Furnace Configuration Theexperimental trial was done by Linde AGA in 2005 using a200 kW pilot cylindrical furnace The experiment at the timewas done as a comparative study on different types of burnersand their operational conditions [20] However in this studythe data from flameless oxy-fuel burner is solely used Duringthe initial stagewhen the furnace atmospherewas at the roomtemperature the burner started the heating process with aconventional combustion After reaching the limit of theself-ignition temperature in the furnace (727∘C) the burnerstarted the combustion in the flameless oxy-fuel modePropane was used as fuel in this study as it is a very desirable

x

y

359mm1116 mm

1862mm4500 mm

Figure 1 Furnace configuration

Burner

Opening 1

Opening 3

Opening 2

x

y

z

x = 1862 mm

x = 1116 mm

x = 359 mm

Figure 2 A photo of the furnace including the three openings

fuel to employ in industrial applications and the oxidizer waspure oxygen

The cylindrical semi-lab scale furnace had a 15m diam-eter and 4m length The burner is installed in the front wallin a coaxial direction with the length of the furnace Also thefurnace was isolated with three layers consisting of 50mmSaffil fibre modules a 200mm SuperG ramming mix and a115mmG-23 insulating brick [19] Also 16 S-type thermocou-ples were installed on the furnace wall at positions of 20mmunder the insulation layer These thermocouples were usedfor constant monitoring of the ldquofurnace wall temperaturerdquoThis configuration is illustrated in Figure 1 The conicalopenings where the probe is horizontally moved across the119909-axis are located at the following three levels of119909 = 359mm119909 = 1116mm and 119909 = 1862mm (Figures 2 and 3) [19]

The measurements were done by employing suctionpyrometers to measure the temperature radiation heat fluxand total surface heat transfer with minimized radiationerrors [20 21] The gas composition was measured using awater-cooled gas-sampling probe to extract the gas from thedesired measurements points Furthermore gas chromatog-raphy was used to analyze the dry-based compositions [19]

22 Burner A REBOX oxy-fuel flameless burner which isdeveloped and commercialized by Linde AGA in Swedenwas used in this studyTheburner which is shown in Figure 4has two injecting nozzles for oxygen and one for fuel Oneessential aspect of these burner types is the injection velocitywhich directly affects the characterization of the combustion


Suction pyrometer

Figure 3 A photo of the positioning of the suction pyrometer

Oxygen inlets

Propane inlet

Figure 4 A photo of the burner configuration including the oxygenand propane inlets

This near sonic injection velocity eventuates an excellentmixing ratio for the oxidant and fuel which forms largeeddies due to a strong turbulence This in turn causes aninternal mixing of flues gases oxygen and fuel in the flameregion The form of the flame in this type of combustionsystem is very spread which is also a reason why it is calleda volumetric combustion [22] The special design of theflameless oxy-fuel combustion requires high speed (nearsonic) injection velocities for both oxygen and propaneAdditionally the asymmetric positioning of feeding nozzlesfor inlets will cause a turbulent flow with internal flue gasrecirculation effects and eddy formations

23 Suction Pyrometers A lot of studies have been doneto validate temperature measurements inside the furnaces(high temperature environments) partly due to the errorsthat radiation causes Specifically Blevins and Pitts [23] havereported that in high temperature conditions correspondingtemperatures above 727∘C radiation errors can be severalhundred degrees In order tomeasure the gas temperature in aradiative condition IFRF developed and introduced ldquosuctionpyrometersrdquo (Figure 5) in 1998 These devices eliminate theerrors due to radiation to a very big extent The details of theuncertainty limits for eachmeasured parameter at 1300∘C areshown in Table 2 while the accuracy of the thermocouple wasmeasured to be plusmn8∘C at 1300∘C [20]

Water jacket

Operating length Water inletWater

outlet

Cooling water

Cooling water

ThermocouplesGuardrings

Stainless steel plug

Figure 5 A sketch of the structure of the suction pyrometerincluding essential parts

Table 2 Uncertainty limits for each measured parameter at 1300∘C

Measurement The minimum andmaximum uncertainties

Flue gas temperature 22ndash57Heat fluxes 2ndash6Flue gas composition 1ndash8

The repeatability of the test was also investigated Specif-ically the measurements were repeated three times by usingidentical experimental conditionsThe standard deviation forthe flame temperature total heat flux and radiative heat fluxwere 4 65 and 2 respectively The repeatability trialto study the flue gas composition showed a maximum andminimumstandard deviation of 8and 4 respectively [20]

3 Mathematical Model

31 Model Assumptions The 3-dimensional mathematicalmodel is based on the following assumptions

(1) The model is a pilot-scale furnace based on thegeometry shown in Figure 1

(2) The standard equations for continuity and momen-tum in Fluent are used [24]

(3) Turbulence is described using either the standard 119896-120576turbulencemodel [25] or the realizable 119896-120576 turbulencemodel [26]

(4) The walls are treated with the standard wall functionrecommended by Launder and Spalding [25]

(5) Chemical reactions are considered using either equi-librium or the steady laminar flamelet model Localvariations in turbulent mixing of the species areconsidered using a Probability Density Function

(6) Radiation is described using discrete ordinatemethodwith the weighted-sum-of-gray-gases model

(7) Leakage of air is neglected so all NO119909 products areeliminated

32 Turbulence The standard 119896-120576 model has been validatedfor similar problems by previous authors [16] However itis known for being overly diffusive and lacking accuracy in


cases with rotating flows and prediction of spreading rateof round jet flows [27] whereas the realizable 119896-120576 modelhas been shown to outperform the standard 119896-120576 model onthese points [28] Therefore in order to investigate whichturbulence model is better suited in this case a comparisonof the standard 119896-120576 model [25] and the realizable 119896-120576 model[26] is made

321 Standard 119896-120576120597 (120588119896)

120597119905 + 120597 (120588119896119906119894)120597119909119894 = 120597120597119909119895 lceil(120583 + 120583119905120590119896)

120597119896120597119909119895rceil + 119866119896 + 119866119887

minus 120588120576(1)

120597 (120588120576)120597119905 + 120597 (120588120576119906119894)120597119909119894 = 120597

120597119909119895 [(120583 + 120583119905120590120576)120597120576120597119909119895]

+ 1198621120576 120576119896 (119866119896 + 1198623120576119866119887)

minus 11986221205761205881205762119896 + 119878120576

(2)

1198623120576 = tanh10038161003816100381610038161003816100381610038161003816V119906perp

10038161003816100381610038161003816100381610038161003816 (3)

where 120590119896 = 10 120590120576 = 13 1198622120576 = 192 1198621120576 = 144 119866119896represents the generation of turbulence kinetic energy dueto the mean velocity gradients and 119866119887 is the generation ofturbulent kinetic energy due to the buoyancy V and 119906perp arethe velocity components parallel and perpendicular to thegravity vector respectively

The turbulent viscosity is given as follows

120583119905 = 120588119862120583 1198962

120576 (4)

where 119862120583 = 009

322 Realizable 119896-120576 The transport equation for turbulentkinetic energy 119896 is the same as (1) with different constantswhile turbulent dissipation 120576 is as follows

120597 (120588120576)120597119905 + 120597 (120588120576119906119895)

120597119909119895 = 120597120597119909119895 [(120583 + 120583119905120590120576)

120597120576120597119909119895] + 1205881198621119878120576

minus 1205881198622 1205762119896 + radic]120576 + 1198621120576 1205761198961198623120576119866119887

1198621 = max [043 120578120578 + 5]

120578 = 119878119896120576

119878 = radic2119878119894119895119878119894119895

(5)

where 120590119896 = 10 120590120576 = 12 1198622 = 19 1198621120576 = 144 and 1198623120576 isdescribed by (3)

The turbulent viscosity is also described by (4) with thedifference that 119862120583 is a function in the realizable 119896-120576 model

33 Combustion For nonpremixed combustion the thermo-chemistry can thus be reduced to a conserved scalar quantitycalled the mixture fraction (119891)

119891 = 119885119894 minus 119885119894ox119885119894fuel minus 119885119894ox (6)

where 119885119894 stands for the 119894th species mass fraction and thesubscripts ox and fuel stand for the oxidizer stream andfuel stream respectively For a nonadiabatic system such asthe present model the instantaneous species mass fractiondensity or temperature is a function of both mixture fractionand enthalpy

A PDF is used to consider subgrid local fluctuationsbetween the species Furthermore the burner causes a rapidchemical reaction with a highly strained flame shape whichleads to forming conspicuous amounts of nonequilibriumspecies Thus to investigate if the infinite rate (local equilib-rium) is good enough to describe this system a comparisonis made between the infinite rate and the SLFM The lattermodel predicts the chemical nonequilibrium flame strainingcaused by turbulence

In SLFM thermochemistry parameters are a functionof mixture fraction and the strain rate or equivalentlydissipation rate (120594)

120594 = 2119863 1003816100381610038161003816nabla11989110038161003816100381610038162 (7)

where 119863 is the diffusion coefficientConsequently in this method the thermochemistry

parameters are fully parameterized by mixture fraction (119891)dissipation rate (120594) and enthalpy (119867) However in orderto moderate the costs of calculation in SLFM the effectof heat lossgain (119867) is neglected Therefore the followingformulation will be used to calculate the thermochemicalparameters

Φ = ∬ Φ (119891 120594st) 119901 (119891 120594st) 119889119891 119889120594st (8)

where Φ represents species mass fractions and temperatureand 120594st represents the stoichiometric dissipation rate

34 Radiation Oxy-fuel combustion strongly promotes theradiative heat transfer since a higher volume fraction of CO2and H2O exists in the flue gas composition in comparisonwith air combustion [11] Thus a careful consideration ofradiation model is needed

In this problem there are no participating media in theradiative space so the effect of scattering will be cancelledThis will transform the general radiative transfer equation(RTE) into the following relationship

119889119868]119889119904 = 119896]119868119887] minus 119896]119868] (9)

where 119868] stands for the intensity in the wavelength 119896] is theabsorption coefficient and 119868119887] is the black body radiative


Qua

lity

inde

x

(0 0 0)

934e minus 01

876e minus 01

817e minus 01

758e minus 01

700e minus 01

641e minus 01

x

y

z

Figure 6 Contours of orthogonal quality

intensity [10] By integrating this equation over the radiativepath of 119878 and after performing a spectral averaging over aband 119896 the following equation will be obtained for each cell[10]

119868]119896119899 = 119868]1198960120591]1198960rarr119899 + 119899minus1sum119894=0

(120591]119896119894+1rarr119899 minus 120591]119896119894rarr119899) 119868119887]119896119894+12 (10)

where overbar stands for a spectral average over a band119896 index 119894 refers to spatial discretization and 120591]119896 can beexpressed as

120591]119896 = 1 minus 119886]119896 (11)

where 119886 is the absorption coefficientIn the nongray model this equation will be solved once

for a single band range by using a constant absorption coeffi-cient However in oxy-fuel combustion the assumption of aconstant absorption coefficient reduces the accuracy [29ndash32]

The DOM was used to solve the RTE for a number ofdiscrete solid angles By incorporating the domain basedstandardWSGGM reasonable predictions of wall heat fluxesand radiative source terms can be expected [10] The trick inWSGGM is that the total emissivity over the distance 119904 can bepresented as follows [33]

120576 = 119868sum119894=0

119886120576119894 (119879) (1 minus 119890minus119896119894119901 119904) (12)

where 119886120576119894 stands for the emissivity weighting factor for the119894th fictitious gray gas and the quantity in the bracket standsfor the 119894th fictitious gray gas emissivity Furthermore 119896119894 is theabsorption coefficient of the 119894th gray gas 119901 is the summationof the partial pressure of all absorbing gases and 119904 is the pathlength

35 Domain The domain of the furnace is illustrated inFigure 6 The mesh was selected to be unstructured with atotal node number of 1 300 000 on the total domainThemeshstructure is finer on the edges close to the wall boundariesand in the area of flame formationTheminimumorthogonal

Table 3 Boundary condition of the inlet flows

Fuel OxygenFlow rates Nm3h 77 404Low heat value MJm3 463Composition C3H8 = 100 O2 = 100Temperature ∘C 25 20Density kgNm3 202 1411Density kgm3 2016 33Velocity ms 94 3517Diameter of nozzle mm 6 4 times 2

quality is 0641 which is adequate regarding general limi-tations Figure 6 shows the contours of orthogonal qualityalongside a horizontal plane in the middle of the simulatedfurnace

4 Boundary Conditions

The capacity of the experimental furnace is 200 kW Detailsof the inlet combustion boundary condition are shown inTable 3 The turbulent intensity and hydraulic diameter forthe propane nozzle were calculated to be 6 and 6mmrespectively Furthermore for the oxygen nozzles the corre-sponding values are 5 and 4mm respectively

Specific heat and thermal conductivity for the interiorwalls are set to 2650 kgm3 880 Jkg-K and 180Wm-Krespectively These are calculated regarding the special caseof isolations on refractory walls in this study

The standard wall function was used to bridge near walllayer with the mean flow in (1) and (2) where 119884+ wascalculated to be between 60 and 200

The density and specific heat of themixture are calculatedlocally according to the PDF table Also the absorptioncoefficient of the mixture is calculated by using WSGGMThe boundary condition on the walls of the furnace wasconsidered as a fixed temperature of 1200∘C The outlet isselected as a pressure type with no gauge pressure Also theturbulent intensity and hydraulic diameter of the outlet werecalculated to have the values of 5 and 500mm respectively

5 Results and Discussions

In this section predicted CFD results for different parame-ters namely temperature and gaseous species (CO CO2 andO2) are comparedwith experimental resultsThese predictedresults are taken from three different CFD models namelymodels 1 2 and 3 For turbulence model 1 uses the standard119896-120576 model whereas models 2 and 3 use the realizable 119896-120576model For combustion a PDF is used in combination withthe chemical reactions assumed at infinite rate (equilibrium)in models 1 and 2 whereas for model 3 the PDF is combinedwith reactions at finite rates (nonequilibrium)with the SLFM

The accuracy of locating the probe for taking the exper-imental data is within plusmn10mm The suction pyrometer hasan insertion diameter of 10mm which sucks the regionalflue gas with a high velocityTherefore the experimental data


1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

0 500 1000 1500 2000 2500Axial distance from burner centerline towards

the end of chamber (mm)ExperimentModel 1

Model 3Model 2

(0 0 0)

T(∘

C)

x

yz

Figure 7 Temperature magnitudes on the axial line of the chamberfrom burner centerline towards the end of the chamber

represent an average volumeof the flue gas composition in thearea of the point of measurement Also the experiments havebeen done 3 times for every point so a stable magnitude foreach point has been determined The scatter associated withthe measured data is 6 [20]

51 Temperature Figure 7 shows the temperature magni-tudes on the axial line of the chamber on its length forboth experimental data and calculated data from three CFDmodels This measurement is done on an axis starting fromburner center and towards the end of the chamber

Overall in regions away from the burner all of the CFDresults are within the uncertainty of the experiments asshown in Figure 7 The measurement point at 359mm issuspected to have additional uncertainty than is shown by theerror bars

In addition to this temperature measurement and inorder to get a comprehensive picture of the temperature pro-file measurements were carried out both in the flame regionand in a region further away from the flame In particularthemeasurements have been done at three different distancesaway from the burner These three lines are located at thedistances 359mm 1116mm and 1862mm from the burnerwall with the origin (119911 = 0) at the axis of the center of theburner This is clearly shown in Figure 2

Figures 8(a) 8(b) and 8(c) show the temperature datameasured at these points and the corresponding calculateddata from the three different CFD models at the locations359mm (a) 1116mm (b) and 1862mm (c) from the burner

As shown in Figure 8(a) there is a major improvementin predicting the temperature in regions close to the reaction

1000

1100

1200

1300

1400

1500

1600

0 200 400 600 800 1000Distance from burner centerline (mm)

minus200minus400minus600minus800minus1000

T(∘

C)

ExperimentModel 1

Model 3Model 2

(a)

11001140118012201260130013401380142014601500

Distance from burner centerline (mm)0 200 400 600 800 1000minus200minus400minus600minus800minus1000

T(∘

C)ExperimentModel 1

Model 3Model 2

(b)

11001140118012201260130013401380142014601500

100 300 500 700 900Distance from burner centerline (mm)

minus100minus300minus500minus700

T(∘

C)

ExperimentModel 1

Model 3Model 2

(c)

Figure 8 Temperature predictions and measurements of the radialaxis at the positions (a) 119909 = 359mm (b) 119909 = 1116mm and (c)119909 = 1862mm

zone in model 3 compared to models 1 and 2 The corre-sponding deviation from experiments reduces from 76 to12 from model 1 to model 3 respectively These resultsaccentuate the difference of considering infinite rate reactionscompared to finite rate reactions in reaction regions wherestretch effects are not negligible A better realization about


0

100

200

300

400

500

600

0 200 400 600 800 1000

Stra

in ra

te (1

s)

Position (mm)minus200minus400minus600minus800minus1000

Figure 9 Strain rate along the radial line of 119909 = 359mm close tothe flame region

Tem

pera

ture

(∘C)1700

15711443131411861057929800

x

y

z

Figure 10 Temperature distribution profile on the 119909-119910 plane in themiddle of the chamber model 1

this might be achieved if one looked at the strain rate in thisregion which is shown in Figure 9 The high strain rate closeto the reaction zone explicitly illustrates the departure fromchemical equilibrium in this region

The improvements from model 1 to model 2 on the otherhand are not significant though the predicted temperatureis generally lower in model 2 This is due to the tendencyof the standard 119896-120576 turbulence model to overpredict theturbulence of the flow which leads to a higher mixing rateand consequently a higher temperature level

In Figures 8(b) and 8(c) an asymmetric temperaturedistribution is obvious which resembles the nature of flame-less oxy-fuel burners The discrepancy between the injectionvelocities for the fuel and oxygen and the asymmetric burnerdesign while increasing the mixing ratio will also lead to adeviated flow path inside the chamber Figure 10 illustratesthis in a more coherent way

52 Gas Composition

521 Flame Region Figures 11(a) 11(b) and 11(c) showthe predicted concentrations by the CFD models and themeasured corresponding values of volume percentages ofO2 CO2 and CO on the dry basis These data are fromthe flame region (359mm from the burner side) on a halfradial axis starting from 100mm from centerline towards thechamber wall Results extrapolated from the models are incorrespondence with the sampling locations

According to Figure 11(a) the assumption of infinite ratereactions eventuates a 9 deviation from experimental data(for model 1 and model 2) This assumption which considersa nonequilibrium chemistry results in the formation of

100 200 300 400 500Distance from centerline of burner (mm) in the flame region

ExperimentModel 1

Model 3Model 2

00

20

40

60

80

100

120

2(

dry

)

(a)

02468

101214161820


ExperimentModel 1

Model 3Model 2

2

(V

dry)

(b)

00

05

10

15

20

25

30


ExperimentModel 1

Model 3Model 2

CO (

Vdr

y)

(c)

Figure 11 (a) Predicted andmeasured CO2 volume percentage (drybasis) on the half radial axis at 119909 = 359 (b) O2 volume percentage(dry basis) at the half radial axis at 119909 = 359 (c) CO volumepercentage (dry basis) at the half radial axis at 119909 = 359

a large amount of intermediate radicals in the reaction zoneOn the contrary modifying the turbulent model does not


000

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

00010203040506070809101112131415

50 150 250 350 450Radial distance from centerline of burner (mm) at x = 1116

2(

Vdr

y)

lowast10

and

2(

Vdr

y)

lowast 10 experiment2 experiment

2 model 2 lowast 10 model 12 model 12 model 3

lowast 10 model 3 lowast 10 model 22 model 12 model 32 model 22 experiment

(a)

00

10

20

30

40

50

60

70

80

90

100

0123456789

101112131415

100 200 300 400 500

2(

Vdr

y)

lowast10

and

2(

Vdr

y) lowast 10 experiment2 experiment 2 model 2

lowast 10 model 12 model 12 model 3 lowast 10 model 3 lowast 10 model 2

2 model 12 model 32 model 2

2 experiment

Radial distance from centerline of burner (mm) at x = 1862

(b)

Figure 12 Predicted and measured volume percentages of CO lowast 10 CO2 and O2 (dry basis) at (a) 119909 = 1116mm and (b) 119909 = 1862mm

show considerable improvements in predicting the gaseousspecies in the flame region

522 Chamber Region Figures 12(a) and 12(b) show thepredicted and measured values of the volume percentages ofO2 CO2 and CO lowast 10 on the dry basis Results are takenfrom the radial axis located at 1116mm and 1862mm fromthe burner (119911 = 0 119909 = 0) respectively The axis reflectingthe CO2 values is shown on the right-hand side of the figurewhile the common axis for CO lowast 10 and O2 dry volumepercentages is shown on the left-hand side

In the chamber region which stands for the overallchamber volume the predictions from all the 3 models arereasonably close to the experimental data Specifically themaximum deviation between the predicted and measureddata predicted bymodel 1 is seen for CO at 1116mm from theorigin (11) and forO2 in 1862mm from the origin (8)Thisdeviation is decreased with the realizable 119896-120576 model while itis reduced to 7 for CO at 1116mm from the origin The pre-dictedCObymodel 3 and experimental data are following thesame trend which shows the strength of this model One canalso note that at some occasions a nonharmonized behavioris seen in the experimental data This raises the suspicion ofsome imperfections in collecting these data like for CO in119911 = 50mm O2 in 119911 = 300mm and CO2 in 119911 = 150mmRegarding this O2 is well predicted by model 3 and less finebymodel 1 andmodel 2 CO2 prediction follows the samepro-gression as in the flame region model 3 with considering the

2m

olar

frac

tion

(wet

)

0384

0288

0192

0096

0000

x

y

z

Figure 13 Distribution profile of volume percentage of CO2 (wetbasis) on the 119909-119910 plane in the middle of the chamber

nonequilibrium effects realizes the full formation of CO2 incontrast with the other twomodelsThe improvement of CO2predictions by model 2 compared to model 1 (by modifyingturbulent model) is also noticeable since both model 2 andmodel 3 are in very close agreement with experimental datain Figures 12(a) and 12(b) Model 3 and model 2 deviate 15and 22 from the experiments respectively In comparisonmodel 1 has a 6 deviation 1862mm away from the burner

As it was observed and described for the temperaturedistribution profile an asymmetric design of this type ofburners forms a nonsymmetric distribution for the gaseousspecies as well as for the temperature Figure 13 which showsthe molar fraction of CO2 predicted by model 1 (given as thewet percentage) illustrates this in a more tangible way


Table 4 The measured and predicted amount of volume percent-ages (dry basis) of O2 CO2 and CO at the exhaust

Predictions Experimentalresults

inconsistencyin total

( 119881 dry) O2 21 28 1332( 119881 dry) CO 0 0 0( 119881 dry) CO2 91 935 263

The measured and predicted amounts of the volumepercentages (dry basis) of O2 CO2 and CO at the exhaustare shown in Table 4

6 Conclusions

Three different sophisticated 3D CFD models have beendeveloped and used to simulate a flameless oxy-fuel combus-tion and results were compared with experimental data froma pilot furnace Model 2 and model 3 are the modified ver-sions of model 1 with respect to turbulence and combustionmodels respectively

The following are the main findings in this study

(i) Model 1 whichwas the least computationally intensivemodel used the standard 119896-120576 for turbulence and PDFwith infinite rate reactions for combustion and hada good agreement in temperature between predictedand experimental data far away from the burnerClose to the burner this model was inferior to theother models

(ii) Model 2 showed clear improvements concerninggaseous species away from the burner radially Thisis a direct cause of the realizable 119896-120576 model for turbu-lence However for temperature large deviations stillremain in the flame region

(iii) Model 3 predicted both the temperature and thegaseous species well in all of the furnace by con-sidering finite rate reactions as well as the realizable119896-120576 modelThe better prediction of temperature closeto the burner was due to the consideration of finitereaction rate which more accurately modeled theheat dissipation in the flame regionThe improvementin prediction of gaseous species overall was due tothe combination of the better combustion modeldescription of the chemical reactions in the flameregion and the more accurate turbulence descriptionof the transport of combustion products

This concludes that it is possible to accurately predict flame-less oxy-fuel combustion by using complex models such asthe realizable 119896-120576 model for turbulence and steady laminarflamelet model for combustion

Competing Interests

The authors declare that they have no competing interests

Acknowledgments

The experiment work has been done at Linde AGA Swedenwhich the authors gratefully acknowledge Special thanks goto Ovako Sweden AB for financing the research In additionthe authors are very thankful to Mohsen Saffari Pour at KTHfor assisting in many ways in this study and Garry Wicks forhelping with the language Finally special thanks are due toHenrik Hall for his support and constructive comments

References

[1] J Von Scheele M Gartz R Paul M T Lantz J P Riegertand S Soderlund ldquoFlameless oxyfuel combustion for increasedproduction and reduced CO2 and NOx emissionsrdquo Stahl undEisen vol 128 no 7 pp 35ndash42 2008

[2] F L Horn and M Steinberg ldquoControl of carbon dioxide emis-sions from a power plant (and use in enhanced oil recovery)rdquoFuel vol 61 no 5 pp 415ndash422 1982

[3] B M Abraham J G Asbury E P Lynch and A P S TeotialdquoCoal-oxygen process provides CO2 for enhanced recoveryrdquoOilGas Journal vol 80 article 11 1982

[4] R Tan G Corragio and S Santos ldquoOxy-coal combustion withflue gas recycle for the power generation industrymdasha literaturereviewrdquo IFRF Doc G 23y1 International Flame ResearchFoundation (IFRF) 2005

[5] B Golchert P Ridenour W Walker M Gu N K Selvarasuand C Zhou ldquoEffect of nitrogen and oxygen concentrations onNOx emissions in an aluminum furnacerdquo in Proceedings of theASME 2006 International Mechanical Engineering Congress andExposition pp 7918ndash4783 Chicago Ill USA 2006

[6] H K Kim Y Kim S M Lee and K Y Ahn ldquoNO reduction in003-02 MW oxy-fuel combustor using flue gas recirculationtechnologyrdquo Proceedings of the Combustion Institute vol 31 pp3377ndash3384 2007

[7] K Andersson F Normann F Johnsson and B Leckner ldquoNOemission during oxy-fuel combustion of ligniterdquo Industrial andEngineering Chemistry Research vol 47 no 6 pp 1835ndash18452008

[8] F Normann K Andersson B Leckner and F JohnssonldquoEmission control of nitrogen oxides in the oxy-fuel processrdquoProgress in Energy and Combustion Science vol 35 no 5 pp385ndash397 2009

[9] M B Toftegaarda J Brixa P A Jensena P Glarborga and AD Jensena ldquoOxy-fuel combustion of solid fuelsrdquo in Progress inEnergy and Combustion Science vol 36 2010

[10] R Johansson K Andersson B Leckner and H ThunmanldquoModels for gaseous radiative heat transfer applied to oxy-fuelconditions in boilersrdquo International Journal of Heat and MassTransfer vol 53 no 1ndash3 pp 220ndash230 2010

[11] C Yin ldquoNongray-gas effects in modeling of large-scale oxy-fuel combustion processesrdquo Energy and Fuels vol 26 no 6 pp3349ndash3356 2012

[12] S Hjartstam R Johansson K Andersson and F JohnssonldquoComputational fluid dynamics modeling of oxy-fuel flamesthe role of soot and gas radiationrdquo Energy and Fuels vol 26 no5 pp 2786ndash2797 2012

[13] C Yin L C R Johansen L A Rosendahl and S K KaeligrldquoNewweighted sum of gray gases model applicable to computa-tional fluid dynamics (CFD) modeling of oxy-fuel combustionderivation validation and implementationrdquo Energy and Fuelsvol 24 no 12 pp 6275ndash6282 2010


[14] P Wang F Fan and Q Li ldquoAccuracy evaluation of the gray gasradiation model in CFD simulationrdquo Case Studies in ThermalEngineering vol 3 pp 51ndash58 2014

[15] R Porter F Liu M Pourkashanian A Williams and D SmithldquoEvaluation of solution methods for radiative heat transferin gaseous oxy-fuel combustion environmentsrdquo Journal ofQuantitative Spectroscopy and Radiative Transfer vol 111 no 14pp 2084ndash2094 2010

[16] A-H Wang J-J Cai and G-W Xie ldquoNumerical simulation ofcombustion characteristics in high temperature air combustionfurnacerdquo Journal of Iron and Steel Research International vol 16no 2 pp 6ndash10 2009

[17] C E Baukal Jr Oxygen-Enhanced Combustion CRC PressBoca Raton Fla USA 2nd edition 2013

[18] P Fredriksson E Claesson P Vesterberg A Lugnet andO Ritzen Application of Oxyfuel Combustion in Reheatingat Ovako Hofors Works SwedenmdashBackground Solutions andResults S l Linde AGA 2006

[19] K Narayanan W Wang W Blasiak and T Ekman ldquoFlamelessoxyfuel combustion technology modeling and benefits in userdquoRevue de Metallurgie Cahiers DrsquoInformations Techniques vol103 no 5 pp 210ndash217 2006

[20] N Krishnamurthy P J Paul and W Blasiak ldquoStudies onlow-intensity oxy-fuel burnerrdquo Proceedings of the CombustionInstitute vol 32 pp 3139ndash3146 2009

[21] J Chedaille and Y Braud Measurement in Flames Hodder ampStoughton Educ 1972

[22] J A Wunning and J G Wunning ldquoFlameless oxidation toreduce thermal no-formationrdquo Progress in Energy and Combus-tion Science vol 23 no 1 pp 81ndash94 1997

[23] L G Blevins and W M Pitts ldquoModeling of bare and aspiratedthermocouples in compartment firesrdquo Fire Safety Journal vol33 no 4 pp 239ndash259 1999

[24] Fluent Theory Guide ANSYS Academic Research Release 15Help System Non-premixed combustion ANSYS Inc

[25] B E Launder and D B Spalding ldquoThe numerical computationof turbulent flowsrdquoComputerMethods inAppliedMechanics andEngineering vol 3 no 2 pp 269ndash289 1974

[26] T-H Shih W W Liou A Shabbir Z Yang and J Zhu ldquoA newk-120576 eddy viscosity model for high reynolds number turbulentflowsrdquo Computers and Fluids vol 24 no 3 pp 227ndash238 1995

[27] Y-S Chen and S-W Kim ldquoComputation of turbulent flowsusing an extended k-epsilon turbulence closure modelrdquo NASAReport CR-179204 1987

[28] T H Shih W W Liou A Shabbir Z Yang and J Zhu ldquoAnew 119896-120576 Eddy viscosity model for high reynolds number tur-bulent flowsmdashmodel development and validationrdquo DocumentID 19950005029 (Acquired Dec 28 1995) Accession Num-ber 95N11442 Subject Category FLUID Mechanics and HeatTransfer ReportPatent Number NASA-TM-106721 ICOMP-94-21 E-9087 NAS 115106721 CMOTT-94-6 Document TypeTechnical Report Publisher Information United States 1994httpntrsnasagovsearchjspR=19950005029

[29] W Yang andW Blasiak ldquoNumerical simulation of properties ofa LPG flamewith high-temperature airrdquo International Journal ofThermal Sciences vol 44 no 10 pp 973ndash985 2005

[30] C Yin L A Rosendahl and S K Kaeligr ldquoChemistry and radi-ation in oxy-fuel combustion a computational fluid dynamicsmodeling studyrdquo Fuel vol 90 no 7 pp 2519ndash2529 2011

[31] M Habermehl J Erfurth D ToporovM Forster and R KneerldquoExperimental and numerical investigations on a swirl oxycoalflamerdquo Applied Thermal Engineering vol 49 pp 161ndash169 2012

[32] M Ghadamgahi P Olund A Lugnet M Saffari Pour and WYang ldquoDesign optimization of flameless-oxyfuel soaking pitfurnace using CFD techniquerdquo Energy Procedia vol 61 pp 611ndash614 2014

[33] M F Modest Radiative Heat Transfer Academic Press 2013

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



2008 Andersson et al [7] experimentally studied the com-bustion chemistry in different combustion processes withrespect to theNO119909 formation in an oxy-fuel and oxy-air com-bustion They showed that the NO119909 emission is considerablyreduced (up to 30) in oxy-fuel combustion compared to itsidentical air-firing system A year later in 2009 Normann etal [8] made another study on the available techniques forNO119909 emission control in oxy-fuel processesTheir study indi-cated that even an implementation of the conventional NO119909control system in a first generation of oxy-fuel power plants isadequate to meet the emission regulation at the time Later in2010 Toftegaarda et al [9] made an extensive literature studyon oxy-fuel combustion with a special focus to investigatethe combustion fundamentals such as flame temperature andemissions They highlighted the deficiency of research inmany areas related to oxy-fuel with both pilot and planttesting such as investigating the heat transfer profiles emis-sion levels and optimal oxygen excess levels at the inlet andvalidating models to predict NO119909 and the formation of SO3

In the recent years improvements in CFD modeling andthe increasing usage of super computers made the CFDtechnique more accessible These studies are mainly forinvestigating the possible design or optimizing methods forburners and furnaces As a result many researchers [10ndash12] started to validate the CFD models for modeling thecomplicated case of oxy-fuel combustion

In 2010 Johansson et al [10] investigated the influence ofdifferent radiation models to simulate gas radiation in oxy-fired boilers Specifically the following radiationmodels weretested the weighted-sum-of-gray-gases model (WSGGM)the spectral line-based WSGGM the two gray-gas approxi-mations and the Statistical Narrow Band (SNB) models Thecomparisons between all the CFD results and experimentaldata show that the WSGGM is the most suitable choice ofall models This shows that this can be used with a lowcomputational cost The authors also discussed that this iseven more important in complicated cases where both wallradiative fluxes and radiative source term are computedLater in 2010 Yin et al [13] derived a modified WSGGMwhich is suitable to use for both oxy-fuel and air-fuel CFDsimulationsThey compared their results with thewidely usedmodel predictions in the literature and derived a set of usefulguidelines for how to model oxy-fuel combustion

As a complement to the former works Hjartstam et al[12] examined the gray and nongrayCFDmodels for gas radi-ation This was done for both oxy-propane and air-propanecombustion systems By comparing the predictions to theexperimental data it was shown that the nongray approachgives an accurate prediction of the source term in both com-bustion systems Furthermore it was found that the simpli-fied gray model fails to predict the source term accurately In2014Wang et al [14] presented research results with a similarfocus Specifically they ran several CFD cases to predict thegas heat transfer by using the gray gas model while they con-figured the accuracy by using the SNB model They showedthat the gray gas model can accurately predict the peak heattransfer flux location but that it overpredicts the heat flux andheat transfer rate by almost 23 for the examined cases

Porter et al [15] also investigated the radiative propertiesof the combustion gases using the P1 model and the DiscreteOrdinate Model (DOM) where both models used a nongrayand a gray method They also concluded that the use of thegray gas model to simulate oxy-fuel combustion might causelarge errors Later Wang et al [16] used many results of theliterature at the time and ran a parametric numerical studyon a propane furnace Specifically they used the 119896-120576 modelincluding a wall function to simulate turbulence DOM tosimulate radiation and the Probability Density Function(PDF)model to calculate the combustion Similarly Yin et al[13] made a parametric study on the use ofWSGGM to fit theparameters for both the air and oxy-fuel combustion systems

By summing up the results from all the former numericalinvestigations and many experimental studies the followingaspects have been found to characterize the oxy-fuel combus-tion

(1) Adrastic decrease in the flue gas volume and thereforean increase in the furnace efficiency and a reductionof the fuel consumption

(2) A reduction of the CO2 emission(3) A higher flame temperature compared to conven-

tional combustion systems(4) An increment in heat transfer due to better radiative

properties of the final products (this will result inincreased productivity)

(5) Improved flame properties with respect to flamestability and ignition characteristics [17]

Parallel to all the proven advantages of this technologymany challenges were also observed This was specificallyreported in the experimental investigations and the industrialimplementations of this technology in existing combustionsystems One of the biggest challenges was seen to be anundesirable increment in the formation of thermal NO119909 incase of air infiltration This is partly due to a very high localtemperature (up to 2000∘C) in the flame region Moreoveran undesirable temperature distribution and a nonuniformheat transfer ratio were also observed in many studies thatemployed this technology [18] Overall extensive efforts andresearch were carried out to find a suitable modification foroxy-fuel combustion in order to get the advantages but avoidthe undesirable effects In this regard the ldquoflameless oxy-fuelrdquo technology was introduced in 1980 as one of the mostpromising solutions to decrease the environmental impactfrom combustion systems

The special burner design in flameless oxy-fuel combus-tion systems surcharges a high internal flue gas recirculation(IFGR) in combination with the other characteristics ofoxy-fuel burners The enhancement of the mixing degreein this mode of the burner is achieved by obtaining anasymmetric oxygen injection with a high (near supersonic)velocity In this configuration the combustion is utilizingthe advantage of a lower exhaust gas volume by using pureoxygen as the oxidant Therefore these types of flames arealmost invisible and they are volumetrically spread alongsidethe chamber The combination of these aspects promises





DOM withWSGGM


DOM withWSGGM







x

y

359mm1116 mm

1862mm4500 mm


Burner

Opening 1

Opening 3

Opening 2

x

y

z

x = 1862 mm

x = 1116 mm

x = 359 mm







Suction pyrometer


Oxygen inlets

Propane inlet




Water jacket


outlet

Cooling water

Cooling water




















321 Standard 119896-120576120597 (120588119896)

120597119905 + 120597 (120588119896119906119894)120597119909119894 = 120597120597119909119895 lceil(120583 + 120583119905120590119896)

120597119896120597119909119895rceil + 119866119896 + 119866119887

minus 120588120576(1)

120597 (120588120576)120597119905 + 120597 (120588120576119906119894)120597119909119894 = 120597

120597119909119895 [(120583 + 120583119905120590120576)120597120576120597119909119895]

+ 1198621120576 120576119896 (119866119896 + 1198623120576119866119887)

minus 11986221205761205881205762119896 + 119878120576

(2)

1198623120576 = tanh10038161003816100381610038161003816100381610038161003816V119906perp

10038161003816100381610038161003816100381610038161003816 (3)



120583119905 = 120588119862120583 1198962

120576 (4)

where 119862120583 = 009


120597 (120588120576)120597119905 + 120597 (120588120576119906119895)

120597119909119895 = 120597120597119909119895 [(120583 + 120583119905120590120576)

120597120576120597119909119895] + 1205881198621119878120576

minus 1205881198622 1205762119896 + radic]120576 + 1198621120576 1205761198961198623120576119866119887

1198621 = max [043 120578120578 + 5]

120578 = 119878119896120576

119878 = radic2119878119894119895119878119894119895

(5)








120594 = 2119863 1003816100381610038161003816nabla11989110038161003816100381610038162 (7)



Φ = ∬ Φ (119891 120594st) 119901 (119891 120594st) 119889119891 119889120594st (8)




119889119868]119889119904 = 119896]119868119887] minus 119896]119868] (9)



Qua

lity

inde

x

(0 0 0)

934e minus 01

876e minus 01

817e minus 01

758e minus 01

700e minus 01

641e minus 01

x

y

z



119868]119896119899 = 119868]1198960120591]1198960rarr119899 + 119899minus1sum119894=0

(120591]119896119894+1rarr119899 minus 120591]119896119894rarr119899) 119868119887]119896119894+12 (10)


120591]119896 = 1 minus 119886]119896 (11)




120576 = 119868sum119894=0

119886120576119894 (119879) (1 minus 119890minus119896119894119901 119904) (12)















1000

1100

1200

1300

1400

1500

1600

1700

1800

1900



Model 3Model 2

(0 0 0)

T(∘

C)

x

yz








1000

1100

1200

1300

1400

1500

1600



T(∘

C)

ExperimentModel 1

Model 3Model 2

(a)

11001140118012201260130013401380142014601500


T(∘

C)ExperimentModel 1

Model 3Model 2

(b)

11001140118012201260130013401380142014601500



T(∘

C)

ExperimentModel 1

Model 3Model 2

(c)




0

100

200

300

400

500

600

0 200 400 600 800 1000

Stra

in ra

te (1

s)



Tem

pera

ture

(∘C)1700

15711443131411861057929800

x

y

z





52 Gas Composition




ExperimentModel 1

Model 3Model 2

00

20

40

60

80

100

120

2(

dry

)

(a)

02468

101214161820


ExperimentModel 1

Model 3Model 2

2

(V

dry)

(b)

00

05

10

15

20

25

30


ExperimentModel 1

Model 3Model 2

CO (

Vdr

y)

(c)




000

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

00010203040506070809101112131415


2(

Vdr

y)

lowast10

and

2(

Vdr

y)




(a)

00

10

20

30

40

50

60

70

80

90

100

0123456789

101112131415

100 200 300 400 500

2(

Vdr

y)

lowast10

and

2(

Vdr




2 experiment


(b)





2m

olar

frac

tion

(wet

)

0384

0288

0192

0096

0000

x

y

z








( 119881 dry) O2 21 28 1332( 119881 dry) CO 0 0 0( 119881 dry) CO2 91 935 263


6 Conclusions







Competing Interests


Acknowledgments


References





































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













DOM withWSGGM


DOM withWSGGM







x

y

359mm1116 mm

1862mm4500 mm


Burner

Opening 1

Opening 3

Opening 2

x

y

z

x = 1862 mm

x = 1116 mm

x = 359 mm







Suction pyrometer


Oxygen inlets

Propane inlet




Water jacket


outlet

Cooling water

Cooling water




















321 Standard 119896-120576120597 (120588119896)

120597119905 + 120597 (120588119896119906119894)120597119909119894 = 120597120597119909119895 lceil(120583 + 120583119905120590119896)

120597119896120597119909119895rceil + 119866119896 + 119866119887

minus 120588120576(1)

120597 (120588120576)120597119905 + 120597 (120588120576119906119894)120597119909119894 = 120597

120597119909119895 [(120583 + 120583119905120590120576)120597120576120597119909119895]

+ 1198621120576 120576119896 (119866119896 + 1198623120576119866119887)

minus 11986221205761205881205762119896 + 119878120576

(2)

1198623120576 = tanh10038161003816100381610038161003816100381610038161003816V119906perp

10038161003816100381610038161003816100381610038161003816 (3)



120583119905 = 120588119862120583 1198962

120576 (4)

where 119862120583 = 009


120597 (120588120576)120597119905 + 120597 (120588120576119906119895)

120597119909119895 = 120597120597119909119895 [(120583 + 120583119905120590120576)

120597120576120597119909119895] + 1205881198621119878120576

minus 1205881198622 1205762119896 + radic]120576 + 1198621120576 1205761198961198623120576119866119887

1198621 = max [043 120578120578 + 5]

120578 = 119878119896120576

119878 = radic2119878119894119895119878119894119895

(5)








120594 = 2119863 1003816100381610038161003816nabla11989110038161003816100381610038162 (7)



Φ = ∬ Φ (119891 120594st) 119901 (119891 120594st) 119889119891 119889120594st (8)




119889119868]119889119904 = 119896]119868119887] minus 119896]119868] (9)



Qua

lity

inde

x

(0 0 0)

934e minus 01

876e minus 01

817e minus 01

758e minus 01

700e minus 01

641e minus 01

x

y

z



119868]119896119899 = 119868]1198960120591]1198960rarr119899 + 119899minus1sum119894=0

(120591]119896119894+1rarr119899 minus 120591]119896119894rarr119899) 119868119887]119896119894+12 (10)


120591]119896 = 1 minus 119886]119896 (11)




120576 = 119868sum119894=0

119886120576119894 (119879) (1 minus 119890minus119896119894119901 119904) (12)















1000

1100

1200

1300

1400

1500

1600

1700

1800

1900



Model 3Model 2

(0 0 0)

T(∘

C)

x

yz








1000

1100

1200

1300

1400

1500

1600



T(∘

C)

ExperimentModel 1

Model 3Model 2

(a)

11001140118012201260130013401380142014601500


T(∘

C)ExperimentModel 1

Model 3Model 2

(b)

11001140118012201260130013401380142014601500



T(∘

C)

ExperimentModel 1

Model 3Model 2

(c)




0

100

200

300

400

500

600

0 200 400 600 800 1000

Stra

in ra

te (1

s)



Tem

pera

ture

(∘C)1700

15711443131411861057929800

x

y

z





52 Gas Composition




ExperimentModel 1

Model 3Model 2

00

20

40

60

80

100

120

2(

dry

)

(a)

02468

101214161820


ExperimentModel 1

Model 3Model 2

2

(V

dry)

(b)

00

05

10

15

20

25

30


ExperimentModel 1

Model 3Model 2

CO (

Vdr

y)

(c)




000

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

00010203040506070809101112131415


2(

Vdr

y)

lowast10

and

2(

Vdr

y)




(a)

00

10

20

30

40

50

60

70

80

90

100

0123456789

101112131415

100 200 300 400 500

2(

Vdr

y)

lowast10

and

2(

Vdr




2 experiment


(b)





2m

olar

frac

tion

(wet

)

0384

0288

0192

0096

0000

x

y

z








( 119881 dry) O2 21 28 1332( 119881 dry) CO 0 0 0( 119881 dry) CO2 91 935 263


6 Conclusions







Competing Interests


Acknowledgments


References





































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Suction pyrometer


Oxygen inlets

Propane inlet




Water jacket


outlet

Cooling water

Cooling water




















321 Standard 119896-120576120597 (120588119896)

120597119905 + 120597 (120588119896119906119894)120597119909119894 = 120597120597119909119895 lceil(120583 + 120583119905120590119896)

120597119896120597119909119895rceil + 119866119896 + 119866119887

minus 120588120576(1)

120597 (120588120576)120597119905 + 120597 (120588120576119906119894)120597119909119894 = 120597

120597119909119895 [(120583 + 120583119905120590120576)120597120576120597119909119895]

+ 1198621120576 120576119896 (119866119896 + 1198623120576119866119887)

minus 11986221205761205881205762119896 + 119878120576

(2)

1198623120576 = tanh10038161003816100381610038161003816100381610038161003816V119906perp

10038161003816100381610038161003816100381610038161003816 (3)



120583119905 = 120588119862120583 1198962

120576 (4)

where 119862120583 = 009


120597 (120588120576)120597119905 + 120597 (120588120576119906119895)

120597119909119895 = 120597120597119909119895 [(120583 + 120583119905120590120576)

120597120576120597119909119895] + 1205881198621119878120576

minus 1205881198622 1205762119896 + radic]120576 + 1198621120576 1205761198961198623120576119866119887

1198621 = max [043 120578120578 + 5]

120578 = 119878119896120576

119878 = radic2119878119894119895119878119894119895

(5)








120594 = 2119863 1003816100381610038161003816nabla11989110038161003816100381610038162 (7)



Φ = ∬ Φ (119891 120594st) 119901 (119891 120594st) 119889119891 119889120594st (8)




119889119868]119889119904 = 119896]119868119887] minus 119896]119868] (9)



Qua

lity

inde

x

(0 0 0)

934e minus 01

876e minus 01

817e minus 01

758e minus 01

700e minus 01

641e minus 01

x

y

z



119868]119896119899 = 119868]1198960120591]1198960rarr119899 + 119899minus1sum119894=0

(120591]119896119894+1rarr119899 minus 120591]119896119894rarr119899) 119868119887]119896119894+12 (10)


120591]119896 = 1 minus 119886]119896 (11)




120576 = 119868sum119894=0

119886120576119894 (119879) (1 minus 119890minus119896119894119901 119904) (12)















1000

1100

1200

1300

1400

1500

1600

1700

1800

1900



Model 3Model 2

(0 0 0)

T(∘

C)

x

yz








1000

1100

1200

1300

1400

1500

1600



T(∘

C)

ExperimentModel 1

Model 3Model 2

(a)

11001140118012201260130013401380142014601500


T(∘

C)ExperimentModel 1

Model 3Model 2

(b)

11001140118012201260130013401380142014601500



T(∘

C)

ExperimentModel 1

Model 3Model 2

(c)




0

100

200

300

400

500

600

0 200 400 600 800 1000

Stra

in ra

te (1

s)



Tem

pera

ture

(∘C)1700

15711443131411861057929800

x

y

z





52 Gas Composition




ExperimentModel 1

Model 3Model 2

00

20

40

60

80

100

120

2(

dry

)

(a)

02468

101214161820


ExperimentModel 1

Model 3Model 2

2

(V

dry)

(b)

00

05

10

15

20

25

30


ExperimentModel 1

Model 3Model 2

CO (

Vdr

y)

(c)




000

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

00010203040506070809101112131415


2(

Vdr

y)

lowast10

and

2(

Vdr

y)




(a)

00

10

20

30

40

50

60

70

80

90

100

0123456789

101112131415

100 200 300 400 500

2(

Vdr

y)

lowast10

and

2(

Vdr




2 experiment


(b)





2m

olar

frac

tion

(wet

)

0384

0288

0192

0096

0000

x

y

z








( 119881 dry) O2 21 28 1332( 119881 dry) CO 0 0 0( 119881 dry) CO2 91 935 263


6 Conclusions







Competing Interests


Acknowledgments


References





































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











321 Standard 119896-120576120597 (120588119896)

120597119905 + 120597 (120588119896119906119894)120597119909119894 = 120597120597119909119895 lceil(120583 + 120583119905120590119896)

120597119896120597119909119895rceil + 119866119896 + 119866119887

minus 120588120576(1)

120597 (120588120576)120597119905 + 120597 (120588120576119906119894)120597119909119894 = 120597

120597119909119895 [(120583 + 120583119905120590120576)120597120576120597119909119895]

+ 1198621120576 120576119896 (119866119896 + 1198623120576119866119887)

minus 11986221205761205881205762119896 + 119878120576

(2)

1198623120576 = tanh10038161003816100381610038161003816100381610038161003816V119906perp

10038161003816100381610038161003816100381610038161003816 (3)



120583119905 = 120588119862120583 1198962

120576 (4)

where 119862120583 = 009


120597 (120588120576)120597119905 + 120597 (120588120576119906119895)

120597119909119895 = 120597120597119909119895 [(120583 + 120583119905120590120576)

120597120576120597119909119895] + 1205881198621119878120576

minus 1205881198622 1205762119896 + radic]120576 + 1198621120576 1205761198961198623120576119866119887

1198621 = max [043 120578120578 + 5]

120578 = 119878119896120576

119878 = radic2119878119894119895119878119894119895

(5)








120594 = 2119863 1003816100381610038161003816nabla11989110038161003816100381610038162 (7)



Φ = ∬ Φ (119891 120594st) 119901 (119891 120594st) 119889119891 119889120594st (8)




119889119868]119889119904 = 119896]119868119887] minus 119896]119868] (9)



Qua

lity

inde

x

(0 0 0)

934e minus 01

876e minus 01

817e minus 01

758e minus 01

700e minus 01

641e minus 01

x

y

z



119868]119896119899 = 119868]1198960120591]1198960rarr119899 + 119899minus1sum119894=0

(120591]119896119894+1rarr119899 minus 120591]119896119894rarr119899) 119868119887]119896119894+12 (10)


120591]119896 = 1 minus 119886]119896 (11)




120576 = 119868sum119894=0

119886120576119894 (119879) (1 minus 119890minus119896119894119901 119904) (12)















1000

1100

1200

1300

1400

1500

1600

1700

1800

1900



Model 3Model 2

(0 0 0)

T(∘

C)

x

yz








1000

1100

1200

1300

1400

1500

1600



T(∘

C)

ExperimentModel 1

Model 3Model 2

(a)

11001140118012201260130013401380142014601500


T(∘

C)ExperimentModel 1

Model 3Model 2

(b)

11001140118012201260130013401380142014601500



T(∘

C)

ExperimentModel 1

Model 3Model 2

(c)




0

100

200

300

400

500

600

0 200 400 600 800 1000

Stra

in ra

te (1

s)



Tem

pera

ture

(∘C)1700

15711443131411861057929800

x

y

z





52 Gas Composition




ExperimentModel 1

Model 3Model 2

00

20

40

60

80

100

120

2(

dry

)

(a)

02468

101214161820


ExperimentModel 1

Model 3Model 2

2

(V

dry)

(b)

00

05

10

15

20

25

30


ExperimentModel 1

Model 3Model 2

CO (

Vdr

y)

(c)




000

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

00010203040506070809101112131415


2(

Vdr

y)

lowast10

and

2(

Vdr

y)




(a)

00

10

20

30

40

50

60

70

80

90

100

0123456789

101112131415

100 200 300 400 500

2(

Vdr

y)

lowast10

and

2(

Vdr




2 experiment


(b)





2m

olar

frac

tion

(wet

)

0384

0288

0192

0096

0000

x

y

z








( 119881 dry) O2 21 28 1332( 119881 dry) CO 0 0 0( 119881 dry) CO2 91 935 263


6 Conclusions







Competing Interests


Acknowledgments


References





































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Qua

lity

inde

x

(0 0 0)

934e minus 01

876e minus 01

817e minus 01

758e minus 01

700e minus 01

641e minus 01

x

y

z



119868]119896119899 = 119868]1198960120591]1198960rarr119899 + 119899minus1sum119894=0

(120591]119896119894+1rarr119899 minus 120591]119896119894rarr119899) 119868119887]119896119894+12 (10)


120591]119896 = 1 minus 119886]119896 (11)




120576 = 119868sum119894=0

119886120576119894 (119879) (1 minus 119890minus119896119894119901 119904) (12)















1000

1100

1200

1300

1400

1500

1600

1700

1800

1900



Model 3Model 2

(0 0 0)

T(∘

C)

x

yz








1000

1100

1200

1300

1400

1500

1600



T(∘

C)

ExperimentModel 1

Model 3Model 2

(a)

11001140118012201260130013401380142014601500


T(∘

C)ExperimentModel 1

Model 3Model 2

(b)

11001140118012201260130013401380142014601500



T(∘

C)

ExperimentModel 1

Model 3Model 2

(c)




0

100

200

300

400

500

600

0 200 400 600 800 1000

Stra

in ra

te (1

s)



Tem

pera

ture

(∘C)1700

15711443131411861057929800

x

y

z





52 Gas Composition




ExperimentModel 1

Model 3Model 2

00

20

40

60

80

100

120

2(

dry

)

(a)

02468

101214161820


ExperimentModel 1

Model 3Model 2

2

(V

dry)

(b)

00

05

10

15

20

25

30


ExperimentModel 1

Model 3Model 2

CO (

Vdr

y)

(c)




000

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

00010203040506070809101112131415


2(

Vdr

y)

lowast10

and

2(

Vdr

y)




(a)

00

10

20

30

40

50

60

70

80

90

100

0123456789

101112131415

100 200 300 400 500

2(

Vdr

y)

lowast10

and

2(

Vdr




2 experiment


(b)





2m

olar

frac

tion

(wet

)

0384

0288

0192

0096

0000

x

y

z








( 119881 dry) O2 21 28 1332( 119881 dry) CO 0 0 0( 119881 dry) CO2 91 935 263


6 Conclusions







Competing Interests


Acknowledgments


References





































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










1000

1100

1200

1300

1400

1500

1600

1700

1800

1900



Model 3Model 2

(0 0 0)

T(∘

C)

x

yz








1000

1100

1200

1300

1400

1500

1600



T(∘

C)

ExperimentModel 1

Model 3Model 2

(a)

11001140118012201260130013401380142014601500


T(∘

C)ExperimentModel 1

Model 3Model 2

(b)

11001140118012201260130013401380142014601500



T(∘

C)

ExperimentModel 1

Model 3Model 2

(c)




0

100

200

300

400

500

600

0 200 400 600 800 1000

Stra

in ra

te (1

s)



Tem

pera

ture

(∘C)1700

15711443131411861057929800

x

y

z





52 Gas Composition




ExperimentModel 1

Model 3Model 2

00

20

40

60

80

100

120

2(

dry

)

(a)

02468

101214161820


ExperimentModel 1

Model 3Model 2

2

(V

dry)

(b)

00

05

10

15

20

25

30


ExperimentModel 1

Model 3Model 2

CO (

Vdr

y)

(c)




000

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

00010203040506070809101112131415


2(

Vdr

y)

lowast10

and

2(

Vdr

y)




(a)

00

10

20

30

40

50

60

70

80

90

100

0123456789

101112131415

100 200 300 400 500

2(

Vdr

y)

lowast10

and

2(

Vdr




2 experiment


(b)





2m

olar

frac

tion

(wet

)

0384

0288

0192

0096

0000

x

y

z








( 119881 dry) O2 21 28 1332( 119881 dry) CO 0 0 0( 119881 dry) CO2 91 935 263


6 Conclusions







Competing Interests


Acknowledgments


References





































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

100

200

300

400

500

600

0 200 400 600 800 1000

Stra

in ra

te (1

s)



Tem

pera

ture

(∘C)1700

15711443131411861057929800

x

y

z





52 Gas Composition




ExperimentModel 1

Model 3Model 2

00

20

40

60

80

100

120

2(

dry

)

(a)

02468

101214161820


ExperimentModel 1

Model 3Model 2

2

(V

dry)

(b)

00

05

10

15

20

25

30


ExperimentModel 1

Model 3Model 2

CO (

Vdr

y)

(c)




000

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

00010203040506070809101112131415


2(

Vdr

y)

lowast10

and

2(

Vdr

y)




(a)

00

10

20

30

40

50

60

70

80

90

100

0123456789

101112131415

100 200 300 400 500

2(

Vdr

y)

lowast10

and

2(

Vdr




2 experiment


(b)





2m

olar

frac

tion

(wet

)

0384

0288

0192

0096

0000

x

y

z








( 119881 dry) O2 21 28 1332( 119881 dry) CO 0 0 0( 119881 dry) CO2 91 935 263


6 Conclusions







Competing Interests


Acknowledgments


References





































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










000

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

00010203040506070809101112131415


2(

Vdr

y)

lowast10

and

2(

Vdr

y)




(a)

00

10

20

30

40

50

60

70

80

90

100

0123456789

101112131415

100 200 300 400 500

2(

Vdr

y)

lowast10

and

2(

Vdr




2 experiment


(b)





2m

olar

frac

tion

(wet

)

0384

0288

0192

0096

0000

x

y

z








( 119881 dry) O2 21 28 1332( 119881 dry) CO 0 0 0( 119881 dry) CO2 91 935 263


6 Conclusions







Competing Interests


Acknowledgments


References





































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













( 119881 dry) O2 21 28 1332( 119881 dry) CO 0 0 0( 119881 dry) CO2 91 935 263


6 Conclusions







Competing Interests


Acknowledgments


References





































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation
































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









Research Article A Comparative CFD Study on Simulating...

Documents

Transcript of Research Article A Comparative CFD Study on Simulating...