Applied Thermal Engineering 58 (2013) 354e362

Contents lists available at SciVerse ScienceDirect

Applied Thermal Engineering

journal homepage: www.elsevier.com/locate/apthermeng

Simulation of transport phenomena in coke oven with staging combustion

Ke Jin a, Yanhui Feng a, *, Xinxin Zhang a, Mingdeng Wang b, Junfeng Yang b, Xiaobo Ma b
a b

School of Mechanical Engineering, University of Science and Technology Beijing, Beijing 100083, China ACRE Coking and Refractory Engineering Consulting Corporation, Liaoning 114001, China

h i g h l i g h t s
 The application of staging combustion in coke ovens and its effects are analyzed.  A 3D model is proposed to describe owecombustionethermo behaviors in coke oven.  Optimizing operation parameters in full-scale coke oven are studied.

a r t i c l e i n f o
Article history: Received 25 January 2013 Accepted 27 April 2013 Available online 7 May 2013 Keywords: Coke oven Coking chamber Combustion chamber Staging combustion Transport phenomena

a b s t r a c t
A three-dimensional transient mathematical model was developed for coupled coking chamber and staging combustion chamber in large-capacity coke ovens, to describe the owecombustionethermo behavior. The model was solved numerically using CFX CFD package and was validated by the central temperature evolution of coke bed. The elds of temperature, uid ow and combustible gas concentration were analyzed, with special reference to the temperature difference of coke bed and NO concentration of exhaust. The results show that staging combustion plays an important role in improving temperature uniformity of the coke bed and reducing NO concentration of exhaust, especially for the large-capacity coke oven. It is benecial for production optimization to decrease the gas mass ow rate at the bottom inlet while increase the rate at the upper inlet in the combustion chamber. In addition, it turns out that some measures such as coal preheating, adjustment of moisture content or/and coal densication may be used to improve the coke production efciency. It is expected the developed model and relevant data in the present research will be benecial to realize large-scale coke oven with a higher energy efciency and lower emission. 2013 Elsevier Ltd. All rights reserved.

1. Introduction Coke produced from blends of coking coals has a wide range of applications and 90% of the coke is used for iron production in the blast furnace [1,2]. Regenerative coke oven is widely used in the world and it contributes more than 90% of the total coke production [3,4]. The combustion chamber and coking chamber are essential components of a regenerative coke oven. The combustion chamber provides thermal energy to the coking chamber where the coal isolated from the air is carbonized to be carbon coke. To guarantee the coke quality, the nal temperature of coke bed needs to be in the range of 950  Ce1050  C, and the temperature uniformity of the bed is satised with the largest temperature difference less than 50  C [5].
* Corresponding author. Tel.: 86 10 62333874; fax: 86 10 62332741. E-mail address: yhfeng@me.ustb.edu.cn (Y. Feng). 1359-4311/$ e see front matter 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.applthermaleng.2013.04.056

Much attention has been paid to the transport phenomena in the coking chamber of coke oven and the related simulation. Tian [6] proposed a simple one-dimensional transient heat conduction model with constant physical properties. Rodhe [7] improved the model by taking into account the temperature-dependent properties, moisture evaporation and chemical reaction. D. Merrick [8,9] developed a series of mathematical models for coking process, to predict the release of volatile matter, the physical properties of the coal/semi-coke/coke during coking, the charge temperature history and effect of blend composition on coke strength and so on. Guo and Tang [10] simultaneously calculated the transient composition, temperatures of the gas and the solid phases, velocity of the gas phase and porosity and density of the semi-coke phase for a coking process using PHOENICS CFD package. To evaluate the effects of the adjacent combustion chamber and improve the accuracy of simulation, coupling adjacent combustion and coking chambers were further pursued by researchers. Chen

K. Jin et al. / Applied Thermal Engineering 58 (2013) 354e362

355

Nomenclature A anisotropic phase function coefcient C specic heat capacity, J kg1 K1 C1,C2,Cm ke turbulence model constant, C1 1.44, C2 1.92, Cm 0.09 E activation energy, J mol1 F1,F2,F3 cross-sectional area of the ue, gas inlet and air inlet, m2 f mixture fraction, % G incident radiation Gb turbulence kinetic energy due to buoyancy, kg m1 s1 Gk turbulence kinetic energy due to velocity gradient, kg m1 s1 H specic enthalpy, J kg1 Hg height of ue, m Hm latent heat of vaporization, J kg1 k turbulent kinetic energy, m2 s1 Kav,Ksv absorption and scattering coefcient, m1 P pressure, Pa PB,Ph bottom pressure of upward and downward owing, Pa R molar gas constant, J mol1 K1 S energy source term, W m3 SG radiative source term T temperature, K ui uid velocity in i direction, i 1,2,3, m s1 V1,V2,V3 mass ow of air, gas and waste gas, m3/s

x YM

amount of vapour condensation, % contribution of the uctuating dilatation in compressible turbulence to the overall dissipation rate, kg m1 s3 thermal diffusion coefcient, m2 s1

Greek letters turbulent dissipation rate, m2 s1 h ratio of circulation, % l thermal conductivity, W m1 K1 m,mt molecular viscosity and turbulent viscosity, Pa s1 r density, kg m3 s StefaneBoltzmann constant, W m2 K4 sk,sz turbulent Prandtl number, sk 1.0, s 1.3 umoisture water content umoisture,0 water content of charging coal umoisture,i amount of water evaporation in temperature range i, umoisture,1 about 75e80%, umoisture,2 about 20e25% wNO NO reaction rate, kmol m3 s1 s time, s [O2],[N2] oxygen and nitrogen concentration, kmol m3 s1 Subscripts m moisture g gas j c represent coal/coke, w represent inter-wall

et al. [11] rst proposed a one-dimensional coupled heat transfer model from the ue in the combustion chamber to the inter-wall and coking chamber. Matsubara et al. [12] presented a onedimensional coupled simulation model, including not only coking chamber, vertical and horizontal ue, but also regenerator with fuel ow rate supplied as inlet boundary conditions. Luo et al. [13] introduced a two-dimensional model, consisting of a couple of vertical ues and two adjacent chambers, and predicted ue temperature variation during a coking cycle and showed the same trend of concavity as experimental measurements, but less uctuation in quantity. Recently, we [14] developed a threedimensional transient mathematical model for coupled coking and combustion chambers. This model was proven to be valid with good accuracy in revealing the temperature-eld evolution of coke bed. Coke ovens without staging combustion (the height of coking chamber is equal to or less than 6 m) have been widely used in actual production [15]. For this kind traditional coke oven, the inlets of BFG (Blast Furnace Gas) and air are only located at the bottom of the ue. With the requirements of improving coke production and quality, along with energy saving, large-capacity coking oven is being developed. To overcome the deciencies of relatively high local combustion temperature and NOx concentration, the staging combustion technology is employed in the large-capacity coking oven [16]. As we know, the staging combustion has been widely applied in various industrial combustions such as coal-red boilers [17] and electrically heated tube reactor [18]. However, little data and literature are available for large-capacity coke oven and its related staging combustion, where the fuel gas and air are supplied into the combustion chamber via more than one inlet along the height direction of combustion chamber. In order to provide some basis and data for the design of large-capacity coke oven and the optimization production with staging combustion, transport phenomena in coke oven were simulated and analyzed in this paper. A three-dimensional transient model was built for coupled coking

and staging combustion chambers. Combustion, uid ow, heat and mass transfer were studied and compared with traditional coke ovens. Effects of staging combustion were evaluated and its optimization was further discussed for improving the temperature uniformity and reducing NO. More operation parameters were also investigated such as the temperature, moisture content and density of charging coal.

2. Three-dimensional model of transport phenomena in coupled coking and combustion chambers 2.1. Physical model The combustion chambers including serials of ues and coking chambers are arranged alternatively in parallel in the upper part of the coke oven. The air and BFG or COG (Coke Oven Gas) are supplied into the ues of combustion chamber and burn off. The hightemperature fume rises along the ue, via the turning port, then goes down. At the bottom of the chamber, some fume goes through the circulation port and mixes with fresh gas for recirculation, the others just ow out of the combustion chamber. The hightemperature fume (1400e1600  C) transfers heat to the inter-wall between combustion chamber and coking chamber by radiation and convection. Then heat is conducted through the wall to the coking chamber, where the coal isolated from the air is heated up gradually until it is stratied to the carbon coke. Three-dimensional model in this paper consisted of a pair of ues in the combustion chamber and two 1/2 coking chamber, as shown in Fig. 1. Different from traditional coking ovens, a largecapacity coke oven is often operated with staging combustion along the height of the chamber, to guarantee the temperature uniformity of the coke bed [16]. The channels of BFG and air for staging combustion are set in parallel in the walls between ues, as illustrated in the combustion chamber of Fig. 1b.

356

K. Jin et al. / Applied Thermal Engineering 58 (2013) 354e362

combustion chamber coking chamber

Table 1 Description of parameters for controlling equations in combustion chamber. Equation name 4 G 1 ui H k f S4

upper inlet of BFG / air coking chamber

middle inlet of BFG / air inlet of air inlet of BFG

Continuity Momentum Energy k Mass fraction

m a m rf Cmk2/ksk m rfCmk2/s mt/st

e vp/vxi e Gk Gb r Ym C1(Gk C3Gb)/k C2rf2/k RRR Brff fox exp E=RT pff ; fox ; T dff dfox dT

Fume

(a)Three-dimensional geometry
periodic boundary symmetric plane in cokiung oven symmetric plane in cokiung oven downward flow fume coking inter oven -wall wall between flues upward flow inlet of COG
inter coking -wall oven

1 qr $VG 3Kav Ksv A$Ksv  V  1 $VG A$G 4A$G$T 4 SG 3Kav Ksv A$Ksv

(3)

(4)

NO formation model (thermal NO model) [19]

wNO 4:52414 1015 T 2 exp 69466=T O0:5 N2

1

(5)

inlet inlet of air of BFG wall between flues periodic boundary

This model is often applicable for the oxygen enriched combustion process.

2.3. Boundary and initial conditions (1) Boundary conditions Coking chamber:

(b)Top view
Fig. 1. Physical model of the coupled coking and combustion chambers.

2.2. Transient mathematical models 2.2.1. Controlling equations in the coking chamber To simply the mathematical model, the following hypotheses were introduced: 1) The coking chamber is symmetric about the midplane. 2) Effects of the moisture vaporization, the volatiles release, and the chemical reaction related with coal pyrolysis were taken into account by temperature-dependent effective thermophysical parameters, such as effective thermal conductivity, effective specic heat and effective bulk density. 3) The production and the ow of raw gas in the coking chamber were neglected. Thus, the heat conduction equation for the coal/coke bed and inter-wall is:

Bottom :

 vT   0; vz z0

Top :

 vT   0; v z  z H

 vT   Midplane : 0; v x x 0
Combustion chamber:

Fuel gas inlet : mgas m0 ; Tgas T0 ; Air inlet : mair m1 ; Tair T1 ; Fume outlets : Pout P0 ;
(2) Initial conditions

rj Cj

vTj vs

vT v l j vx j vx

vT v l j vy j vy

vT v l j vz j vz

 S (1)

Inter wall : T jwall;t0 Tw ;

2.2.2. Controlling equations in the combustion chamber The general conservation equation describing the combustion, uid ow, heat and mass transfer is,

Charging coal : T jcoal;t 0 Tc ;

In order to improve simulation accuracy and efciency, the elds of the velocity, temperature and concentration were individually simulated for the combustion chamber in advance by employing adiabatic boundary conditions. Then the results were input as the initial conditions of combustion chamber for the following two chamber coupled simulation.

    v v  v vf rg f rg ui f G Sf vs vxi vxi vxi

(2)

The specic variable f and relevant parameters are listed in Table 1. Radiation model

K. Jin et al. / Applied Thermal Engineering 58 (2013) 354e362 Table 2 Operation parameters of coking ovens. Parameter Inlet volume ux of the BFG Inlet temperature of the BFG Inlet volume ux of the air Inlet temperature of the air Initial temperature of brick wall Temperature of the charging coal Mass fraction of moisture in the charging coal Density of the charging coal Value (6 m-height) 0.1026 1373 0.0956 1373 1400 363 9.3 736 Value (7 m-height) 0.129 1353 0.126 1353 1400 293 10 740 Unit m3/s K m3/s K K K % kg/m3 Table 3 Components of BFG. Components Volume fraction (%) CO2 10.52 O2 0.29 CO 26.78 H2 2.58 CH4 0.19 N2 55.28

357

H2O 4.36

equations were solved by matrix factorization algorithm. The residuals of iteration were less than 104. The initial time step was set to be 0.001 s and then increased with the advance of the calculation. 3. Results and discussion 3.1. Validation of the model

2.4. Chemical reaction heat and vaporization latent heat The chemical reaction heat and latent heat of vaporization were considered by introducing equivalent specic heat, denoted as Ce. The equivalent specic heat represented the variation of chemical reaction heat and evaporation latent heat with temperature. According to the experiment [20], the evaporation was divided into three phases: moisture condensation period T0eT1 (20  Ce 70  C), constant-rate drying period T1eT2 (70  Ce150  C), deceleration drying period T2eT3 (150  Ce300  C). For T0eT1 (moisture condensation), the amount of condensed moisture was, Due to the scantiness of the test data for large-capacity coking oven with staging combustion, here, a 6 m-height coking oven without staging combustion in the literature [22] was simulated to validate the developed model in this paper. Relevant parameters for calculation are listed in Table 2. The numerical results were achieved and the midpoint temperature of coke bed is illustrated in Fig. 2. It is clear that the numerical results are in good agreement with the measured data from real coke oven, with the relative error less than 5%. 3.2. Results of typical operation condition The typical operation parameters of a 7 m-height large-capacity coking oven with staging combustion are listed in Table 2.

umoisture 1 x$umoisture;0

(6)

For T1eT2 (constant-rate drying), T2eT3 (deceleration drying), the equivalent specic heat of evaporation latent heat was,

Ce;i

Hm $umoisture;i i 1; 2 Ti1 Ti

(7)

In the temperature range of Tm to Tn [21], if the chemical reaction heat was Hc (J kg1), the corresponding equivalent specic heat would be,

Ce

Hc Tm Tn

(8)

The above model was solved numerically using CFX CFD package. The hexahedral structured grid was used for the discretization of computation zone. The grid independence was studied and the number of the grids was set to be 0.9 million. The fully implicit coupled solver was set, which solves the hydrodynamic equations (for u, v, w, p) as a single system. The discretization of governing equations employed shape functions for diffusion terms and the second-order difference upwind scheme for advection terms. The algebraic

(a)Top view
0.6m centre of height direction 5 4 3 2 1 turning port

Z circulation port 0.6m X

(b) Height direction

Fig. 2. Comparison of numerical and measured midpoint temperature of coke bed. Fig. 3. Temperature-monitoring points in the coking chamber.

358

K. Jin et al. / Applied Thermal Engineering 58 (2013) 354e362

(1) Temperature evolution in the coking chamber The distribution of monitoring points in coal/coke bed and interwall is displayed in Fig. 3. The evolution of temperatures at monitoring points is shown in Fig. 4. It could be observed that the consumed time for coking completion is 21 h when the central temperature of coke bed achieves 1000  C, which is about 22 h in the actual production [16]. A very good agreement is found between the predicted and onsite consumed time with a relative error of 4.54%. Compared with 6 mheight coking oven, the consumed time for coking completion extends because of the larger amount of charging coal in largecapacity oven. For the inter-wall, at the quite early stage, the temperature does not increase, but decreases at the coke side because the inter-wall initially loses more heat to the lowtemperature coal than it absorbs from the ue gas. About 2 h later, the temperature begins to increase. For the coal/coke bed, the temperature rises all the time. However, the local temperature of the coke bed does not show an obvious rise until the moisture stops evaporation. The temperature proles of coke bed at ending time are shown in Fig. 5 for different location. From the temperature prole of the

Fig. 4. Evolution of temperature at monitoring points.

The specic heat, thermal conductivity and density of the coal/ coke and brick wall in the literature [21] were adopted. The components of BFG are shown in Table 3. The BFG/air equivalence ratio was 1.31.

(a) Interface between inter-wall and coal bed

(b) Plane near the inter-wall

(c) 1/4 plane (from the midplane)

(d) Midplane

Fig. 5. Temperature proles of coke bed (K).

K. Jin et al. / Applied Thermal Engineering 58 (2013) 354e362

359

upper inlet of BFG upper inlet of BFG

middle inlet of BFG middle inlet of BFG

X=0 plane

bottom inlet of BFG

plane A-A

bottom inlet of BFG

(a) Temperature profile

(c) CO concentration profile

turning port upper inlet of BFG upper inlet of BFG middle inlet of BFG middle inlet of BFG

plane A-A

circulation port

plane A-A

bottom inlet of BFG

(b) Velocity profile

(d) NO concentration profile

Fig. 6. Proles of temperature, velocity, CO and NO concentration at the midplane of the combustion chamber.

central plane (Fig. 5d), it is seen that the maximum temperature difference is 36  C along the height direction, while this value for traditional coking ovens without staging combustion is about 55  C [14]. Even if the bed height of large-capacity coking oven is higher, the temperature uniformity of the coke bed is improved, because the staging combustion is employed. (2) Distributions of velocity, temperature and concentration in the combustion chamber The switch of the ow directions of fume, fuel gas and air is neglectable in one cycle, and the transport phenomena in the combustion chamber are quasi-steady. The temperature and velocity proles at the midplane in combustion chamber are shown in Fig. 6a and b. The local distributions at turning port, upper inlet of BFG, middle inlet of BFG and

circulation port are magnied for display. Some high-temperature zones, so-called ame front, are observed around the ames. The gas temperature near the inlets is relatively high due to the radiation effect of the ame, and it decreases along the ow direction as the energy is absorbed by the coal in the coking chamber. The ratio of gas recirculation was dened, denoted as h, as the fraction of recirculated fume in the total amount of fume produced by combustion. The ratio of recirculation was 35% gained by numerical calculation. The theoretical value obtained from the

Table 4 Mass ow distribution of BFG/air for staging combustion. Upper inlet Mass Mass Mass Mass ow ow ow ow rate of BFG (kg/s) distribution of BFG (%) rate of air (kg/s) distribution of air (%) 0.006 17.8 0.00687 20.9 Middle inlet 0.0032 9.5 0.0034 10.4 Bottom inlet 0.0245 72.7 0.0226 68.7

Fig. 7. Evolution of NO concentration at the outlets of combustion chamber.

360

K. Jin et al. / Applied Thermal Engineering 58 (2013) 354e362 Table 5 BFG/air mass ow distribution of different cases. Case 1 Upper inlet (%) Middle inlet (%) Bottom inlet (%) 10 5 85 Case 2 (typical condition) 20 10 70 Case 3 30 10 60

empirical formula (9) [5] is 37.5%, which is consistent with our simulation result. The ratio of recirculation has great effects on the gas temperature distribution and it is an important parameter in the actual operation of a coke oven.
2 rgas Tgas V1 V2 V 2 1 h2 ro T0 r T 2 air air 3 2 F1  F2 273 F1  F3 273 273 F1

(9)

Hg r2 r1 Ph PB S1H DP

The mass ow distribution of BFG/air at each inlet for staging combustion was calculated and listed in Table 4. The bottom inlets have the maximum mass ow rate and middle inlets have the minimum value. Fig. 6c displays the concentration distribution of CO at the midplane in the combustion chamber. It is relatively straightforward to see the ame lengths, thereby the effect of staging

combustion. From top to bottom, the ame lengths are 0.87 m, 0.28 m and 1.27 m, respectively. The ame length is related with the gas ow rate. Some NO is produced in the combustion and NO concentration distribution at the midplane is shown in Fig. 6d. It can be seen that NO concentrates at the ame front where temperature is the highest as shown in Fig. 6a. The variation of NO concentration with the time is shown in Fig. 7 at the outlets of combustion chamber. It displays that the amount of NO increases with the time. Along with the coking process, the temperature difference between combustion and coking chambers decreases, so does the heat loss of combustion chamber, and thereby more NO produces with increasing temperature in the combustion chamber. 3.3. Optimization of staging combustion Different cases of staging combustion in 7 m-height largecapacity coking oven were numerically simulated and compared with traditional 6 m-height coking oven without staging

A: Large-capacity oven with three staging combustion; B: Large-capacity oven with two staging combustion; C: Large-capacity oven without staging combustion; D: Traditional oven without staging combustion; (a) Maximum temperature difference of coke bed midplane

(a) Maximum temperature difference of coke bed midplane

A: Traditional oven without staging combustion; B: Large-capacity oven without staging combustion; C: Large-capacity oven with two staging combustion; D: Large-capacity oven with three staging combustion; (b) Evolution of NO concentration at combustion chamber outlets
Fig. 8. Results under different staging combustion conditions.

(b) Evolution of NO concentration at combustion chamber outlets

Fig. 9. Results under different BFG/air distributions.

K. Jin et al. / Applied Thermal Engineering 58 (2013) 354e362

361

combustion [14]. For analysis, the number of fuel gas inlets was changed only, while the air inlets were xed to be three. Maximum temperature differences at the coke bed midplane in the end of coking process and the evolutions of NO concentration are illustrated in Fig. 8 under different staging combustion conditions. The maximum temperature difference of large-capacity coking oven would be 89  C if without staging combustion, which is much higher than traditional coking oven and even the value of 50  C generally required for the coke quality [5]. Once the staging combustion are employed, e.g. two staging combustion, the temperature uniformity could be improved to an acceptable level with the temperature difference of 52  C. The maximum temperature difference goes down to exciting 36  C for three staging combustion. It is indicated in Fig. 8b that staging combustion is benecial

for NO reducing. The staging input decreases local gas concentration, so the ame front temperature drops, and thereby reducing the formation of NO. An appropriate mass ow distribution of BFG/air input is important for staging combustion of coke ovens since it affects the temperature uniformity of coke bed, and NO concentration of exhaust. In this article, three cases of mass ow distribution, as listed in Table 5 were discussed for production optimization. The maximum temperature difference of coke bed midplane and evolution of NO concentration at combustion chamber outlet are shown in Fig. 9. It turns out that case 3 has the minimum temperature difference and NO concentration. To some extent, decreasing the mass ow rate at the bottom inlet but increasing the rate at the upper inlet is benecial to the temperature uniformity improvement and NO concentration reduction. For three staging combustion, as the mass ow rate ratio of the upper to the bottom is larger than 2/7, the nal temperature difference of coke bed is less than 40  C. 3.4. Effects of operational parameters According to production practice, the temperature, moisture content and density of charging coal have signicant impacts on coking process. These factors were investigated and the central temperature of the coke bed was chosen to exhibit the inuences, as displayed in Fig. 10. It is shown that coal preheating, coal moisture decreasing and coal compacting are benecial to the improvement of the production efciency. 4. Conclusions

(a) Density
For the coupled coking and combustion chambers in largecapacity coking ovens, a three-dimensional unsteady model which could describe the burning, owing, mass and heat transfer process was set up and numerically solved. The model was validated by measurements from real coke oven without staging combustion. Effects and optimization of staging combustion were discussed. More operation parameters were also analyzed. The following conclusions were obtained: (1) For 7 m-height large-capacity coking oven with three staging combustion, the consumed time for coking completion is about 21 h and the maximum temperature difference of coke bed is 36  C in the end. The ratio of gas recirculation in the combustion chamber is 35%, the gas mass ow rate at the bottom inlet is the largest and the value at the middle inlet is the least. (2) Staging combustion in large-capacity coke oven can improve the temperature uniformity of the coke bed and reduce NO concentration of exhaust, which is even superior to the traditional small-capacity coke oven. It is benecial for production optimization to decrease the gas mass ow rate at the bottom inlet but increase the rate at the upper inlet in the combustion chamber. For three staging combustion, as the mass ow rate ratio of the upper to the bottom is larger than 2/7, the nal maximum temperature difference of coke bed is less than 40  C. (3) Coal preheating, coal moisture decreasing and coal compacting are benecial to the improvement of production efciency. Acknowledgements The authors are grateful for the support from the National Hightech Research and Development Program of China (2009AA063301).

(b) Temperature

(c) Moisture content

Fig. 10. Effects of density, temperature and moisture content of charging coal on the midpoint temperature of coke bed.

362

K. Jin et al. / Applied Thermal Engineering 58 (2013) 354e362 [12] K. Matsubara, O. Tajima, N. Suzuki, Y. Nakayama, T. Kato, New simulation model for coke oven and some applications (special issue on ironmaking), ISIJ Int. 68 (15) (1982) 2148e2155. [13] G.Q. Luo, Z. Wen, H.F. Chen, Z.L. Gao, A ue-oven mathematical model for coke oven, Fuel Chem. Process. 29 (2) (1998) 78e82. [14] K. Jin, Y.H. Feng, X.X. Zhang, W. Lin, C.Q. Zhang, J.F. Yang, X.B. Ma, Numerical analysis on thermal processes in coupled coking and combustion chambers of coke oven, J. Chem. Ind. Eng. 63 (3) (2012) 788e795. [15] Y. Wang, Analysis on 6 m coke oven thermotechnical parameters and energy saving measure, Fuel Chem. Process. 40 (4) (2009) 17e22. [16] B.Q. Jin, The practice of large capacity coke oven, Fuel Chem. Process. 42 (5) (2011) 21e22. [17] L.S. Xiao, H.C. Zeng, F. Jin, J. Han, Effects of air staging and particle size on NOx emissions and unburnt carbon in y ash, Power Eng. 1 (2001) 1042e1045. [18] H. Spliethoff, U. Greul, H. Rudiger, K. Hein, Basic effects on NOx emissions in air staging and reburning at a bench-scale test facility, Fuel 75 (5) (1996) 560e564. [19] G. Lofer, R. Sieber, M. Haraseka, H. Hofbauer, R. Hauss, J. Landauf, NOx formation in natural gas combustion d a new simplied reaction scheme for CFD calculations, Fuel 86 (2005) 512e523. [20] H. Kunmasa, T. Chikasa, K. Nishioka, E. Inoue, Mechanism of moisture transfer across the oven width at early stage of carbonization, Steel 82 (5) (1996) 393e396. [21] J.L. Si, Modelling of Layered Transmission Process in Heterogeneous Porous Media and Its Application. Ph.D Thesis, University of Science and Technology Beijing, China, 2009. [22] F. Huo, The Mathematical Model of Longitudinal Temperature and Its Application in the 3# Coke Oven in Baosteel. Master Thesis, University of Science and Technology Beijing, China, 2003.

References
[1] M.A. Diez, R. Alvarez, C. Barriocanal, Coal for metallurgical coke production: predictions of coke quality and future requirements for cokemaking, Int. J. Coal Geol. 50 (4) (2012) 389e412. [2] P. Wang, K.L. Bei, G.J. Zhang, Y.F. Zhang, Research present status and developmental trend of mathematical model during coke oven heating process, Chem. Ind. 26 (1) (2006) 47e49. [3] B.H. Yue, X.G. Wang, X.P. Ai, J. Yang, L. Li, X.G. Lu, W.Z. Ding, Catalytic reforming of model tar compounds from hot coke oven gas with low steam/ carbon ratio over Ni/MgOeAl2O3 catalysts, Fuel Process. Technol. 91 (9) (2010) 1098e1104. [4] H.X. Zhu, H.B. Wu, X.L. Wang, F. Li, G.Q. Luo, Z. Wen, Present status and developmental trend of mathematical model for coke oven heating process, Ind. Heat. 3 (2000) 4e7. [5] Z.Z. Yao, M.D. Zheng, Coking Plant Science, third ed., Metallurgical Industry Press, Beijing, 2008. [6] D.Q. Tian, Computer analysis of heat transfer in coke oven, Tech. Rep. Fuji Steel 8 (1969) 31e43. [7] Rodhe, Calculation and measurement of the non-stationary temperature eld in the coke soft and pylolysis, Chem 1 (1969) 1e8. [8] D. Merrick, Mathematic model of the thermal decomposition of coal: 1. The evolution of volatile matter, Fuel 62 (5) (1983) 534e539. [9] D. Merrick, Mathematical modelling of the coking process, Ironmak. Conf. Proc. 43 (1984) 287e302. [10] Z.C. Guo, H.Q. Tang, Numerical simulation for a process analysis of a coke oven, China Particuol. 3 (6) (2005) 373e378. [11] H.F. Chen, P.J. Readyhough, J.K. Brimacombe, Physical model of combustion chamber in coke oven, J. Univ. Sci. Technol. B 3 (1982) 24e43.