- Email: [email protected]
Numerical simulation of CH4-H2-AIR non-premixed flame stabilized by a bluff body
Available online at www.sciencedirect.com
ScienceDirect Availableonline onlineatatwww.sciencedirect.com www.sciencedirect.com Available Energy Procedia 00 (2017) 000–000
ScienceDirect ScienceDirect
www.elsevier.com/locate/procedia
Energy (2017) 000–000 530–536 EnergyProcedia Procedia139 00 (2017) www.elsevier.com/locate/procedia
International Conference On Materials And Energy 2015, ICOME 15, 19-22 May 2015, Tetouan, Morocco, and the International Conference On Materials And Energy 2016, ICOME 16, 17-20 May 2016, La Rochelle, France The 15th International Symposium on District Heating and Cooling
Numerical simulation of CH4-H2-AIR non-premixed flame Assessing the feasibility thebody heat demand-outdoor stabilizedofbyusing a bluff temperature function for a 1long-term district 2 heat demand 1 forecast 12-
KHALADI Fatma Zohra , ALLICHE Mounir *, CHIKH Salah a,b,c a a b c c I. Andrić *, A. Pina , P. Ferrão , J. Fournier ., B. Lacarrière O. Le Corre Laboratoire de Transports Polyphasiques et Milieux Poreux (LTPMP), Université de ,Technologie Houari
a
Boumediene, Alger, Algeria
IN+ Center for Innovation, Physique Technology and Policy Research - Instituto SuperiorUniversité Técnico, Av. RoviscoQuartier Pais 1, 1049-001 Laboratoire de Mécanique et Modélisation Mathématique (LMP2M), de Médéa, Ain Dheb,Lisbon, Medea,Portugal Algeria b Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France c Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France
Abstract Abstract The aim of this study is to perform a numerical simulation of a non-premixed turbulent flame of methane-air enriched by hydrogen. The selected axisymmetric configuration is composed of a central injector of methane-hydrogen mixture surrounded heatingwhich networks are commonly theThe literature onesoftware of the most effective solutions for decreasing the byDistrict a bluff-body, is surrounded by a addressed co-axial airinjet. Ansys as CFX is used to solve the equations governing greenhouse gas emissions fromStokes the building sector. These require high investments which returned through heat turbulent reactive flow (Navier averaged in sense of systems Favre). The Turbulence is modeled usingarethe k-ε model. Thethe EDM sales. Dissipation Due to themodel), changedthen climate conditions building policies, heatwith demand the used future decrease, (Eddy the FRC model and (Finite Rate renovation Combustion) combined EDMin are to could modeling the prolonging the investment return period. combustion phenomena. The results show some concordance with the temperature profile given by experience to a hydrogen rate main scope of this paper is to assess the feasibility of using the heat demand – outdoor temperature function for heat demand ofThe 50%. forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 vary Published in both construction ©buildings 2017 Thethat Authors. by Elsevierperiod Ltd. and typology. Three weather scenarios (low, medium, high) and three district renovation scenarios were developed (shallow, intermediate, deep). 2015 To estimate the error, Peer-review under responsibility of the scientific committee of ICOME and ICOME 2016.obtained heat demand values were compared with results from a dynamic heat demand model, previously developed and validated by the authors. The results showed that when only weather change isEDM considered, margin errorCFX could be acceptable for some applications Keywords: Non-premixed combustion, Hydrogen enrichment, and FRC the models, CFD,ofAnsys (the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered). value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the 1.The Introduction decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and renovation scenarios considered). On the other hand, function intercept increased for 7.8-12.7% per decade (depending on the Over the past three decades, there has been considerable effort in the world to develop and introduce coupled scenarios). The values suggested could be used to modify the function parameters for the scenarios considered, and alternative transportation fuels to replace conventional fuels such as gasoline and diesel, environmental issues, most improve the accuracy of heat demand estimations.
notably air pollution and limited availability of conventional fuels are among the principle driving forces behind this © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling. * Corresponding author. Keywords: Heat demand; Forecast; Climate change E-mail address: [email protected] 1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of ICOME 2015 and ICOME 2016. 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the scientific committee of ICOME 2015 and ICOME 2016 10.1016/j.egypro.2017.11.249
2
KHALADI Fatma Zohra et al. / Energy Procedia 139 (2017) 530–536 F-Z. Khaladi et al. / Energy Procedia 00 (2017) 000–000
531
movement. Thus, if one tries to find for the definition of perfect fuel, hydrogen probably satisfies most of the desirable characteristics of such a fuel. Plentiful and clean burning, hydrogen has very high-energy content. [1, 2] Due to difficulties in conducting spatially resolved measurements of combustion characteristics in devices, the numerical simulation can be cost effective approach to study the combustion mechanism. In this work, Computational Fluid Dynamics (CFD) based numerical simulations have been performed to study the combustion of non-premixed and various aspects in order to use them for solving the realistic problems. This study allows us: • The understanding of the basics of Hydrogen-oxygen reaction mechanism, its combustion and the geometry of the cylindrical chamber used in this study is very important for simulating Methane-Air combustion system enriched by Hydrogen. • To develop a two-dimensional numerical mesh and flow model which adequately and accurately represent the physical model of combustion chamber and is simple enough to limit the amount of computational time for obtaining a solution. The objective of this study is to find and apply appropriate model that improve the simulation of combustion with the commercial CFD as Ansys CFX. • Generate numerical data/solutions, which correlate as much as possible with the experimental data for various conditions including equivalence ratios, mass flow rates of hydrogen-air mixture. In the same context, Zhuyin Ren [3] presents a numerical simulation of a non-premixed combustion flame methane-air enriched by hydrogen, stabilized by a recirculation zone created by an obstacle or a bluff-body : a simple approach in which we apply a different combustion model to see if the same results were been obtained. Zhuyin Ren [3] uses a reduced description with chemical tabs implemented in FLUENT is combined with the EDC model (Eddy Dissipation Concept) considers that a moderate or chemistry with model PDF (probability density function) for combustion. We decided, in the simulation of combustion, for EDM in its infinitely fast chemistry limited compared to the scale of the turbulent times. Then for model Combined EDM/FRC, firstly, that it is valid for several reactions classified from a low to a high number of Damköhler (slow or fast chemistry compared to the scale of the turbulence time) and secondly, reaction rates are firstly calculated for each model separately and then the minimum of both is used. This procedure is applied to each reaction separately, so that when the level in onestep, would be limited by the chemical kinetics, certain other steps would be limited by the turbulent mixing at the same time and in the same physical position. [4]. Our approach is a first step in choosing the appropriate mesh following a test of several types of mesh, with an infinitely fast chemistry, applying the EDM model for the combustion of methane-air flame. Then, we compare the methane-air flame at the hydrogen-air flame. In the other hand, we compare the application of EDM model to combined EDM/FRC model for simulating the combustion of an air- methane sulfonate flame enriched by hydrogen. The results of our calculations are presented in the form of temperature profiles. 1.1. Hydrogen as a fuel Hydrogen is a colorless, odorless, tasteless, and nonpoisonous gas under normal conditions on Earth. It typically exists as a diatomic molecule. Hydrogen is the most abundant element in the universe, accounting for 90 percent of the universe by weight. However, it is not commonly found in its pure form. [1, 2, 5] The properties of hydrogen are largely listed in reference [6] with conventional fuels i.e. Gasoline & Diesel and other alternative fuels such as CNG, LPG, and Biogas. Hydrogen has wide range of flammability in comparison with other fuels. One of the significant advantages is that hydrogen engine can run on a lean mixture. When engine is run on slightly lean mixtures fuel economy is greater and the combustion reaction is more complete. Additionally, the final combustion temperature can be lowered by using ultra-lean mixtures, reducing the amount of NOx emissions. Indeed, the minimum energy required for ignition for hydrogen is about an order of magnitude less than that required for gasoline. This enables hydrogen engines to run well on lean mixtures and ensures prompt ignition. Unfortunately, since very little energy is necessary to ignite a hydrogen combustion reaction, and almost any hydrogen/air mixture can be ignited due to wide limits of flammability of hydrogen, hot gases and hot spots on the cylinder can serve as sources of ignition, creating problems of premature ignition and flashback [7].
KHALADI Fatma et al.Procedia / Energy00 Procedia 139 (2017) 530–536 F-Z. Khaladi et al.Zohra / Energy (2017) 000–000
532
3
1.2. Hydrogen Combustion In the combustion chamber, the combustion flame is typically extinguished at certain distance from the cylinder wall due to heat losses called as quenching distance. For hydrogen, the quenching distance is less than that of gasoline, so that flame comes closer to the wall before it is extinguished. Thus, it is more difficult to quench a hydrogen flame than a gasoline flame [8]. The flame speed of hydrogen is nearly an order of magnitude higher than that of gasoline. For stoichiometric mixtures, hydrogen engines can more closely approach the thermodynamically ideal engine cycle. At leaner mixtures, the flame velocity decreases significantly. In other hand, one the reasons for which we are interested in hydrogen is because its chemistry is considered a starting point for the more complex hydrocarbon chemistry. It is important to stress that in the auto ignition stages of any flame, the fuel air mixture may follow a low temperature reaction mechanism and in the latter stages, an explosive reaction due to the increase in temperature and/or pressure causing the operating point to shift between the regions of the explosion limits of a stoichiometric mixture. 2. GEOMETRY AND BOUNDARY CONDITIONS In our study, we adopted the geometry of Zhuyin Ren [3] with the same conditions of entry of fuel and air: methane-hydrogen is surrounded by a bluff-body and an air co-flow. The diameter of the bluff-body, Db, is 50 mm and the jet is 3.6 mm. The computational domain is shown in Fig 1. Origin coordinate system is taken at the center of the output plane of the jet. The numerical conditions selected for calculating a non- premixed turbulent flame stabilized by bluff-body are : (fuel axial velocity is 118 m/s with a 10% turbulence intensity and Hydraulic diameter is 3.6 mm to 300K temperature, air cocurrently velocity is 40 m/s with 10% turbulence intensity and Hydraulic diameter equal to 0.25 m at a temperature of 300 K). The domain is discretized with a mesh multizone: hexahedral prisms with 10954 to 10051 nodes. Axis-symmetric boundary conditions are applied along the central axis of the combustion chamber. At the exit, a pressure outlet boundary condition is specified with a fixed pressure of 1.01325*105 Pa. At the chamber wall, no-slip boundary condition and no species flux normal to the wall surface are applied. The thermal boundary condition on the chamber wall is taken as adiabatic wall condition. Fuel inlet
(0,0)
Bluff Body
Db 3Db
Air Inlet cocourant
7.9 Db
Fig 1. Schematic of the combustion chamber (Db=50mm and Øfuel inj=3.6 mm)
3. MATHEMATICAL MODELING The governing equations include the set of the Navier-Stokes equations, continuity equation, and any additional conservation equations, such as energy or species concentrations. The fluid flow is modeled by the governing equations, which show the effect of the governing phenomena on the fluid flow. These governing phenomena may include conduction, convection, diffusion, turbulence, radiation and combustion [9]. For turbulence modelling, we used the Standard k-ε model. In this simple model, two additional transport equation are solved for the two turbulence quantities:. the turbulent kinetic energy k and the energy dissipation rate ε. These two quantities are related to the primary variables and can give a length scale and time scale to form a quantity with dimension of , thus making the model complete (no more flow-dependent specifications are required). This is a widely used model in CFD simulations.
4
KHALADI Fatma Zohra et al. / Energy Procedia 139 (2017) 530–536 F-Z. Khaladi et al. / Energy Procedia 00 (2017) 000–000
533
4. Results and discussion 4.1. Mesh sensitivity A number of numerical simulations have been performed to study the combustion phenomena under adiabatic wall conditions when hydrogen air mixture changes from lean to rich and at different mass flow rate of mixture. The first step was a choice of the mesh to discretize the volume occupied by the fluid, a mesh that ensures convergence of the numerical procedure. To achieve this were tested several types of mesh. For testing of the mesh, methane-air combustion was simulated without adding hydrogen with EDM (Eddy dissipation model). The figures below show the evolution of the temperature for each type of mesh. Fig 2 shows that for an axial section of the injector very close (a) the change in temperature is almost the same regardless of the mesh type. But away from the injector, this trend changes from one mesh to another (a) and (b) except for the multizone mesh with three types of mapped mesh (b). 1800
1000
automatic tetraedrons hexaedral to quadrilators hexaedral to quadrilators-triangles multizone hexaedral multizone to prisms multizone hexaedres/prisms
1600 1400 1200 1000 800 600
500
0,01
0,02
radius (m) Y/Db=0,26
0,03
0,04
Y/Db=0,26 Y/Db=0,6
1500
Tem pe rature(k)
1500
0,00
2000
automatic tetraehdrons hexaedral to quadrilators hexaedral to quadrilators-triangles multizone hexaedral multizone to prisms multizone to hexaedrons-prisms
T e m p é ra tu re (k )
Temperature ( k)
2000
Y/Db=1,3 Y/Db=1,8 Y/Db=3,5 Y/Db=5
1000
Y/Db=7
500
400
0,00 0,00
(a)
Fig 2. Temperature Profiles for different mesh
0,01
rayon (m) Y/Db=1,3
0,02
0,03
0,04
0,01
0,02
0,03
0,04
radius (m) mesh multizone(hexaedrons-prism)
Fig 3. Temperature Profiles for different positions in the axial centre line
Fig 3 shows clearly that the clearly change in the temperature curve is better represented by a multizone mesh than by automatic mesh. Indeed, in the first case we see that the temperature logically evolves in an increasing manner from one section to another to exit the chamber. While this trend is not stable in the second case: decreasing between 0.26, 0.6 and 1.3 and 1.8 de1.3 increasing then decreasing from 1.8 to exit the combustion chamber. Fig 4a shows an asymmetric shape of the iso-contours in a hexahedral-dominated meshing, and this is the case in all types tested mesh, except for the multizone mesh where the shape of the iso-contour is axisymmetric as shown in Fig 4b. From the above we see that the evolution of the temperature is homogeneous and uniform axisymmetric for multizone mesh (prism, hexahedral) or non-uniform (hexahedral prisms) than for other types of mesh. In addition, it does not require a lot of computing time compared to other types of mesh (500 iterations against 1000). That is why we opted for the multizone mesh. Only between uniform and non-uniform mesh, we choose the latter.
(a) (b) Fig 4. Temperature iso-contours for different mesh: (a) Method for dominance hexahedral: mesh free only quadrilateral faces, (b) Method Multizone with a mapped hexahedral mesh/prism
KHALADI Fatma Zohra et al. / Energy Procedia 139 (2017) 530–536 F-Z. Khaladi et al. / Energy Procedia 00 (2017) 000–000
534
5
4.2. Hydrogen-air flame Fig 5 shows the temperature iso-contours in a hydrogen-air flame with the same basis as those of air methane: an asymmetrical shape of a flame that seems blown but where the maximum temperature is substantially greater than that of methane-air (2273 k / 1887k). The Figure shows the contours of temperature (K) on the cross section along central axis of combustion chamber at stoichiometric air fuel ratio (at Ф=1). It shows the gas temperature distribution along the central axis. It can be seen from this Figure that the highest temperature is obtained at the exit of combustion chamber. The flame temperature can be as high as 2365K which is almost the same as the adiabatic flame temperature of the Combustion of non-premixed stoichiometric hydrogen-air mixture.
H2-Air CH4- Air Fig 5. Temperature iso-contours for Hydrogen-Air and Methane-Air Flames
4.3. Enrichment with hydrogen Fig 6 below clearly illustrates the difference in the radial variation in temperature between three types of flames: airmethane, hydrogen-air and methane-hydrogen-air. Although the initial temperature and that at the end of combustion (flame resistance to extinction), the Figure shows that hydrogen flame, is higher than in a methane flame, the widest radial development of this temperature is observed only away from the injector. As to the radial profile of the temperature of a methane-air flame, enriched with 10% hydrogen, it coincides with that of a methaneair flame without addition of hydrogen but with a slight increase of the temperature in the first flame. Indicating the effect of enrichment with hydrogen, ie the percentage remains suitable for the adopted configuration. Fig 7a shows the evolution of the temperature in a methane-air flame enriched with hydrogen at different enrichment levels. We notice a significant difference in the evolution of temperature increases with increasing hydrogen content when combustion is simulated model combined with EDM/FRC. A slight increase when combustion is simulated with the EDM model. However, the maximum temperature may decrease when the proportion of hydrogen exceeds 50% (Fig 8a). The temperature iso-contours (Fig 7b) for each percent of added hydrogen confirm this imbalance: the flame loses its axially symmetrical shape and homogeneous and appears blown. The best percentage of hydrogen addition would be a priori 50% a result confirmed by [3]. 100% CH4
100 % CH4
100% H2
2000
10% H2
2000
100 % H2 10% H2
1500
Température (k)
Température(k)
1500
1000
500
500 0,00
1000
0,01
rayon(m) Y/Db=1,8 (d)
0,02
0,03
0,04
0,00
0,01
rayon(m) Y/Db=0,26 (a)
0,02
0,03
Fig 6- Radial Profile of for different sections and three flames types
0,04
F-Z. Khaladi Fatma et al. / Zohra EnergyetProcedia 00 (2017) 000–000 KHALADI al. / Energy Procedia 139 (2017) 530–536
Température(k)
6
1900 1800 1700 1600 1500 1400 1300 1200 1100 1000 900 800 700 600 500 400 300 0,00
535
10 % H2 20 % H2 30 % H2 40 % H2 50 % H2 60 % H2 70 % H2 80 % H2 90 % H2
0,01
rayon (m) Y/Db=0,26
0,02
0,03
0,04
10%H2 30% H2 60%H2 90% H2 (a) (b) Fig 7. Radial temperature profile (a) and Iso-contours of temperature for different hydrogen percentages Added (b) : Combustion simulated by EDM/FRC Combined (a)
10% H2
2000
20% H2 30% H2 40% H2 50% H2
Température(k)
1500
1000
500
0,00
0,01
rayon (m) Y/Db=0,26
0,02
0,03
0,04
10%H2
(a)
30% H2
90% H2
(a) (b) Fig 8. Radial temperature profile (a) and Iso-temperature contours for different percentages of added hydrogen (b) simulated combustion by EDM
4.4. Comparison with experience The question is what model is best for simulating this flame. For this, a comparison is made with experimental data from [3]. Fig 9 shows a good agreement with simulation and experiment results. The combined model EDM/FRC give better result than simple EDM model in the outlet zone of the combustion chamber. Near the burner region, the EDM model can be used with an acceptable precision. 2000
2000
Température(k)
1500
1000
500 0,00
1500
Température(k)
EDM EDM/FRC Expérience
EDM EDM/FRC Expérience
1000
500
0,01
rayon (m) Y/Db=0,26 (a)
0,02
0,03
0,04
0,00
0,01 rayon
(m) 0,02 Y/Db=1,8 (d)
0,03
0,04
Fig 9: Radial profile of the temperature of a methane-air flame enriched with 50% hydrogen
5. CONCLUSION In this work, the CFD based combustion simulations have been applied to analyze the combustion characteristics of non-premixed hydrogen-air in a 2D combustor. The CFD simulations, taking into account the coupling of fluid dynamics, heat transfer and detailed chemical kinetics, are used to investigate the effects of various operating conditions. The combustor performance is evaluated by predicting the temperatures of exit gas of the combustor and outer wall of the combustor. To make the combustor operable, the heat output should meet the design criteria, the wall temperature should be lower than the material allowable temperature and the exit gas
536
KHALADI Fatma Zohra et al. / Energy Procedia 139 (2017) 530–536 F-Z. Khaladi et al. / Energy Procedia 00 (2017) 000–000
7
temperature should be high enough. In other hand, this study presents a numerical simulation of a non-premixed turbulent flame of methane-air enriched by hydrogen. The selected axisymmetric configuration is composed of a central injector of methane-hydrogen mixture surrounded by a bluff-body, which is surrounded by a co-axial air jet. The EDM (Eddy Dissipation model), then the FRC model (Finite Rate Combustion) combined with EDM are used to modeling the combustion phenomena. The results show some concordance with the temperature profile given by experience to a hydrogen rate of 50%. References [1]- Das L.M., Hydrogen-oxygen reaction mechanism and its implication to hydrogen engine combustion, 21:703-715 International Journal of Hydrogen Energy, 1996 [2]- S. Verhelst, R. Sierens, Hydrogen engine-specific properties, International Journal of Hydrogen Energy 26 (2001) 987990. [3]- Zhuyin Ren , Graham M. Goldin , Varun Hiremath et Stephen B. Pope, "Simulations of a turbulent non-premixed flame using combined dimension reduction and tabulation for combustion chemistry ", Elsevier, 1er septembre 2012. [4]- Jingsong Hua, MengWu, Kurichi Kumar, Numerical simulation of the combustion of hydrogen air mixture in micro-scaled chambers. Part II: CFD analysis for a micro-combustor 60:3507-3515, chemical engineering science, 2005. [5]- Belahhomley Sauvi, Introduction aux combustibles et à la combustion, Édition Technip, Paris, 1981. [6]- Sources Gaz De France, Carburants et Moteurs, Guibet, Combustibles gazeux (oct 93) [7]- R. Mouangue, M. Obounou, " Simulation numerique d’une flamme de diffusion turbulente h2/air ", Phys. Chem. News 50 (2009) 69-78 [8]- John GRONDIN ," Etude du mélange et de la combustion dans les flammes-jets subsoniques à haute vitesse : influence des vitesses, des densités et de la composition du combustible", Thèse de doctorat , ENSMA Poitiers, 7 janvier 2010. [9]- BORGHI. R., 1988, Turbulent combustion modelling, Prog. Energy Comb. Sci., 14 245 [10]- Mounir Alliche, Pierre Haldenwang, Salah Chikh, Extinction conditions of a premixed flame in a channel, Combustion and Flame, 157 (2010), 1060-1070.
ScienceDirect Availableonline onlineatatwww.sciencedirect.com www.sciencedirect.com Available Energy Procedia 00 (2017) 000–000
ScienceDirect ScienceDirect
www.elsevier.com/locate/procedia
Energy (2017) 000–000 530–536 EnergyProcedia Procedia139 00 (2017) www.elsevier.com/locate/procedia
International Conference On Materials And Energy 2015, ICOME 15, 19-22 May 2015, Tetouan, Morocco, and the International Conference On Materials And Energy 2016, ICOME 16, 17-20 May 2016, La Rochelle, France The 15th International Symposium on District Heating and Cooling
Numerical simulation of CH4-H2-AIR non-premixed flame Assessing the feasibility thebody heat demand-outdoor stabilizedofbyusing a bluff temperature function for a 1long-term district 2 heat demand 1 forecast 12-
KHALADI Fatma Zohra , ALLICHE Mounir *, CHIKH Salah a,b,c a a b c c I. Andrić *, A. Pina , P. Ferrão , J. Fournier ., B. Lacarrière O. Le Corre Laboratoire de Transports Polyphasiques et Milieux Poreux (LTPMP), Université de ,Technologie Houari
a
Boumediene, Alger, Algeria
IN+ Center for Innovation, Physique Technology and Policy Research - Instituto SuperiorUniversité Técnico, Av. RoviscoQuartier Pais 1, 1049-001 Laboratoire de Mécanique et Modélisation Mathématique (LMP2M), de Médéa, Ain Dheb,Lisbon, Medea,Portugal Algeria b Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France c Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France
Abstract Abstract The aim of this study is to perform a numerical simulation of a non-premixed turbulent flame of methane-air enriched by hydrogen. The selected axisymmetric configuration is composed of a central injector of methane-hydrogen mixture surrounded heatingwhich networks are commonly theThe literature onesoftware of the most effective solutions for decreasing the byDistrict a bluff-body, is surrounded by a addressed co-axial airinjet. Ansys as CFX is used to solve the equations governing greenhouse gas emissions fromStokes the building sector. These require high investments which returned through heat turbulent reactive flow (Navier averaged in sense of systems Favre). The Turbulence is modeled usingarethe k-ε model. Thethe EDM sales. Dissipation Due to themodel), changedthen climate conditions building policies, heatwith demand the used future decrease, (Eddy the FRC model and (Finite Rate renovation Combustion) combined EDMin are to could modeling the prolonging the investment return period. combustion phenomena. The results show some concordance with the temperature profile given by experience to a hydrogen rate main scope of this paper is to assess the feasibility of using the heat demand – outdoor temperature function for heat demand ofThe 50%. forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 vary Published in both construction ©buildings 2017 Thethat Authors. by Elsevierperiod Ltd. and typology. Three weather scenarios (low, medium, high) and three district renovation scenarios were developed (shallow, intermediate, deep). 2015 To estimate the error, Peer-review under responsibility of the scientific committee of ICOME and ICOME 2016.obtained heat demand values were compared with results from a dynamic heat demand model, previously developed and validated by the authors. The results showed that when only weather change isEDM considered, margin errorCFX could be acceptable for some applications Keywords: Non-premixed combustion, Hydrogen enrichment, and FRC the models, CFD,ofAnsys (the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered). value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the 1.The Introduction decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and renovation scenarios considered). On the other hand, function intercept increased for 7.8-12.7% per decade (depending on the Over the past three decades, there has been considerable effort in the world to develop and introduce coupled scenarios). The values suggested could be used to modify the function parameters for the scenarios considered, and alternative transportation fuels to replace conventional fuels such as gasoline and diesel, environmental issues, most improve the accuracy of heat demand estimations.
notably air pollution and limited availability of conventional fuels are among the principle driving forces behind this © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling. * Corresponding author. Keywords: Heat demand; Forecast; Climate change E-mail address: [email protected] 1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of ICOME 2015 and ICOME 2016. 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the scientific committee of ICOME 2015 and ICOME 2016 10.1016/j.egypro.2017.11.249
2
KHALADI Fatma Zohra et al. / Energy Procedia 139 (2017) 530–536 F-Z. Khaladi et al. / Energy Procedia 00 (2017) 000–000
531
movement. Thus, if one tries to find for the definition of perfect fuel, hydrogen probably satisfies most of the desirable characteristics of such a fuel. Plentiful and clean burning, hydrogen has very high-energy content. [1, 2] Due to difficulties in conducting spatially resolved measurements of combustion characteristics in devices, the numerical simulation can be cost effective approach to study the combustion mechanism. In this work, Computational Fluid Dynamics (CFD) based numerical simulations have been performed to study the combustion of non-premixed and various aspects in order to use them for solving the realistic problems. This study allows us: • The understanding of the basics of Hydrogen-oxygen reaction mechanism, its combustion and the geometry of the cylindrical chamber used in this study is very important for simulating Methane-Air combustion system enriched by Hydrogen. • To develop a two-dimensional numerical mesh and flow model which adequately and accurately represent the physical model of combustion chamber and is simple enough to limit the amount of computational time for obtaining a solution. The objective of this study is to find and apply appropriate model that improve the simulation of combustion with the commercial CFD as Ansys CFX. • Generate numerical data/solutions, which correlate as much as possible with the experimental data for various conditions including equivalence ratios, mass flow rates of hydrogen-air mixture. In the same context, Zhuyin Ren [3] presents a numerical simulation of a non-premixed combustion flame methane-air enriched by hydrogen, stabilized by a recirculation zone created by an obstacle or a bluff-body : a simple approach in which we apply a different combustion model to see if the same results were been obtained. Zhuyin Ren [3] uses a reduced description with chemical tabs implemented in FLUENT is combined with the EDC model (Eddy Dissipation Concept) considers that a moderate or chemistry with model PDF (probability density function) for combustion. We decided, in the simulation of combustion, for EDM in its infinitely fast chemistry limited compared to the scale of the turbulent times. Then for model Combined EDM/FRC, firstly, that it is valid for several reactions classified from a low to a high number of Damköhler (slow or fast chemistry compared to the scale of the turbulence time) and secondly, reaction rates are firstly calculated for each model separately and then the minimum of both is used. This procedure is applied to each reaction separately, so that when the level in onestep, would be limited by the chemical kinetics, certain other steps would be limited by the turbulent mixing at the same time and in the same physical position. [4]. Our approach is a first step in choosing the appropriate mesh following a test of several types of mesh, with an infinitely fast chemistry, applying the EDM model for the combustion of methane-air flame. Then, we compare the methane-air flame at the hydrogen-air flame. In the other hand, we compare the application of EDM model to combined EDM/FRC model for simulating the combustion of an air- methane sulfonate flame enriched by hydrogen. The results of our calculations are presented in the form of temperature profiles. 1.1. Hydrogen as a fuel Hydrogen is a colorless, odorless, tasteless, and nonpoisonous gas under normal conditions on Earth. It typically exists as a diatomic molecule. Hydrogen is the most abundant element in the universe, accounting for 90 percent of the universe by weight. However, it is not commonly found in its pure form. [1, 2, 5] The properties of hydrogen are largely listed in reference [6] with conventional fuels i.e. Gasoline & Diesel and other alternative fuels such as CNG, LPG, and Biogas. Hydrogen has wide range of flammability in comparison with other fuels. One of the significant advantages is that hydrogen engine can run on a lean mixture. When engine is run on slightly lean mixtures fuel economy is greater and the combustion reaction is more complete. Additionally, the final combustion temperature can be lowered by using ultra-lean mixtures, reducing the amount of NOx emissions. Indeed, the minimum energy required for ignition for hydrogen is about an order of magnitude less than that required for gasoline. This enables hydrogen engines to run well on lean mixtures and ensures prompt ignition. Unfortunately, since very little energy is necessary to ignite a hydrogen combustion reaction, and almost any hydrogen/air mixture can be ignited due to wide limits of flammability of hydrogen, hot gases and hot spots on the cylinder can serve as sources of ignition, creating problems of premature ignition and flashback [7].
KHALADI Fatma et al.Procedia / Energy00 Procedia 139 (2017) 530–536 F-Z. Khaladi et al.Zohra / Energy (2017) 000–000
532
3
1.2. Hydrogen Combustion In the combustion chamber, the combustion flame is typically extinguished at certain distance from the cylinder wall due to heat losses called as quenching distance. For hydrogen, the quenching distance is less than that of gasoline, so that flame comes closer to the wall before it is extinguished. Thus, it is more difficult to quench a hydrogen flame than a gasoline flame [8]. The flame speed of hydrogen is nearly an order of magnitude higher than that of gasoline. For stoichiometric mixtures, hydrogen engines can more closely approach the thermodynamically ideal engine cycle. At leaner mixtures, the flame velocity decreases significantly. In other hand, one the reasons for which we are interested in hydrogen is because its chemistry is considered a starting point for the more complex hydrocarbon chemistry. It is important to stress that in the auto ignition stages of any flame, the fuel air mixture may follow a low temperature reaction mechanism and in the latter stages, an explosive reaction due to the increase in temperature and/or pressure causing the operating point to shift between the regions of the explosion limits of a stoichiometric mixture. 2. GEOMETRY AND BOUNDARY CONDITIONS In our study, we adopted the geometry of Zhuyin Ren [3] with the same conditions of entry of fuel and air: methane-hydrogen is surrounded by a bluff-body and an air co-flow. The diameter of the bluff-body, Db, is 50 mm and the jet is 3.6 mm. The computational domain is shown in Fig 1. Origin coordinate system is taken at the center of the output plane of the jet. The numerical conditions selected for calculating a non- premixed turbulent flame stabilized by bluff-body are : (fuel axial velocity is 118 m/s with a 10% turbulence intensity and Hydraulic diameter is 3.6 mm to 300K temperature, air cocurrently velocity is 40 m/s with 10% turbulence intensity and Hydraulic diameter equal to 0.25 m at a temperature of 300 K). The domain is discretized with a mesh multizone: hexahedral prisms with 10954 to 10051 nodes. Axis-symmetric boundary conditions are applied along the central axis of the combustion chamber. At the exit, a pressure outlet boundary condition is specified with a fixed pressure of 1.01325*105 Pa. At the chamber wall, no-slip boundary condition and no species flux normal to the wall surface are applied. The thermal boundary condition on the chamber wall is taken as adiabatic wall condition. Fuel inlet
(0,0)
Bluff Body
Db 3Db
Air Inlet cocourant
7.9 Db
Fig 1. Schematic of the combustion chamber (Db=50mm and Øfuel inj=3.6 mm)
3. MATHEMATICAL MODELING The governing equations include the set of the Navier-Stokes equations, continuity equation, and any additional conservation equations, such as energy or species concentrations. The fluid flow is modeled by the governing equations, which show the effect of the governing phenomena on the fluid flow. These governing phenomena may include conduction, convection, diffusion, turbulence, radiation and combustion [9]. For turbulence modelling, we used the Standard k-ε model. In this simple model, two additional transport equation are solved for the two turbulence quantities:. the turbulent kinetic energy k and the energy dissipation rate ε. These two quantities are related to the primary variables and can give a length scale and time scale to form a quantity with dimension of , thus making the model complete (no more flow-dependent specifications are required). This is a widely used model in CFD simulations.
4
KHALADI Fatma Zohra et al. / Energy Procedia 139 (2017) 530–536 F-Z. Khaladi et al. / Energy Procedia 00 (2017) 000–000
533
4. Results and discussion 4.1. Mesh sensitivity A number of numerical simulations have been performed to study the combustion phenomena under adiabatic wall conditions when hydrogen air mixture changes from lean to rich and at different mass flow rate of mixture. The first step was a choice of the mesh to discretize the volume occupied by the fluid, a mesh that ensures convergence of the numerical procedure. To achieve this were tested several types of mesh. For testing of the mesh, methane-air combustion was simulated without adding hydrogen with EDM (Eddy dissipation model). The figures below show the evolution of the temperature for each type of mesh. Fig 2 shows that for an axial section of the injector very close (a) the change in temperature is almost the same regardless of the mesh type. But away from the injector, this trend changes from one mesh to another (a) and (b) except for the multizone mesh with three types of mapped mesh (b). 1800
1000
automatic tetraedrons hexaedral to quadrilators hexaedral to quadrilators-triangles multizone hexaedral multizone to prisms multizone hexaedres/prisms
1600 1400 1200 1000 800 600
500
0,01
0,02
radius (m) Y/Db=0,26
0,03
0,04
Y/Db=0,26 Y/Db=0,6
1500
Tem pe rature(k)
1500
0,00
2000
automatic tetraehdrons hexaedral to quadrilators hexaedral to quadrilators-triangles multizone hexaedral multizone to prisms multizone to hexaedrons-prisms
T e m p é ra tu re (k )
Temperature ( k)
2000
Y/Db=1,3 Y/Db=1,8 Y/Db=3,5 Y/Db=5
1000
Y/Db=7
500
400
0,00 0,00
(a)
Fig 2. Temperature Profiles for different mesh
0,01
rayon (m) Y/Db=1,3
0,02
0,03
0,04
0,01
0,02
0,03
0,04
radius (m) mesh multizone(hexaedrons-prism)
Fig 3. Temperature Profiles for different positions in the axial centre line
Fig 3 shows clearly that the clearly change in the temperature curve is better represented by a multizone mesh than by automatic mesh. Indeed, in the first case we see that the temperature logically evolves in an increasing manner from one section to another to exit the chamber. While this trend is not stable in the second case: decreasing between 0.26, 0.6 and 1.3 and 1.8 de1.3 increasing then decreasing from 1.8 to exit the combustion chamber. Fig 4a shows an asymmetric shape of the iso-contours in a hexahedral-dominated meshing, and this is the case in all types tested mesh, except for the multizone mesh where the shape of the iso-contour is axisymmetric as shown in Fig 4b. From the above we see that the evolution of the temperature is homogeneous and uniform axisymmetric for multizone mesh (prism, hexahedral) or non-uniform (hexahedral prisms) than for other types of mesh. In addition, it does not require a lot of computing time compared to other types of mesh (500 iterations against 1000). That is why we opted for the multizone mesh. Only between uniform and non-uniform mesh, we choose the latter.
(a) (b) Fig 4. Temperature iso-contours for different mesh: (a) Method for dominance hexahedral: mesh free only quadrilateral faces, (b) Method Multizone with a mapped hexahedral mesh/prism
KHALADI Fatma Zohra et al. / Energy Procedia 139 (2017) 530–536 F-Z. Khaladi et al. / Energy Procedia 00 (2017) 000–000
534
5
4.2. Hydrogen-air flame Fig 5 shows the temperature iso-contours in a hydrogen-air flame with the same basis as those of air methane: an asymmetrical shape of a flame that seems blown but where the maximum temperature is substantially greater than that of methane-air (2273 k / 1887k). The Figure shows the contours of temperature (K) on the cross section along central axis of combustion chamber at stoichiometric air fuel ratio (at Ф=1). It shows the gas temperature distribution along the central axis. It can be seen from this Figure that the highest temperature is obtained at the exit of combustion chamber. The flame temperature can be as high as 2365K which is almost the same as the adiabatic flame temperature of the Combustion of non-premixed stoichiometric hydrogen-air mixture.
H2-Air CH4- Air Fig 5. Temperature iso-contours for Hydrogen-Air and Methane-Air Flames
4.3. Enrichment with hydrogen Fig 6 below clearly illustrates the difference in the radial variation in temperature between three types of flames: airmethane, hydrogen-air and methane-hydrogen-air. Although the initial temperature and that at the end of combustion (flame resistance to extinction), the Figure shows that hydrogen flame, is higher than in a methane flame, the widest radial development of this temperature is observed only away from the injector. As to the radial profile of the temperature of a methane-air flame, enriched with 10% hydrogen, it coincides with that of a methaneair flame without addition of hydrogen but with a slight increase of the temperature in the first flame. Indicating the effect of enrichment with hydrogen, ie the percentage remains suitable for the adopted configuration. Fig 7a shows the evolution of the temperature in a methane-air flame enriched with hydrogen at different enrichment levels. We notice a significant difference in the evolution of temperature increases with increasing hydrogen content when combustion is simulated model combined with EDM/FRC. A slight increase when combustion is simulated with the EDM model. However, the maximum temperature may decrease when the proportion of hydrogen exceeds 50% (Fig 8a). The temperature iso-contours (Fig 7b) for each percent of added hydrogen confirm this imbalance: the flame loses its axially symmetrical shape and homogeneous and appears blown. The best percentage of hydrogen addition would be a priori 50% a result confirmed by [3]. 100% CH4
100 % CH4
100% H2
2000
10% H2
2000
100 % H2 10% H2
1500
Température (k)
Température(k)
1500
1000
500
500 0,00
1000
0,01
rayon(m) Y/Db=1,8 (d)
0,02
0,03
0,04
0,00
0,01
rayon(m) Y/Db=0,26 (a)
0,02
0,03
Fig 6- Radial Profile of for different sections and three flames types
0,04
F-Z. Khaladi Fatma et al. / Zohra EnergyetProcedia 00 (2017) 000–000 KHALADI al. / Energy Procedia 139 (2017) 530–536
Température(k)
6
1900 1800 1700 1600 1500 1400 1300 1200 1100 1000 900 800 700 600 500 400 300 0,00
535
10 % H2 20 % H2 30 % H2 40 % H2 50 % H2 60 % H2 70 % H2 80 % H2 90 % H2
0,01
rayon (m) Y/Db=0,26
0,02
0,03
0,04
10%H2 30% H2 60%H2 90% H2 (a) (b) Fig 7. Radial temperature profile (a) and Iso-contours of temperature for different hydrogen percentages Added (b) : Combustion simulated by EDM/FRC Combined (a)
10% H2
2000
20% H2 30% H2 40% H2 50% H2
Température(k)
1500
1000
500
0,00
0,01
rayon (m) Y/Db=0,26
0,02
0,03
0,04
10%H2
(a)
30% H2
90% H2
(a) (b) Fig 8. Radial temperature profile (a) and Iso-temperature contours for different percentages of added hydrogen (b) simulated combustion by EDM
4.4. Comparison with experience The question is what model is best for simulating this flame. For this, a comparison is made with experimental data from [3]. Fig 9 shows a good agreement with simulation and experiment results. The combined model EDM/FRC give better result than simple EDM model in the outlet zone of the combustion chamber. Near the burner region, the EDM model can be used with an acceptable precision. 2000
2000
Température(k)
1500
1000
500 0,00
1500
Température(k)
EDM EDM/FRC Expérience
EDM EDM/FRC Expérience
1000
500
0,01
rayon (m) Y/Db=0,26 (a)
0,02
0,03
0,04
0,00
0,01 rayon
(m) 0,02 Y/Db=1,8 (d)
0,03
0,04
Fig 9: Radial profile of the temperature of a methane-air flame enriched with 50% hydrogen
5. CONCLUSION In this work, the CFD based combustion simulations have been applied to analyze the combustion characteristics of non-premixed hydrogen-air in a 2D combustor. The CFD simulations, taking into account the coupling of fluid dynamics, heat transfer and detailed chemical kinetics, are used to investigate the effects of various operating conditions. The combustor performance is evaluated by predicting the temperatures of exit gas of the combustor and outer wall of the combustor. To make the combustor operable, the heat output should meet the design criteria, the wall temperature should be lower than the material allowable temperature and the exit gas
536
KHALADI Fatma Zohra et al. / Energy Procedia 139 (2017) 530–536 F-Z. Khaladi et al. / Energy Procedia 00 (2017) 000–000
7
temperature should be high enough. In other hand, this study presents a numerical simulation of a non-premixed turbulent flame of methane-air enriched by hydrogen. The selected axisymmetric configuration is composed of a central injector of methane-hydrogen mixture surrounded by a bluff-body, which is surrounded by a co-axial air jet. The EDM (Eddy Dissipation model), then the FRC model (Finite Rate Combustion) combined with EDM are used to modeling the combustion phenomena. The results show some concordance with the temperature profile given by experience to a hydrogen rate of 50%. References [1]- Das L.M., Hydrogen-oxygen reaction mechanism and its implication to hydrogen engine combustion, 21:703-715 International Journal of Hydrogen Energy, 1996 [2]- S. Verhelst, R. Sierens, Hydrogen engine-specific properties, International Journal of Hydrogen Energy 26 (2001) 987990. [3]- Zhuyin Ren , Graham M. Goldin , Varun Hiremath et Stephen B. Pope, "Simulations of a turbulent non-premixed flame using combined dimension reduction and tabulation for combustion chemistry ", Elsevier, 1er septembre 2012. [4]- Jingsong Hua, MengWu, Kurichi Kumar, Numerical simulation of the combustion of hydrogen air mixture in micro-scaled chambers. Part II: CFD analysis for a micro-combustor 60:3507-3515, chemical engineering science, 2005. [5]- Belahhomley Sauvi, Introduction aux combustibles et à la combustion, Édition Technip, Paris, 1981. [6]- Sources Gaz De France, Carburants et Moteurs, Guibet, Combustibles gazeux (oct 93) [7]- R. Mouangue, M. Obounou, " Simulation numerique d’une flamme de diffusion turbulente h2/air ", Phys. Chem. News 50 (2009) 69-78 [8]- John GRONDIN ," Etude du mélange et de la combustion dans les flammes-jets subsoniques à haute vitesse : influence des vitesses, des densités et de la composition du combustible", Thèse de doctorat , ENSMA Poitiers, 7 janvier 2010. [9]- BORGHI. R., 1988, Turbulent combustion modelling, Prog. Energy Comb. Sci., 14 245 [10]- Mounir Alliche, Pierre Haldenwang, Salah Chikh, Extinction conditions of a premixed flame in a channel, Combustion and Flame, 157 (2010), 1060-1070.