- Email: [email protected]
3D numerical modelling of turbulent biogas combustion in a newly generated 10 KW burner
Accepted Manuscript 3D Numerical modelling of turbulent biogas combustion in a newly generated 10 kw burner Mustafa İlbaş, Murat Şahin, Serhat Karyeyen PII:
S1743-9671(16)30488-3
DOI:
10.1016/j.joei.2016.10.004
Reference:
JOEI 276
To appear in:
Journal of the Energy Institute
Received Date: 22 August 2016 Revised Date:
18 October 2016
Accepted Date: 19 October 2016
Please cite this article as: M. İlbaş, M. Şahin, S. Karyeyen, 3D Numerical modelling of turbulent biogas combustion in a newly generated 10 kw burner, Journal of the Energy Institute (2016), doi: 10.1016/ j.joei.2016.10.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 3D NUMERICAL MODELLING OF TURBULENT BIOGAS COMBUSTION IN A NEWLY GENERATED 10 KW BURNER Mustafa İLBAŞ*1, Murat ŞAHİN** and Serhat KARYEYEN* *
Gazi University, Technology Faculty, Department of Energy Systems Engineering, 06500,
**
RI PT
Teknikokullar/Ankara/TURKEY Gazi University, Graduate School of Natural and Applied Science, 06500, Teknikokullar, Ankara/TURKEY
SC
Abstract
This study concentrates on the 3D numerical modelling of combustion of different biogases in a generated burner and combustor. The main goal of this study is to investigate the
M AN U
combustion characteristics (such as temperature and emissions) of biogases through a combustor due to depletion of natural gas. Moreover, the effect of the preheated air on flame temperatures of biogases have been studied in the present study. Finally, the effect of H2S amount in biogas on SO2 emissions has been investigated within these predictions. The numerical modelling of turbulent diffusion flames has been performed by using the standard
TE D
k–ε model of turbulent flow, the PDF/Mixture Fraction combustion model and P-1 radiation model in the combustor. A CFD code has been used for all predictions. Temperature gradients have been determined on axial and radial directions for better understanding combustion characteristics of biogases. Modelling has been studied for thermal power of 10 kW and
EP
excess air ratio of λ= 1.2 for each biogas combustion. The first finding is that combustion of biogases is possible via the newly generated burner. Moreover, the results show that the one of biogas is very close to methane in terms of temperature distributions in the combustor due
AC C
to including high amount of methane compared to other biogases. It is also concluded that the flame temperatures of biogases increase with preheating the combustion air as expected. It is finally revealed that SO2 emissions increase as amount of H2S in biogas is increased through the combustor.
Keywords: Biogas, Burner, Combustion, Emission, CFD Modelling
1
Corresponding author. Email: [email protected]; Tel: + 90 312 202 86 09; Fax: + 90 312 202 89 47
ACCEPTED MANUSCRIPT 1. Introduction Although fossil fuels are still used in power generation systems, buildings and elsewhere, these have two problems. First, fossil fuels are restricted and are not equally distributed all over the World. Scientists estimate that coal, crude oil and natural gas will be rapidly depleted in the near future. Second, fossil fuels are responsible for global warming. Because, CO2 is
RI PT
majorly released when fossil fuel are burned in any system. Because, there are aforementioned problems, the scientists try to find new energy resources such as solar energy, wind energy, hydrogen energy, biomass and the others.
SC
The growing interest in the use of biogas has promoted nowadays. Biogas arising from effluent, municipal solid waste, residuals, gasification of biomass are very appropriate for power generation, heating systems and elsewhere. Biogas consists chiefly of methane and
M AN U
carbondioxide as well as it has a trace amount of nitrogen, hydrogen sulphur and oxygen depending on production or gasification methods [1]. Methane in biogas has high amount of energy as major component of natural gas. However, the other components are noncombustible.
TE D
Biogas combustion has to be investigated for better understanding of combustion characteristics of biogases. There are some studies about combustions of biogas or blending fuels in the literature. Adouane et al. [2] have conducted an experimental study related to reduction of NOX emissions arising from fuel-bound nitrogen. They conclude that ammonia in
EP
the fuel drastically affects NOX formation. Bhoi and Channiwala [3] have experimentally investigated emission characteristics and axial flame temperature distributions of producer gas including high amount of nitrogen by varying thermal inputs and equivalance ratios. The
AC C
results show that the maximum temperature emerges in flame front for this gas. Hosseini et al. [4] have performed combustion characteristics of biogas under flameless combustion conditions. Hosseini et al. estimate that biogas is not appropriate for any furnaces due to its low calorific value (LCV). Huynh and Kong [5] have determined NOx emissions coming from syngas combustion obtained three different biomass feedstock under different conditions. It can be concluded that highest NOx levels form under syngas combustion obtained seed corn as the seed corn has the highest nitrogen content. Leung and Wierzba [6] have investigated the effect of hydrogen addition on stability limits of non-premixed biogas flames. It is demonstrated that hydrogen addition on the enhancement of biogas highly affects.
ACCEPTED MANUSCRIPT İlbaş and Karyeyen [7] have modelled the combustion characteristics of coal gases including high amount of hydrogen or nitrogen depending on gasification or carbonization methods. They reveal that the coke oven gas is very suitable in combustion systems in terms of combustion performances (e.g. temperature values). Sethuraman et al. [8] have focused on the effects of nitrogen content in biomass feedstock on the producer gas composition and NOx
RI PT
emissions. They have found that there are relationships between nitrogen in biomass, ammonia in the producer gas, and NOx emissions in the flue gas. Somehsaraei et al. [9] have examined the fuel flexibility and performance analysis of biogas in micro gas turbines. When minor modifications to fuel valves and compressor were taken place, They have claimed that
SC
these modifications were assumed to allow engine operation with the simulated biogas composition. Chen and Zheng [10] have predicted hydrogen-enriched biogas MILD oxy-fuel conditions. It is found that biogas flame can be sustained under the MILD oxy-fuel
M AN U
combustion. Nikpey et al. [11] have investigated the impact of biogas and natural gas on combustion performance and emissions in a micro gas turbine. The results indicate that there is almost no change in electrical efficiency comparing with the natural gas fired case. Selim et al. [12] have studied effect of CO2 and N2 concentration during H2S combustion. They have found that injection of CO2 highly affected temperature values. Mordaunt and Pierce [13]
TE D
have designed a combustion device and investigated the effects of CO2 on combustion of CH4 and emissions. Hosseini and Wahid [14] have studied combustion characteristics of biogas under hydrogen-enriched combustion conditions. The results show that NOX formation increases as the flame temperature increases. Lafay et al. [15] have compared stability
EP
combustion domains, flame structures and Dynamics between methane and biogas flames. They have revealed that laminar flame speed strongly depends on fuel composition. Jahangirian et al. [16] have examined chemical and thermal influences of biogas CO2 content.
AC C
It is demonstrated that the presence of CO2 in the fuel highly affects NOX emissions (reducing). Hosseini and Wahid [17] have modelled biogas under flameless combustion conditions in a tangential burner. The results indicate that the temperature uniformity in the tangential flameless burner is more than coaxial configurations.
Some studies related to methane combustions were also reviewed as the methane combustion has been also modelled in the present study in order to validate the modellings. Feyz et al. [18], for example, have researched the effect of recess length on the combustion performance of methane flame. They have observed that the flame temperature increase as the recess length is increased. Saqr et al. [19] have modelled effect of free stream turbulence on NOX
ACCEPTED MANUSCRIPT formation of methane flames. They have reported that change in free stream turbulence affects NOX formation level considerably. In addition to these studies, there is a study regarding combustion of low calorific value gas by authors. İlbaş and Karyeyen [20] have also studied combustion behaviours of low calorific value coal gases by enriching hydrogen in order to improve their combustion characteristics in that study. They have concluded that
RI PT
hydrogen addition highly affects the flame temperature of the low calorific value coal gases.
Although there are some aforementioned studies related to biogas or blending fuels combustion, there needs to be extremely studied combustion characteristics of biogas.
SC
Because, there are a lot of problems concerning combustion of biogas such as flame stability, flame propogation, flame temperature and emissions. However, the authors think that the newly generated of burner can contribute flame stability and relative high flame temperature
M AN U
of biogas because it has radial fuel inlets and angular and straight air inlets. These inlets can provide better fuel-air mixture and as a result of this, better combustion performance. If they are all to be considered, this study focuses on combustion performances and emission characteristics of biogases including different gas components by means of the newly
2. CFD Modelling
TE D
generated of burner under thermal power of 10 kW and excess air ratio of 1.2
2.1. Descriptions of burner and model combustor 3D CFD modelling has been performed by using the newly generated diffusion flame of
EP
burner in order to model combustions of methane (as baseline fuel) and biogases in the present study. This burner is shown in Figure 1a. A and B in Figure 1a represent the fuel and air inlets, respectively. It has also turbulators with 15° angles in the side of air stream in order
AC C
to provide flame stabilization. Moreover, the fuel outlet of this burner has radial inlets for better mixing of air and fuel can be seen in Figure 1b. Hydraulic diameters of fuel and air inlets are 6 mm and 32 mm. Also, hydraulic diameter of the combustor outlet is 110 mm. Fuel and air inlet temperatures, combustion gauge pressures are also fixed at 283 K and 21 mbar, respectively. The combustor outlet is open to atmosphere.
RI PT
ACCEPTED MANUSCRIPT
(a)
(b)
SC
Figure 1a. Fuel and air inlets in the burner; Figure 1b. Radial fuel inlets
The burner and combustor mesh structure used in this study can be seen in Figure 2. This
EP
TE D
M AN U
combustor is sudden expansion type and has length of 100 cm and diameter of 40 cm.
AC C
Figure 2. The burner and combustor mesh structure
2.2. Governing Equations
The mathematical modelling described for the gas mixture fuel combustion is based on the steady-state condition, 3D continuity, momentum, energy and species equations. The general form of transport equation is below [21]: Ф
+ Ф = Ф + Ф
(1)
ACCEPTED MANUSCRIPT where Ф represents the dependent variables. shows the transport coefficient for variable Ф
and Ф states the source term of the transport equaiton for Ф.
The Mixture Fraction/PDF Model has been used as combustion model in this study. This combustion model consists of the solution of transport equations for a single conserved scalar
RI PT
(the mixture fraction). In this model transport equations for individual species are not solved. Instead, individual component concentrations for species of interest are derived form the predicted mixture fraction distribution. The interaction of turbulence and chemistry are
SC
accounted for with the help of a probability density function (PDF) [22].
The PDF modelling approach was specifically developed for the simulation of turbulent
of the local fuel mass fraction as: =
M AN U
diffusion flames. For a fuel/oxidant system, the mixture fraction, can be expressed in terms
(2)
TE D
Where and are mass fraction of fuel and oxidant respectively. The mixture fraction, is a conserved quantity whose value at each point in the flow domain is computed by the solution of the following conservation equation for the time averaged
̅
̅
=
!
"# ̅
$#
% +
(3)
AC C
+
EP
value of in the turbulent flow field [22].
where is the source term which is due solely to transfer of mass into the gas phase from liquid fuel droplets.
In addition to solving for the mean mixture fraction, a conservation equation is solved for the '
mixture fraction variance, ′ which is used in the closure model describing turbulencechemistry interactions [22]: )
(
+
)
(
)
"# (
= *$
#
)
(
'
1
+ + ,- . * + − ,0 2 ′
'
(4)
ACCEPTED MANUSCRIPT where 3 , ,- and ,0 are constants used in the mixture fraction/PDF model. The radiation heat transfer happens at high temperatures. In combustors, the flame temperature is generally high (1000-1600°C) especially at near stoichiometric combustion
RI PT
conditions. Therefore, the heat transfer from swirl combustors is significant. Thus, It is essential to include the radiation model for a more accurate prediction of the temperature distribution in combustors [22]. The P-1 model is the simplest case of the more general P–N model. The P-1 model has some advantages over the other radiation models. With the P-1
SC
model, it is easy to solve with little CPU demand, the model includes the effect of scattering. For combustion applications where the optical thickness is large, the P-1 model works reasonably well. The P-1 model can be easily applied to complicated geometries with
M AN U
curvilinear coordinates [23]. For these reason, P-1 radiation model was used for a more accurate prediction of biogas combustion in the present study.
Numerical simulation of the turbulence reacting flow in the newly generated of burner and the combustor was carried out using FLUENT code. It uses the finite volume method to solve the
TE D
governing equations for a fluid and species and provides the capability to use different physical models such as inviscid or viscous, laminar or turbulent and the other [24]. The Simple algorithm was applied for the pressure–velocity coupling. Second order upwind scheme was used for the convection terms in all transport equations, and the second order
EP
method was used for the pressure discretization. The solution convergence is assumed when all of the residuals parameters fall below 10-5. The numerical prediction was carried out with
AC C
the following assumptions: The flow was steady-state, three dimensional and turbulent.
2.3.Definition of biogases
Components of biogases consumed in this study and methane as baseline fuel are given in Table 1.These biogases are defined as phases 1, 2, 3 and 4 depending on contents in this study.
ACCEPTED MANUSCRIPT Table 1. Biogas components Methane
Phase 1
Phase 2
Phase 3
Phase 4
(%)
(%)
(%)
(%)
100
55
60
65
55
CO2
-
43,1
38
33
43,1
N2
-
1,53
1,5
1,3
1,53
H2S
-
10 ppm
0 ppm
O2
-
0,3
0,5
8040,00
4422,06
4824,00
(kcal/m3)
0 ppm
0 ppm
0,7
0,37
5226,00
4422,00
SC
Calorific value
RI PT
CH4
M AN U
3. Experimental Setup
In the experimental part of the present study, temperature values have been measured on some axial and radial locations through the combustor in order to validate the modellings. The
AC C
EP
TE D
schematic view of the combustion system is shown in Figure 3.
Figure 3. The layout of the combustion system
A general view of the combustion chamber is illustrated in Figure 4. The length and diameter of the combustion chamber are fixed at 100 cm and 40 cm, respectively. The combustion
ACCEPTED MANUSCRIPT chamber consists of a sight glass made of tempered glass and five measuring ports. These measuring ports are placed on the combustor wall in order to measure temperature values on axial and radial positions within the combustor. The first measuring port is designed to determine the flame characteristics of the biogases and location of this measuring port is 10 cm away from the combustor inlet. The other four measuring ports are placed at intervals of
M AN U
SC
RI PT
20 cm from the first measuring port to the combustion chamber outlet.
TE D
Figure 4. The combustion chamber and the burner
Ceramic coated R-type thermocouples capable of withstanding high temperatures up to 1700°C were used to measure temperature values through the combustion chamber. The diameters of these thermocouples are 5 mm. At high temperatures, the radiation correction is
EP
necessary the temperature levels precisely. In particular, radiative heat transfer takes places from the flame to the combustor wall. Mean bulk velocities were calculated theoretically in
AC C
the flame front considering fuel and air velocities. Then, temperature correction was performed for each axial and radial measurements. It was computed to be between 2–50 °C depending on the measured axial, radial and wall temperatures in the combustor.
4. Results and Discussion 4.1. Mesh independence
Mesh independence is very important situation in numerical studies as excessive mesh structure needs advanced computer technology, long computational time and etc. For this reason, mesh independence can be taken into account before predictions. The methane has been selected as baseline fuel in order to determine fine mesh structure. Grid structures
ACCEPTED MANUSCRIPT including 311812, 447198, 613947, 893508, 1523435 elements have been tested in the predictions. The influence of the mesh refinement on the axial temperature profiles for these mesh structures is shown in Figure 3. Because there is no difference between predictions at 613947 elements mesh structure and further two mesh structures, the 613947 elements mesh
M AN U
SC
RI PT
structure was selected as the fine mesh structure for further 3D modelling cases.
Figure 5. Axial temperature gradients for different mesh structure
TE D
4.2. Experimental validation and the effect of the turbulence models Predictions are compared with experimental measurements. Axial and radial temperature values of methane flames have been measured at 10 cm, 30 cm, 50 cm, 70 cm and 90 cm axial distances through the combustor. Combustion conditions such as stream flow rates, inlet
EP
temperatures and pressures, fuel (methane for validation) and ect. have been identically selected for better capturing the predictions (thermal power of 10 kW and excess air ratio of
AC C
1.2). In addition to the validation, the effect of the turbulence models on temperature profiles have been investigated in order to determine more appropriate turbulence model for this modelling. It is clearly seen in Figure 6, the axial temperature profiles are in good agreement with the experimental data. There is a very slight difference between measurements and the predictions through the combustor.
SC
RI PT
ACCEPTED MANUSCRIPT
M AN U
Figure 6. Axial temperature gradients for different turbulence models and model validation with experimental data
Figure 7 shows the radial temperature gradients of methane at 10 cm and 50 cm axial positions. The predictions are in satisfactorily agreement with the measurements as can be seen in Figure 7. Moreover, the effect of the turbulence models on temperature distributions
TE D
has been studied as previously mentioned. One of the most prominent turbulence models, the k-Ɛ standard turbulence model, is implemented in most general purpose CFD codes and is considered the industry standard model. It has proven to be stable and numerically robust and has a well established regime of predictive capability. For general purpose simulations, the
EP
model offers a good compromise in terms of accuracy and robustness [25]. Although general trends of the predictions are very similar for three turbulence models (Figure 6 and Figure 7),
AC C
the best consistency between the measurements and the predictions emerged while using the k-Ɛ standard turbulence models. Therefore, k-Ɛ standard turbulence model has been chosen to be used further 3D modellings.
RI PT
ACCEPTED MANUSCRIPT
models and model validation with the experimental data
SC
(a) (b) Figure 7a. Radial temperature gradients at 10 cm axial distance for different turbulence
Figure 7b. Radial temperature gradients at 50 cm axial distance for different turbulence
M AN U
models and model validation with the experimental data.
Predictions are also compared to the measurements in axial and radial directions for biogas (phase 1). Predicted and measured temperature distributions of biogas (phase 1) combustion are shown in Figure 8 and Figure 9. As can be seen in Figure 8 and Figure 9, the predictions
AC C
EP
combustion conditions.
TE D
in both directions are in good agreement with the measurements under biogas (phase 1)
Figure 8. Axial temperature distributions of biogas (phase 1) combustion
M AN U
SC
50 cm
RI PT
ACCEPTED MANUSCRIPT
90 cm
AC C
EP
TE D
30 cm
10 cm
70 cm
Burner walls Swirl vanes
Figure 9. Radial temperature distributions of biogas (phase 1) combustion Fuel Inlet
Air Inlet
ACCEPTED MANUSCRIPT 4.3. Temperature distributions 4.3.1. Axial and radial temperatures of biogases Figure 10 depicts the axial temperature distributions of different biogas contents and methane. The maximum temperature level emerges under the methane combustion through the combustor due to its calorific value. The temperature levels of phase 3 biogas are very closer
RI PT
to that of methane in the flame zone. The authors estimate that this situation is related to calorific value as phase 3 includes high amount of methane comparison to the other biogases. All temperature distributions increase in the flame zone initially and then, decrease gradually
TE D
M AN U
SC
towards the combustor outlet due to convective and radiative heat transfers.
EP
Figure 10. Axial temperature gradients of biogases and methane Radial temperature profiles between center of the combustor and combustor walls are shown
AC C
in Figure 11. Initially, temperature profiles at axial distance of 10 cm decrease rapidly and remain the same towards the combustor wall as this combustor type is sudden expansion. When general temperature trends are evaluated, the temperature levels of methane are generally high comparison to the these of biogases. In particular, tcombustion performance of phase 3 biogas is relatively high in comparison to the other biogases. This circumstance may be explained the presence of more CH4 as combustible component in phase 3. However, the general trends for all biogases are very similar at radial directions. Moreover, It is revealed that the temperature values between center and combustor wall are getting closer towards the combustor outlet because the flame propogates more effectively under all combustion conditions.
M AN U
SC
50 cm
RI PT
ACCEPTED MANUSCRIPT
90 cm
AC C
EP
TE D
30 cm
10 cm
70 cm
Burner walls Swirl vanes
Figure 11. Radial temperature gradients of biogases and methane through the combustor Fuel Inlet
Air Inlet
ACCEPTED MANUSCRIPT 4.3.2. The effect of preheated air on flame temperature Combustion air has been preheated before entering the burner in order to understand the effects of this circumstance on flame temperatures of biogases in the present study. This influence is clearly shown in Figure 12. As It can be seen in Figure 12, the flame temperatures of biogases increase as the combustion air is preheated before entering the burner air inlet.
RI PT
The maximum flame temperatures for all preheat conditions are high under the phase 3 biogas conditions due to including high amount of methane compared to the other biogases as It was mentioned before. Moreover, there are slight differences between the flame temperatures of
TE D
M AN U
SC
Phase1 and 4 biogases. Phase 1 biogas additionally comprises H2S as combustible gas.
EP
Figure 12. The effect of preheated air on flame temperatures of biogases
4.4. Emission distributions
AC C
4.4.1. CO emission
Carbon monoxide (CO) is incomplete combustion product. For this reason, in combustion process, CO emissions have to be examined in order to understand whether the combustion process is efficiently completed or not. Axial CO emissions of biogases are presented in Figure 13. CO levels increase rapidly in flame zone as can be seen in Figure 13. Because, chemical reactions are not generally completed in the flame zone. Then, CO levels decrease gradually towards the combustor outlet because of complete of combustion process. The maximum CO levels show up under the Phase 3 biogas combustion conditions due to including high amount of C atoms. The other CO formation levels take place depending on C atoms as can be seen in Figure 9.
M AN U
Figure 13. Axial CO gradients of biogases
SC
RI PT
ACCEPTED MANUSCRIPT
4.4.2. CO2 emission
The characteristics of CO2 emissions upon combustion of biogases have been also predicted and analyzed and are given in Figure 14. The results show that general CO2 distributions of biogases are very similar through the combustor. Because, amounts of methane and carbon
TE D
dioxide in biogases are nearly same. The amounts of CO2 in percentages decrease rapidly for all biogases because combustion has not yet completed in flame zone, and then, these
AC C
EP
percentages increase gradually and finally, remain the same value till the combustor outlet.
Figure 14. Axial CO2 gradients of biogases
ACCEPTED MANUSCRIPT 4.4.3. The effect of H2S amount in biogas on SO2 emission The effect of H2S as a combustible and toxic component on SO2 emissions is a very important parameter in biogas combustion system as biogases include a trace amount of hydrogen sulphur depending on production method. Because of that, this emission has been examined in the present modelling. Figure 15 shows SO2 distributions of biogases through the combustor.
RI PT
10 ppm, 300 ppm and 600 ppm values represent amount of H2S in biogas in Figure 15. It is
M AN U
SC
concluded that SO2 emissions increase as amount of H2S in biogas is increased.
TE D
Figure 15. Axial SO2 gradients of biogases
5. Conclusions
EP
Numerical investigation of combustion characteristics of biogases has been performed via the newly generated burner in the present study. The effects of the preheated combustion air on
AC C
the flame temperatures and H2S amount in biogases on SO2 emissons have also been studied within these predictions.
It has been found out that the predictions are in good agreement with the measurements when The PDF/Mixture Fraction combustion model, the P-1 radiation model and the k-Ɛ standard turbulence model have been used during the modellings. It may be said that the k-Ɛ standard turbulence model gives better predictons compared to the other turbulence models for this modelling study. On the other hand it may be declared that the P-1 radiation model is quite appropriate for this modelling work as it can be easily applied to complicated geometries.
ACCEPTED MANUSCRIPT Combustion characteristics of various biogas types containing different amounts of combustible and non-combustible gases have been determined in the present study. It is concluded that the maximum flame temperature emerges during the phase 3 biogas combustion and is predicted to be very close to the methane flame temperature due to its
RI PT
calorific value. It means that phase 3 types of biogas may be used as an alternative fuel to methane or natural gas in industrial natural gas-fired furnaces, boilers and etc. It is revealed that the general temperature trends of methane and biogases combustion are nearly the same at radial positions towards the combustor outlet.
SC
It is also concluded that the flame temperatures of biogases increase as the combustion air is preheated before entering the burner. It has been determined that the maximum CO forms
M AN U
during the Phase 3 biogas combustion due to high amount of methane in comparison to the other biogas phases. It is demonstrated that the maximum CO2 formation shows up during the combustion of Phase 1 and Phase 4 biogases due to containing high amount of CO2. It is predicted that SO2 emissions increase as amount of H2S content in biogas is increased as expected.
TE D
Therefore, it may be declared that biogas combustion may be modelled by the use of aforementioned models in industrial applications such as furnaces and boilers. From the comparison of combustion and emission characteristics of biogases with methane, it can also
EP
be concluded that the studied burner can be used as an biogas combustion system in industrial applications when it is scaled up for higher thermal power needs.
AC C
Acknowledgements
The authors gratefully acknowledge to TUBITAK (Project Number: 114M668) and Gazi University (Project Number: 07/2016-01) for their technical and financial supports. Nomenclature
3D: three dimensional H2S: hydrogen sulphur PDF: probability density function CFD: computational fluid dynamics λ: excess air ratio
ACCEPTED MANUSCRIPT SO2: sulphur dioxide CO2: carbon dioxide CO: carbon monoxide NOX: nitrogen oxide CH4: methane N2: nitrogen O2: oxygen ρ = density, kg/m3 Φ = dependent variable u = velocity of the mixture, m/s Γ = transport coefficent
f = mixture fraction mF = mass fraction of fuel mO = mass fraction of oxidant ̅ = time averaged value of f
M AN U
Ф = source term of the transport equation
SC
RI PT
LCV: lower calorific value
σt = constant used in the mixture fraction/PDF model
TE D
'
′ = mixture fraction variance
3 = constant used in the mixture fraction/PDF model Cg = constant used in the mixture fraction/PDF model
EP
Cd = constant used in the mixture fraction/PDF model
AC C
RNG = renormalization group theory References [1]
E. H. M. Dirkse, Biogas upgrading using the DMT TS-PWS® Technology, Report,
Page 2-12, DMT Environmental Technology.
[2]
B. Adouane, W. de Jong, J. P. van Buijtenen, G. Vitteveen, Fuel-NOx emissions
reduction during the combustion of LCV gas in an air staged Winnox-TUD combustor, Appl. Therm. Engineering 30 (2010) 1034-1038. [3]
P. R. Bhoi, S. A. Channiwala, Emission characteristics and axial flame
temperature
distribution of producer gas fired premixed burner, Biomass and Bioenergy, (2009) 469-477.
33
ACCEPTED MANUSCRIPT [4]
S. E. Hosseini, G. Bagheri, M. A. Wahid, Numerical investigation of biogas flameless combustion, Energy Convers. Manag. 81 (2014) 41-50.
[5]
C. V. Huynh, S. C. Kong, Combustion and NOx emissions of biomass-derived syngas under various gasification conditions utilizing oxygen-enriched-air and steam, FUEL 107 (2013) 457-464. T. Leung, I. Wierzba, The effect of hydrogen addition on biogas non-premixed jet
RI PT
[6]
flame stability in a co-flowing air stream, Int. J. Hydrogen Energy 33 (2008) 3856– 3862. [7]
M. İlbaş, S. Karyeyen, Modelling of combustion performances and emission
SC
characteristics of coal gases in a model gas turbine combustor, Int. J. Energy Research 38 (2014) 1171-1180. [8]
S. Sethuraman, C. V. Huynh, S. C. Kong, Producer Gas Composition and NOx
M AN U
Emissions from a Pilot-Scale Biomass Gasification and Combustion System Using Feedstock with Controlled Nitrogen Content, Energy&Fuels 25 (2011) 813–822. [9]
H. N. Somehsaraei, M. M. Majoumerda, P. Breuhausb, M. Assadia, Performance analysis of a biogas-fueled micro gas turbine using a validated thermodynamic model, Appl. Therm. Engineering 66 (2014) 181-190.
S. Chen, C. Zheng, Counterflow diffusion flame of hydrogen-enriched biogas
TE D
[10]
under
MILD oxy-fuel condition, Int. J. Hydrogen Energy 36 (2011) 15403-15413. [11]
H. Nikpey, M. Assadi, P. Breuhaus, P.T. Mørkved, Experimental evaluation and ANN modeling of a recuperative micro gas turbine burning mixtures of natural gas
and
[12]
EP
biogas, Appl. Energy 117 (2014) 30-41. H. Selim, A.K. Gupta, A. Al Shoaibi, Effect of CO2 and N2 concentration in acid gas stream on H2S combustion, Appl. Energy 98 (2012) 53-58. C. J. Mordaunt, W. C. Pierce, Design and preliminary results of an atmospheric-
AC C
[13]
pressure model gas turbine combustor utilizing varying CO2 doping concentration in
CH4 to emulate biogas combustion, FUEL 124 (2014) 258-268.
[14]
S. E. Hosseini, M. A. Wahid, Development of biogas combustion in combined heat
and power generation, Renew. Sustainable Energy Rev. 40 (2014) 868-875. [15]
Y. Lafay, B. Taupin, G. Martins, G. Cabot, B. Renou, A. Boukhalfa, Experimental study of biogas combustion using a gas turbine configuration, Experimental in Fluids 43 (2007) 395-410.
[16]
S. Jahangirian, A. Engeda, I. S. Wichman, Thermal and Chemical Structure of Biogas Counterflow Diffusion Flames, Energy&Fuels 23 (2009) 5312-5321.
ACCEPTED MANUSCRIPT [17]
S. E. Hosseini, M. A. Wahid, Effects of burner configuration on the characteristics of biogas flameless combustion, Combust. Sci. Technol. 187 (2015) 1240-1262.
[18]
M.E. Feyz, J.A. Esfahani, I. Pishbin, S.M.R. Modarres Razavi, Effect of recess length on the flame parameters and combustion performance of a low swirl burner, Appl. Therm. Engineering 89 (2015) 609-617. K. M. Saqr, H. S. Aly, M. M. Sies, M. A. Wahid, Effect of free stream turbulence on
RI PT
[19]
NOx and soot formation in turbulent diffusion CH4-air flames, Int. J. Heat and Mass Transf. 37 (2010) 611-617. [20]
M. İlbaş, S. Karyeyen, A numerical study on combustion behaviours of hydrogen-
15218-15226. [21]
H. K. Versteeg, W. Malalasekera, An introduction to computational fluid Dynamics,
M. Ilbas, Studies of ultra low NOX burner. PhD Thesis: University of Wales, College of Cardiff, UK, 1997.
[23]
M AN U
Second Edition, PEARSON Prentice Hall, 1995. [22]
SC
enriched low calorific value coal gases, Int. J. Hydrogen Energy 40 (2015)
M. Ilbas, The effect of thermal radiation and radiation models on hydrogen– hydrocarbon combustion modelling, Int. J. Hydrogen Energy 30 (2005) 1113-1126. İ. Yılmaz, Effect of swirl number on combustion characteristics in a natural gas
TE D
[24]
diffusion flame, J. Energy Resour. Techn. 135 (2013) 042204, 1-8.
EP
Fluent Incorporated. Fluent 13.0, User’s Guide, 2011.
AC C
[25]
ACCEPTED MANUSCRIPT TITLE: 3D NUMERICAL MODELLING OF TURBULENT BIOGAS COMBUSTION IN A NEWLY GENERATED 10 KW DIFFUSION FLAME OF BURNER
RI PT
SC M AN U TE D EP
• • •
The newly generated burner has been described in detail. The PDF/Mixture Fraction combustion model, standard k-Ɛ turbulence model and P-1 radiation model have been used in order to model biogases. Model validation has been performed. Temperature and emission distributions of the biogases have been obtained. The effects of the preheated air and H2S amount in the biogases have been investigated.
AC C
• •
S1743-9671(16)30488-3
DOI:
10.1016/j.joei.2016.10.004
Reference:
JOEI 276
To appear in:
Journal of the Energy Institute
Received Date: 22 August 2016 Revised Date:
18 October 2016
Accepted Date: 19 October 2016
Please cite this article as: M. İlbaş, M. Şahin, S. Karyeyen, 3D Numerical modelling of turbulent biogas combustion in a newly generated 10 kw burner, Journal of the Energy Institute (2016), doi: 10.1016/ j.joei.2016.10.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 3D NUMERICAL MODELLING OF TURBULENT BIOGAS COMBUSTION IN A NEWLY GENERATED 10 KW BURNER Mustafa İLBAŞ*1, Murat ŞAHİN** and Serhat KARYEYEN* *
Gazi University, Technology Faculty, Department of Energy Systems Engineering, 06500,
**
RI PT
Teknikokullar/Ankara/TURKEY Gazi University, Graduate School of Natural and Applied Science, 06500, Teknikokullar, Ankara/TURKEY
SC
Abstract
This study concentrates on the 3D numerical modelling of combustion of different biogases in a generated burner and combustor. The main goal of this study is to investigate the
M AN U
combustion characteristics (such as temperature and emissions) of biogases through a combustor due to depletion of natural gas. Moreover, the effect of the preheated air on flame temperatures of biogases have been studied in the present study. Finally, the effect of H2S amount in biogas on SO2 emissions has been investigated within these predictions. The numerical modelling of turbulent diffusion flames has been performed by using the standard
TE D
k–ε model of turbulent flow, the PDF/Mixture Fraction combustion model and P-1 radiation model in the combustor. A CFD code has been used for all predictions. Temperature gradients have been determined on axial and radial directions for better understanding combustion characteristics of biogases. Modelling has been studied for thermal power of 10 kW and
EP
excess air ratio of λ= 1.2 for each biogas combustion. The first finding is that combustion of biogases is possible via the newly generated burner. Moreover, the results show that the one of biogas is very close to methane in terms of temperature distributions in the combustor due
AC C
to including high amount of methane compared to other biogases. It is also concluded that the flame temperatures of biogases increase with preheating the combustion air as expected. It is finally revealed that SO2 emissions increase as amount of H2S in biogas is increased through the combustor.
Keywords: Biogas, Burner, Combustion, Emission, CFD Modelling
1
Corresponding author. Email: [email protected]; Tel: + 90 312 202 86 09; Fax: + 90 312 202 89 47
ACCEPTED MANUSCRIPT 1. Introduction Although fossil fuels are still used in power generation systems, buildings and elsewhere, these have two problems. First, fossil fuels are restricted and are not equally distributed all over the World. Scientists estimate that coal, crude oil and natural gas will be rapidly depleted in the near future. Second, fossil fuels are responsible for global warming. Because, CO2 is
RI PT
majorly released when fossil fuel are burned in any system. Because, there are aforementioned problems, the scientists try to find new energy resources such as solar energy, wind energy, hydrogen energy, biomass and the others.
SC
The growing interest in the use of biogas has promoted nowadays. Biogas arising from effluent, municipal solid waste, residuals, gasification of biomass are very appropriate for power generation, heating systems and elsewhere. Biogas consists chiefly of methane and
M AN U
carbondioxide as well as it has a trace amount of nitrogen, hydrogen sulphur and oxygen depending on production or gasification methods [1]. Methane in biogas has high amount of energy as major component of natural gas. However, the other components are noncombustible.
TE D
Biogas combustion has to be investigated for better understanding of combustion characteristics of biogases. There are some studies about combustions of biogas or blending fuels in the literature. Adouane et al. [2] have conducted an experimental study related to reduction of NOX emissions arising from fuel-bound nitrogen. They conclude that ammonia in
EP
the fuel drastically affects NOX formation. Bhoi and Channiwala [3] have experimentally investigated emission characteristics and axial flame temperature distributions of producer gas including high amount of nitrogen by varying thermal inputs and equivalance ratios. The
AC C
results show that the maximum temperature emerges in flame front for this gas. Hosseini et al. [4] have performed combustion characteristics of biogas under flameless combustion conditions. Hosseini et al. estimate that biogas is not appropriate for any furnaces due to its low calorific value (LCV). Huynh and Kong [5] have determined NOx emissions coming from syngas combustion obtained three different biomass feedstock under different conditions. It can be concluded that highest NOx levels form under syngas combustion obtained seed corn as the seed corn has the highest nitrogen content. Leung and Wierzba [6] have investigated the effect of hydrogen addition on stability limits of non-premixed biogas flames. It is demonstrated that hydrogen addition on the enhancement of biogas highly affects.
ACCEPTED MANUSCRIPT İlbaş and Karyeyen [7] have modelled the combustion characteristics of coal gases including high amount of hydrogen or nitrogen depending on gasification or carbonization methods. They reveal that the coke oven gas is very suitable in combustion systems in terms of combustion performances (e.g. temperature values). Sethuraman et al. [8] have focused on the effects of nitrogen content in biomass feedstock on the producer gas composition and NOx
RI PT
emissions. They have found that there are relationships between nitrogen in biomass, ammonia in the producer gas, and NOx emissions in the flue gas. Somehsaraei et al. [9] have examined the fuel flexibility and performance analysis of biogas in micro gas turbines. When minor modifications to fuel valves and compressor were taken place, They have claimed that
SC
these modifications were assumed to allow engine operation with the simulated biogas composition. Chen and Zheng [10] have predicted hydrogen-enriched biogas MILD oxy-fuel conditions. It is found that biogas flame can be sustained under the MILD oxy-fuel
M AN U
combustion. Nikpey et al. [11] have investigated the impact of biogas and natural gas on combustion performance and emissions in a micro gas turbine. The results indicate that there is almost no change in electrical efficiency comparing with the natural gas fired case. Selim et al. [12] have studied effect of CO2 and N2 concentration during H2S combustion. They have found that injection of CO2 highly affected temperature values. Mordaunt and Pierce [13]
TE D
have designed a combustion device and investigated the effects of CO2 on combustion of CH4 and emissions. Hosseini and Wahid [14] have studied combustion characteristics of biogas under hydrogen-enriched combustion conditions. The results show that NOX formation increases as the flame temperature increases. Lafay et al. [15] have compared stability
EP
combustion domains, flame structures and Dynamics between methane and biogas flames. They have revealed that laminar flame speed strongly depends on fuel composition. Jahangirian et al. [16] have examined chemical and thermal influences of biogas CO2 content.
AC C
It is demonstrated that the presence of CO2 in the fuel highly affects NOX emissions (reducing). Hosseini and Wahid [17] have modelled biogas under flameless combustion conditions in a tangential burner. The results indicate that the temperature uniformity in the tangential flameless burner is more than coaxial configurations.
Some studies related to methane combustions were also reviewed as the methane combustion has been also modelled in the present study in order to validate the modellings. Feyz et al. [18], for example, have researched the effect of recess length on the combustion performance of methane flame. They have observed that the flame temperature increase as the recess length is increased. Saqr et al. [19] have modelled effect of free stream turbulence on NOX
ACCEPTED MANUSCRIPT formation of methane flames. They have reported that change in free stream turbulence affects NOX formation level considerably. In addition to these studies, there is a study regarding combustion of low calorific value gas by authors. İlbaş and Karyeyen [20] have also studied combustion behaviours of low calorific value coal gases by enriching hydrogen in order to improve their combustion characteristics in that study. They have concluded that
RI PT
hydrogen addition highly affects the flame temperature of the low calorific value coal gases.
Although there are some aforementioned studies related to biogas or blending fuels combustion, there needs to be extremely studied combustion characteristics of biogas.
SC
Because, there are a lot of problems concerning combustion of biogas such as flame stability, flame propogation, flame temperature and emissions. However, the authors think that the newly generated of burner can contribute flame stability and relative high flame temperature
M AN U
of biogas because it has radial fuel inlets and angular and straight air inlets. These inlets can provide better fuel-air mixture and as a result of this, better combustion performance. If they are all to be considered, this study focuses on combustion performances and emission characteristics of biogases including different gas components by means of the newly
2. CFD Modelling
TE D
generated of burner under thermal power of 10 kW and excess air ratio of 1.2
2.1. Descriptions of burner and model combustor 3D CFD modelling has been performed by using the newly generated diffusion flame of
EP
burner in order to model combustions of methane (as baseline fuel) and biogases in the present study. This burner is shown in Figure 1a. A and B in Figure 1a represent the fuel and air inlets, respectively. It has also turbulators with 15° angles in the side of air stream in order
AC C
to provide flame stabilization. Moreover, the fuel outlet of this burner has radial inlets for better mixing of air and fuel can be seen in Figure 1b. Hydraulic diameters of fuel and air inlets are 6 mm and 32 mm. Also, hydraulic diameter of the combustor outlet is 110 mm. Fuel and air inlet temperatures, combustion gauge pressures are also fixed at 283 K and 21 mbar, respectively. The combustor outlet is open to atmosphere.
RI PT
ACCEPTED MANUSCRIPT
(a)
(b)
SC
Figure 1a. Fuel and air inlets in the burner; Figure 1b. Radial fuel inlets
The burner and combustor mesh structure used in this study can be seen in Figure 2. This
EP
TE D
M AN U
combustor is sudden expansion type and has length of 100 cm and diameter of 40 cm.
AC C
Figure 2. The burner and combustor mesh structure
2.2. Governing Equations
The mathematical modelling described for the gas mixture fuel combustion is based on the steady-state condition, 3D continuity, momentum, energy and species equations. The general form of transport equation is below [21]: Ф
+ Ф = Ф + Ф
(1)
ACCEPTED MANUSCRIPT where Ф represents the dependent variables. shows the transport coefficient for variable Ф
and Ф states the source term of the transport equaiton for Ф.
The Mixture Fraction/PDF Model has been used as combustion model in this study. This combustion model consists of the solution of transport equations for a single conserved scalar
RI PT
(the mixture fraction). In this model transport equations for individual species are not solved. Instead, individual component concentrations for species of interest are derived form the predicted mixture fraction distribution. The interaction of turbulence and chemistry are
SC
accounted for with the help of a probability density function (PDF) [22].
The PDF modelling approach was specifically developed for the simulation of turbulent
of the local fuel mass fraction as: =
M AN U
diffusion flames. For a fuel/oxidant system, the mixture fraction, can be expressed in terms
(2)
TE D
Where and are mass fraction of fuel and oxidant respectively. The mixture fraction, is a conserved quantity whose value at each point in the flow domain is computed by the solution of the following conservation equation for the time averaged
̅
̅
=
!
"# ̅
$#
% +
(3)
AC C
+
EP
value of in the turbulent flow field [22].
where is the source term which is due solely to transfer of mass into the gas phase from liquid fuel droplets.
In addition to solving for the mean mixture fraction, a conservation equation is solved for the '
mixture fraction variance, ′ which is used in the closure model describing turbulencechemistry interactions [22]: )
(
+
)
(
)
"# (
= *$
#
)
(
'
1
+ + ,- . * + − ,0 2 ′
'
(4)
ACCEPTED MANUSCRIPT where 3 , ,- and ,0 are constants used in the mixture fraction/PDF model. The radiation heat transfer happens at high temperatures. In combustors, the flame temperature is generally high (1000-1600°C) especially at near stoichiometric combustion
RI PT
conditions. Therefore, the heat transfer from swirl combustors is significant. Thus, It is essential to include the radiation model for a more accurate prediction of the temperature distribution in combustors [22]. The P-1 model is the simplest case of the more general P–N model. The P-1 model has some advantages over the other radiation models. With the P-1
SC
model, it is easy to solve with little CPU demand, the model includes the effect of scattering. For combustion applications where the optical thickness is large, the P-1 model works reasonably well. The P-1 model can be easily applied to complicated geometries with
M AN U
curvilinear coordinates [23]. For these reason, P-1 radiation model was used for a more accurate prediction of biogas combustion in the present study.
Numerical simulation of the turbulence reacting flow in the newly generated of burner and the combustor was carried out using FLUENT code. It uses the finite volume method to solve the
TE D
governing equations for a fluid and species and provides the capability to use different physical models such as inviscid or viscous, laminar or turbulent and the other [24]. The Simple algorithm was applied for the pressure–velocity coupling. Second order upwind scheme was used for the convection terms in all transport equations, and the second order
EP
method was used for the pressure discretization. The solution convergence is assumed when all of the residuals parameters fall below 10-5. The numerical prediction was carried out with
AC C
the following assumptions: The flow was steady-state, three dimensional and turbulent.
2.3.Definition of biogases
Components of biogases consumed in this study and methane as baseline fuel are given in Table 1.These biogases are defined as phases 1, 2, 3 and 4 depending on contents in this study.
ACCEPTED MANUSCRIPT Table 1. Biogas components Methane
Phase 1
Phase 2
Phase 3
Phase 4
(%)
(%)
(%)
(%)
100
55
60
65
55
CO2
-
43,1
38
33
43,1
N2
-
1,53
1,5
1,3
1,53
H2S
-
10 ppm
0 ppm
O2
-
0,3
0,5
8040,00
4422,06
4824,00
(kcal/m3)
0 ppm
0 ppm
0,7
0,37
5226,00
4422,00
SC
Calorific value
RI PT
CH4
M AN U
3. Experimental Setup
In the experimental part of the present study, temperature values have been measured on some axial and radial locations through the combustor in order to validate the modellings. The
AC C
EP
TE D
schematic view of the combustion system is shown in Figure 3.
Figure 3. The layout of the combustion system
A general view of the combustion chamber is illustrated in Figure 4. The length and diameter of the combustion chamber are fixed at 100 cm and 40 cm, respectively. The combustion
ACCEPTED MANUSCRIPT chamber consists of a sight glass made of tempered glass and five measuring ports. These measuring ports are placed on the combustor wall in order to measure temperature values on axial and radial positions within the combustor. The first measuring port is designed to determine the flame characteristics of the biogases and location of this measuring port is 10 cm away from the combustor inlet. The other four measuring ports are placed at intervals of
M AN U
SC
RI PT
20 cm from the first measuring port to the combustion chamber outlet.
TE D
Figure 4. The combustion chamber and the burner
Ceramic coated R-type thermocouples capable of withstanding high temperatures up to 1700°C were used to measure temperature values through the combustion chamber. The diameters of these thermocouples are 5 mm. At high temperatures, the radiation correction is
EP
necessary the temperature levels precisely. In particular, radiative heat transfer takes places from the flame to the combustor wall. Mean bulk velocities were calculated theoretically in
AC C
the flame front considering fuel and air velocities. Then, temperature correction was performed for each axial and radial measurements. It was computed to be between 2–50 °C depending on the measured axial, radial and wall temperatures in the combustor.
4. Results and Discussion 4.1. Mesh independence
Mesh independence is very important situation in numerical studies as excessive mesh structure needs advanced computer technology, long computational time and etc. For this reason, mesh independence can be taken into account before predictions. The methane has been selected as baseline fuel in order to determine fine mesh structure. Grid structures
ACCEPTED MANUSCRIPT including 311812, 447198, 613947, 893508, 1523435 elements have been tested in the predictions. The influence of the mesh refinement on the axial temperature profiles for these mesh structures is shown in Figure 3. Because there is no difference between predictions at 613947 elements mesh structure and further two mesh structures, the 613947 elements mesh
M AN U
SC
RI PT
structure was selected as the fine mesh structure for further 3D modelling cases.
Figure 5. Axial temperature gradients for different mesh structure
TE D
4.2. Experimental validation and the effect of the turbulence models Predictions are compared with experimental measurements. Axial and radial temperature values of methane flames have been measured at 10 cm, 30 cm, 50 cm, 70 cm and 90 cm axial distances through the combustor. Combustion conditions such as stream flow rates, inlet
EP
temperatures and pressures, fuel (methane for validation) and ect. have been identically selected for better capturing the predictions (thermal power of 10 kW and excess air ratio of
AC C
1.2). In addition to the validation, the effect of the turbulence models on temperature profiles have been investigated in order to determine more appropriate turbulence model for this modelling. It is clearly seen in Figure 6, the axial temperature profiles are in good agreement with the experimental data. There is a very slight difference between measurements and the predictions through the combustor.
SC
RI PT
ACCEPTED MANUSCRIPT
M AN U
Figure 6. Axial temperature gradients for different turbulence models and model validation with experimental data
Figure 7 shows the radial temperature gradients of methane at 10 cm and 50 cm axial positions. The predictions are in satisfactorily agreement with the measurements as can be seen in Figure 7. Moreover, the effect of the turbulence models on temperature distributions
TE D
has been studied as previously mentioned. One of the most prominent turbulence models, the k-Ɛ standard turbulence model, is implemented in most general purpose CFD codes and is considered the industry standard model. It has proven to be stable and numerically robust and has a well established regime of predictive capability. For general purpose simulations, the
EP
model offers a good compromise in terms of accuracy and robustness [25]. Although general trends of the predictions are very similar for three turbulence models (Figure 6 and Figure 7),
AC C
the best consistency between the measurements and the predictions emerged while using the k-Ɛ standard turbulence models. Therefore, k-Ɛ standard turbulence model has been chosen to be used further 3D modellings.
RI PT
ACCEPTED MANUSCRIPT
models and model validation with the experimental data
SC
(a) (b) Figure 7a. Radial temperature gradients at 10 cm axial distance for different turbulence
Figure 7b. Radial temperature gradients at 50 cm axial distance for different turbulence
M AN U
models and model validation with the experimental data.
Predictions are also compared to the measurements in axial and radial directions for biogas (phase 1). Predicted and measured temperature distributions of biogas (phase 1) combustion are shown in Figure 8 and Figure 9. As can be seen in Figure 8 and Figure 9, the predictions
AC C
EP
combustion conditions.
TE D
in both directions are in good agreement with the measurements under biogas (phase 1)
Figure 8. Axial temperature distributions of biogas (phase 1) combustion
M AN U
SC
50 cm
RI PT
ACCEPTED MANUSCRIPT
90 cm
AC C
EP
TE D
30 cm
10 cm
70 cm
Burner walls Swirl vanes
Figure 9. Radial temperature distributions of biogas (phase 1) combustion Fuel Inlet
Air Inlet
ACCEPTED MANUSCRIPT 4.3. Temperature distributions 4.3.1. Axial and radial temperatures of biogases Figure 10 depicts the axial temperature distributions of different biogas contents and methane. The maximum temperature level emerges under the methane combustion through the combustor due to its calorific value. The temperature levels of phase 3 biogas are very closer
RI PT
to that of methane in the flame zone. The authors estimate that this situation is related to calorific value as phase 3 includes high amount of methane comparison to the other biogases. All temperature distributions increase in the flame zone initially and then, decrease gradually
TE D
M AN U
SC
towards the combustor outlet due to convective and radiative heat transfers.
EP
Figure 10. Axial temperature gradients of biogases and methane Radial temperature profiles between center of the combustor and combustor walls are shown
AC C
in Figure 11. Initially, temperature profiles at axial distance of 10 cm decrease rapidly and remain the same towards the combustor wall as this combustor type is sudden expansion. When general temperature trends are evaluated, the temperature levels of methane are generally high comparison to the these of biogases. In particular, tcombustion performance of phase 3 biogas is relatively high in comparison to the other biogases. This circumstance may be explained the presence of more CH4 as combustible component in phase 3. However, the general trends for all biogases are very similar at radial directions. Moreover, It is revealed that the temperature values between center and combustor wall are getting closer towards the combustor outlet because the flame propogates more effectively under all combustion conditions.
M AN U
SC
50 cm
RI PT
ACCEPTED MANUSCRIPT
90 cm
AC C
EP
TE D
30 cm
10 cm
70 cm
Burner walls Swirl vanes
Figure 11. Radial temperature gradients of biogases and methane through the combustor Fuel Inlet
Air Inlet
ACCEPTED MANUSCRIPT 4.3.2. The effect of preheated air on flame temperature Combustion air has been preheated before entering the burner in order to understand the effects of this circumstance on flame temperatures of biogases in the present study. This influence is clearly shown in Figure 12. As It can be seen in Figure 12, the flame temperatures of biogases increase as the combustion air is preheated before entering the burner air inlet.
RI PT
The maximum flame temperatures for all preheat conditions are high under the phase 3 biogas conditions due to including high amount of methane compared to the other biogases as It was mentioned before. Moreover, there are slight differences between the flame temperatures of
TE D
M AN U
SC
Phase1 and 4 biogases. Phase 1 biogas additionally comprises H2S as combustible gas.
EP
Figure 12. The effect of preheated air on flame temperatures of biogases
4.4. Emission distributions
AC C
4.4.1. CO emission
Carbon monoxide (CO) is incomplete combustion product. For this reason, in combustion process, CO emissions have to be examined in order to understand whether the combustion process is efficiently completed or not. Axial CO emissions of biogases are presented in Figure 13. CO levels increase rapidly in flame zone as can be seen in Figure 13. Because, chemical reactions are not generally completed in the flame zone. Then, CO levels decrease gradually towards the combustor outlet because of complete of combustion process. The maximum CO levels show up under the Phase 3 biogas combustion conditions due to including high amount of C atoms. The other CO formation levels take place depending on C atoms as can be seen in Figure 9.
M AN U
Figure 13. Axial CO gradients of biogases
SC
RI PT
ACCEPTED MANUSCRIPT
4.4.2. CO2 emission
The characteristics of CO2 emissions upon combustion of biogases have been also predicted and analyzed and are given in Figure 14. The results show that general CO2 distributions of biogases are very similar through the combustor. Because, amounts of methane and carbon
TE D
dioxide in biogases are nearly same. The amounts of CO2 in percentages decrease rapidly for all biogases because combustion has not yet completed in flame zone, and then, these
AC C
EP
percentages increase gradually and finally, remain the same value till the combustor outlet.
Figure 14. Axial CO2 gradients of biogases
ACCEPTED MANUSCRIPT 4.4.3. The effect of H2S amount in biogas on SO2 emission The effect of H2S as a combustible and toxic component on SO2 emissions is a very important parameter in biogas combustion system as biogases include a trace amount of hydrogen sulphur depending on production method. Because of that, this emission has been examined in the present modelling. Figure 15 shows SO2 distributions of biogases through the combustor.
RI PT
10 ppm, 300 ppm and 600 ppm values represent amount of H2S in biogas in Figure 15. It is
M AN U
SC
concluded that SO2 emissions increase as amount of H2S in biogas is increased.
TE D
Figure 15. Axial SO2 gradients of biogases
5. Conclusions
EP
Numerical investigation of combustion characteristics of biogases has been performed via the newly generated burner in the present study. The effects of the preheated combustion air on
AC C
the flame temperatures and H2S amount in biogases on SO2 emissons have also been studied within these predictions.
It has been found out that the predictions are in good agreement with the measurements when The PDF/Mixture Fraction combustion model, the P-1 radiation model and the k-Ɛ standard turbulence model have been used during the modellings. It may be said that the k-Ɛ standard turbulence model gives better predictons compared to the other turbulence models for this modelling study. On the other hand it may be declared that the P-1 radiation model is quite appropriate for this modelling work as it can be easily applied to complicated geometries.
ACCEPTED MANUSCRIPT Combustion characteristics of various biogas types containing different amounts of combustible and non-combustible gases have been determined in the present study. It is concluded that the maximum flame temperature emerges during the phase 3 biogas combustion and is predicted to be very close to the methane flame temperature due to its
RI PT
calorific value. It means that phase 3 types of biogas may be used as an alternative fuel to methane or natural gas in industrial natural gas-fired furnaces, boilers and etc. It is revealed that the general temperature trends of methane and biogases combustion are nearly the same at radial positions towards the combustor outlet.
SC
It is also concluded that the flame temperatures of biogases increase as the combustion air is preheated before entering the burner. It has been determined that the maximum CO forms
M AN U
during the Phase 3 biogas combustion due to high amount of methane in comparison to the other biogas phases. It is demonstrated that the maximum CO2 formation shows up during the combustion of Phase 1 and Phase 4 biogases due to containing high amount of CO2. It is predicted that SO2 emissions increase as amount of H2S content in biogas is increased as expected.
TE D
Therefore, it may be declared that biogas combustion may be modelled by the use of aforementioned models in industrial applications such as furnaces and boilers. From the comparison of combustion and emission characteristics of biogases with methane, it can also
EP
be concluded that the studied burner can be used as an biogas combustion system in industrial applications when it is scaled up for higher thermal power needs.
AC C
Acknowledgements
The authors gratefully acknowledge to TUBITAK (Project Number: 114M668) and Gazi University (Project Number: 07/2016-01) for their technical and financial supports. Nomenclature
3D: three dimensional H2S: hydrogen sulphur PDF: probability density function CFD: computational fluid dynamics λ: excess air ratio
ACCEPTED MANUSCRIPT SO2: sulphur dioxide CO2: carbon dioxide CO: carbon monoxide NOX: nitrogen oxide CH4: methane N2: nitrogen O2: oxygen ρ = density, kg/m3 Φ = dependent variable u = velocity of the mixture, m/s Γ = transport coefficent
f = mixture fraction mF = mass fraction of fuel mO = mass fraction of oxidant ̅ = time averaged value of f
M AN U
Ф = source term of the transport equation
SC
RI PT
LCV: lower calorific value
σt = constant used in the mixture fraction/PDF model
TE D
'
′ = mixture fraction variance
3 = constant used in the mixture fraction/PDF model Cg = constant used in the mixture fraction/PDF model
EP
Cd = constant used in the mixture fraction/PDF model
AC C
RNG = renormalization group theory References [1]
E. H. M. Dirkse, Biogas upgrading using the DMT TS-PWS® Technology, Report,
Page 2-12, DMT Environmental Technology.
[2]
B. Adouane, W. de Jong, J. P. van Buijtenen, G. Vitteveen, Fuel-NOx emissions
reduction during the combustion of LCV gas in an air staged Winnox-TUD combustor, Appl. Therm. Engineering 30 (2010) 1034-1038. [3]
P. R. Bhoi, S. A. Channiwala, Emission characteristics and axial flame
temperature
distribution of producer gas fired premixed burner, Biomass and Bioenergy, (2009) 469-477.
33
ACCEPTED MANUSCRIPT [4]
S. E. Hosseini, G. Bagheri, M. A. Wahid, Numerical investigation of biogas flameless combustion, Energy Convers. Manag. 81 (2014) 41-50.
[5]
C. V. Huynh, S. C. Kong, Combustion and NOx emissions of biomass-derived syngas under various gasification conditions utilizing oxygen-enriched-air and steam, FUEL 107 (2013) 457-464. T. Leung, I. Wierzba, The effect of hydrogen addition on biogas non-premixed jet
RI PT
[6]
flame stability in a co-flowing air stream, Int. J. Hydrogen Energy 33 (2008) 3856– 3862. [7]
M. İlbaş, S. Karyeyen, Modelling of combustion performances and emission
SC
characteristics of coal gases in a model gas turbine combustor, Int. J. Energy Research 38 (2014) 1171-1180. [8]
S. Sethuraman, C. V. Huynh, S. C. Kong, Producer Gas Composition and NOx
M AN U
Emissions from a Pilot-Scale Biomass Gasification and Combustion System Using Feedstock with Controlled Nitrogen Content, Energy&Fuels 25 (2011) 813–822. [9]
H. N. Somehsaraei, M. M. Majoumerda, P. Breuhausb, M. Assadia, Performance analysis of a biogas-fueled micro gas turbine using a validated thermodynamic model, Appl. Therm. Engineering 66 (2014) 181-190.
S. Chen, C. Zheng, Counterflow diffusion flame of hydrogen-enriched biogas
TE D
[10]
under
MILD oxy-fuel condition, Int. J. Hydrogen Energy 36 (2011) 15403-15413. [11]
H. Nikpey, M. Assadi, P. Breuhaus, P.T. Mørkved, Experimental evaluation and ANN modeling of a recuperative micro gas turbine burning mixtures of natural gas
and
[12]
EP
biogas, Appl. Energy 117 (2014) 30-41. H. Selim, A.K. Gupta, A. Al Shoaibi, Effect of CO2 and N2 concentration in acid gas stream on H2S combustion, Appl. Energy 98 (2012) 53-58. C. J. Mordaunt, W. C. Pierce, Design and preliminary results of an atmospheric-
AC C
[13]
pressure model gas turbine combustor utilizing varying CO2 doping concentration in
CH4 to emulate biogas combustion, FUEL 124 (2014) 258-268.
[14]
S. E. Hosseini, M. A. Wahid, Development of biogas combustion in combined heat
and power generation, Renew. Sustainable Energy Rev. 40 (2014) 868-875. [15]
Y. Lafay, B. Taupin, G. Martins, G. Cabot, B. Renou, A. Boukhalfa, Experimental study of biogas combustion using a gas turbine configuration, Experimental in Fluids 43 (2007) 395-410.
[16]
S. Jahangirian, A. Engeda, I. S. Wichman, Thermal and Chemical Structure of Biogas Counterflow Diffusion Flames, Energy&Fuels 23 (2009) 5312-5321.
ACCEPTED MANUSCRIPT [17]
S. E. Hosseini, M. A. Wahid, Effects of burner configuration on the characteristics of biogas flameless combustion, Combust. Sci. Technol. 187 (2015) 1240-1262.
[18]
M.E. Feyz, J.A. Esfahani, I. Pishbin, S.M.R. Modarres Razavi, Effect of recess length on the flame parameters and combustion performance of a low swirl burner, Appl. Therm. Engineering 89 (2015) 609-617. K. M. Saqr, H. S. Aly, M. M. Sies, M. A. Wahid, Effect of free stream turbulence on
RI PT
[19]
NOx and soot formation in turbulent diffusion CH4-air flames, Int. J. Heat and Mass Transf. 37 (2010) 611-617. [20]
M. İlbaş, S. Karyeyen, A numerical study on combustion behaviours of hydrogen-
15218-15226. [21]
H. K. Versteeg, W. Malalasekera, An introduction to computational fluid Dynamics,
M. Ilbas, Studies of ultra low NOX burner. PhD Thesis: University of Wales, College of Cardiff, UK, 1997.
[23]
M AN U
Second Edition, PEARSON Prentice Hall, 1995. [22]
SC
enriched low calorific value coal gases, Int. J. Hydrogen Energy 40 (2015)
M. Ilbas, The effect of thermal radiation and radiation models on hydrogen– hydrocarbon combustion modelling, Int. J. Hydrogen Energy 30 (2005) 1113-1126. İ. Yılmaz, Effect of swirl number on combustion characteristics in a natural gas
TE D
[24]
diffusion flame, J. Energy Resour. Techn. 135 (2013) 042204, 1-8.
EP
Fluent Incorporated. Fluent 13.0, User’s Guide, 2011.
AC C
[25]
ACCEPTED MANUSCRIPT TITLE: 3D NUMERICAL MODELLING OF TURBULENT BIOGAS COMBUSTION IN A NEWLY GENERATED 10 KW DIFFUSION FLAME OF BURNER
RI PT
SC M AN U TE D EP
• • •
The newly generated burner has been described in detail. The PDF/Mixture Fraction combustion model, standard k-Ɛ turbulence model and P-1 radiation model have been used in order to model biogases. Model validation has been performed. Temperature and emission distributions of the biogases have been obtained. The effects of the preheated air and H2S amount in the biogases have been investigated.
AC C
• •