Study of combustion optimization for low NO x reform coal ... · PDF fileStudy of combustion optimization for low NO x reform coal-fired ... (Transformation and upgrading of coal power

Study of combustion optimization for low NOx reform coal-fired

boiler in a 600MW unit

Xiaoyu Zhang, Lin Fu, Yongsheng Fan, Junjie Zhang,

(Shenhua Guohua (Beijing) Electric Power Research Institute Co., Ltd., Beijing, 100025, China

Email: [email protected])

Jiong Shen, Yiguo Li

(School of Energy and Environment, Southeast University, Nanjing, 210096, China)

ABSTRACT:

Along with the most stringent emission standards in China, high efficiency and low NOx burner reform in coal-fired boiler has been largely performed. Reduction of boiler combustion emissions requires the development of coal air-staged combustion and slagging prevention. Steam parameters, carbon content in fly ash and regulation quality after low NOx burner reform are significantly different from the typical coal-fired boiler.

For retrofitted boiler, main combustion zone is condensed (size of primary air nozzle is narrowed), and a large reduction zone (vertical height nearly 7-9 meters) is formed between main combustion zone and separated over-fired zone (SOFA, consisting of seven-layer auxiliary air). This new design can make the most of reduction of nitrogen removal theory to reduce NOx, while it also brings the risk of furnace slagging. Characteristics of steam temperature, boiler efficiency and NOx emissions are studied under different load. Numerical and test results are compared.

Reheat temperature is lower than designed value, and desuperheating water and carbon content in fly ash are ascended after burner reformed. Flame is stretched, furnace temperature is more homogeneous. NOx concentration can be lowered to within 150 mg/Nm3, and carbon content in fly ash is less than 2% at the furnace outlet under the whole load range. Efficiency of retrofitted boiler can reach above 94%. A portion of combustion (10%-30%) is accomplished in over-fired area. Endothermic portion of water wall and radiant superheater is adjusted through moving the flame center position. With increase of bellows differential pressure, upper combustion portion raised. Upper combustion portion is condensed and burner tilts up during the underload performance (below 40% load) to gain the rating steam temperature. Numerical results indicate that high temperature area enlarged with increase of height in the main combustion zone, and NOx concentration decreased evidently in the large reducing zone. Air-staged combustion is the dominating factor to lower NOx concentration, instead of the low oxygen running. The results are expected to provide valuable information for the design and operation of new low NOx tangentially coal-fired boiler. KEYWORDS: Combustion Optimization, Coal-fired Boiler, Numerical Simulation, Test, Multi-objectives

1. Introduction Along with the most stringent emission standards (Transformation and upgrading of coal power

for energy conservation and emissions reduction action plan (2014-2020)) in China, high efficiency and low NOx burner reform in coal-fired boiler has been largely performed. Due to the large-scale application in short time, and technical digestion, absorption and improvement are delayed, occurring some common problems, such as reheat temperature is lower than designed value and desuperheating water and carbon content in fly ash are ascended. Combustion optimization is an effective way to solve these problems, through exploiting units’ potentialities. The paper compares the unit test results and numerical results, revealing the furnace combustion share and NOx generation variation, also giving some operation suggestions about low nitrogen transformed boiler.

Using the three dimensional simulation, we can get lots of information about boiler combustion working conditions, such as temperature distribution, flow field and species field. Effects of secondary air distribution, excess air coefficient and pulverized coal concentration under different load on furnace average temperature field and NOx formation have been studied extensively [1-5], which helps providing predictions on optimization of furnace air distribution, the variable load operation and NOx emissions. Sun et al. [6] utilized the FLUENT software to study the effects of different SOFA (Separated Over-Fire Air) distribution mode on furnace flow, combustion, and NOx formation in a 600MW tangentially coal-fired boiler, it is concluded that the depth of the air-staged combustion is a main inhibitory factor to NOx formation in main combustion zone, which shows good agreement with the test results.

Some others made studies on combustion characteristics [7-9], slagging in the furnace [10-14] and soot formation [8]. A 500 MW tangentially-fired boiler is studied by Choi and Kimb [7], considering air classification and grade of air distribution mode and fuel's impact on NOx formation. Park et al. [9] used a new kind of grid technology to reduce the numerical pseudo diffusion. Zhang et al. [10] investigated low NOx combustion strategies in tangentially coal-fired boilers, effects of HBC (Horizontal Bias Combustion) and OFA (Over-Fire Air) on NOx reduction are compared. HBC makes a significant NOx reduction in primary combustion zone when there is no air staging, and OFA has a remarkable effect on the reduction of NOx emissions. The prediction has a good agreement with on-site measurement results. Li et al. [11] studied load effects on combustion performance and NOx emission of a 220 MW coal fired boiler, it concluded that the distributions of temperature, O2 and CO concentration inside furnace with different loads show good similarly. Zhi et al. [12] investigated the formation and evolution of reductive and corrosive gases during air-staged combustion of blended coals. The blended coal mentioned in the paper has less impact on the NOx reduction compared with single coal respectively. Sun et al. [13] studied the flow, combustion and NOx emission characteristics in a 660MW tangential firing ultra-supercritical boiler. Results indicate that the temperature of flue gas in the combustion zone decreases a little with the application OFA and AA. Yang et al. [14] tried to use computational fluid dynamics to diagnose the combustion characteristics in two identical coal-fired boilers. The swirl strength of burners has a linear relationship with the gap of air register vanes.

Recently, much experimental research has been conducted to improve the understanding of tangentially fired boiler combustion optimization. Using combustion adjustment to improve boiler steam parameters and reduce the emissions has been carried out early, can be divided into furnace

parameter optimization and parameters outside the furnace. Furnace parameters optimization including air distribution optimization [15], the swinging Angle of burners[16] and bellows differential pressure [17], oxygen content [18,19], and so on. Parameters outside the furnace incorporate pulverized system optimization [20], auxiliary air temperature [21], wind speed [22] and the combination of coal mill [23,24].

Although combustion optimization of tangentially coal-fired boiler has been extensively studied, the influence factors on distributions of species along the furnace height are still limited, and further, timely adjustment measures on combustion optimization need to be formulated, especially in the low NOx retrofitted boilers. In practice, it is useful to have a general chart to describe the characteristics of combustion in low NOx retrofitted boiler with variations in the dominant factors. The main objectives of the present work were to investigate the temperature and species distribution in low NOx reformed boiler, analyzing the effects of deep air staging technology on combustion and pollutants formation, comparing the numerical results with test results, and giving combustion optimization suggestions on improving boiler efficiency and reducing NOx emissions, eliminating the steam temperature deviation. In the next section, the utility boiler parameters and research methods are described. The results and discussions are presented in the following part. Finally, general conclusions and further work are provided.

2. Boiler description and Research Methods 2.1 Utility boiler description

The experiments were carried out on a 600 MW tangentially bitumite-fired boiler showed in Fig. 1, with the furnace width of 17.558m, 19.558m in depth and 73.6m in height. The boiler was initially produced by Shanghai boiler plant with type of SG–2028/17.5–M908 subcritical parameter controlled circulation boiler. Horizontal bias combustion (HBC) and air-staging combustion technologies were employed in the low NOx reformed procedure. There were twenty burner nozzles at each corner, including six primary air (PA) nozzles, seven auxiliary air or secondary air (SA) nozzles, and seven separated over-fired air (SOFA) nozzles. The angle between PA and SA is 7 degrees. The initial primary combustion zone (PCZ) is compressed to 8.2m, and the height of SOFA reached 4.3m, then a huge reduction zone between PCZ and SOFA is formed. This new design can make the most of reduction of nitrogen removal theory to reduce NOx, while it also brings the risk of furnace slagging.

The designed coal was Shenhua blended bitumite coal with volatile content 30.8% and low heat value 24.12 MJ/kg. The proximate and ultimate analysis of the used coal is shown in table 1. The designed pulverized coal fineness R90 was 18%.

Table 1 Ultimate and ash fusibility of coal analysis

Industry analysis / %

Element analysis / %

Fusion characteristics / ℃

LHV / MJ•kg-1

Mar Aar Var Car Har Nar St,ar Oar DT ST FT Qnet,ar 9.92 11.04 30.83 58.4 3.80 0.97 0.34 9.45 1090 1120 1160 24.12

Figure 1 Schematic of the studied utility boiler

2.2 Numerical models and mesh system The numerical simulation in this study is supported using a commercial soft package, ANSYS

FLUENT 6.3.26. The coal combustion is a complex process including turbulent flow, coal devolatilization, homogeneous volatile combustion, heterogeneous char combustion, and radiative heat transfer [25]. The realizable k–ε model with a correction via the buoyancy effect was used to simulate the turbulent flow. The flow of the pulverized coal particles was simulated by the stochastic particle trajectory model. The devolatilization was estimated via the dual rate competition model, which considers the different release rates of the gas components. Char combustion was calculated by a kinetic/diffusion-limited model. Turbulent gas-phase combustion was simulated by the species transport model, which directly solves the transport equations for all species included in the problem definition [26]. The P1 model is employed for the radiation modeling. In this study, the eddy-dissipation concept model was used to calculate the interaction between the turbulence and the gas-phase reactions, which provides good predictions of flue gas mixture and carbon monoxide concentration [27].

Additionally, the Lagrangian discrete phase model is used to consider the pulverized-coal injection and the mass, momentum, and heat exchange between the discrete and continuous phases. The incompressible ideal gas and mass-weighted mixing law are chosen to define the density, viscosity, and absorption coefficient of the gas phase mixture. The particle size distribution is simulated using the Rosin–Rammler equation. Wall condition is no slip, and standard wall function is employed. The temperature of wall is 450K, radiation coefficient is 0.5. NOx is formed mainly by thermal NOx, fuel NOx, and prompt NOx, in this study prompt NOx is ignored.

The inlet temperature and speed of primary air and coal is 77 degree centigrade, and 24.8m/s, respectively. Secondary air temperature is 323 degree centigrade, with speed of 52.8 m/s. The coal composition is the same with designed value which mentioned above.

The calculation domain is considered to be from the hopper to the furnace outlet, including the primary nozzles and secondary nozzles, as presented in figure 2. The total structured mesh is

AAA

ABB

BCC

CDD

DEE

EFF

FF

SOFA1SOFA2SOFA3SOFA4SOFA5SOFA6SOFA7

9.1m

8.2m

4.3m

Primary Combustion

Zone

Reduction Zone

OFA

17.448m73.6m

around 2.18 million hexahedron cells. Grid refinement is employed in the PCZ and SOFA domain. The minimum volume of grid is 3.176e-6 m3, and the maximum is 6.520e-1 m3. To reduce the pseudo-diffusion in the combustion zone, the primary zone cross section is divided into radial grid, showed in figure 2 (b).

(a) the overall grid

(b) furnace cross-section grid

Figure 2 Mesh systems for different cross-section

3. Results and Discussions Fig.3 shows the main parameters of a low nitrogen reformed boiler. NOx concentration can be

lowered to within 150 mg/Nm3 in the stable condition, exhaust gas temperature is below 130 degree centigrade, and operation oxygen varies from 3.1% to 5.2%. Efficiency of retrofitted boiler can reach above 94% for this new design.

Figure 3 Operating parameters of a 600 MW low NOx retrofitted boiler

3.1 Temperature and Species Distribution In order to understand the effects of reformed scheme on combustion characteristics, numerical

0 2000 4000 6000200

400

600

800

t / s

Loa

d / M

W

Load

0 2000 4000 6000100

150

200

250

t / s

NO

x / m

g/m

3

NO

x

0 2000 4000 6000110

120

130

140

t / s

T /

℃

Flue gas temperature

0 2000 4000 60003

4

5

6

t / s

O2 /

%

Operation Oxygen

study on stable conditions was performed. Figure 4 presented the temperature distribution in the full load condition. The flame is stretched, and high temperature area is extended from the PCZ to the OFA. It is evident that the temperature in the hopper section is still high, which is different from the traditional boiler. The temperature in the reduction zone is nearly 2000K, and slagging risk is also ascended.

Figure 4 Temperature field in the center cross-section of reformed boiler

Figure 5 shows the compared temperature results between retrofitted boiler and non-reformed one. Temperature distribution is more homogeneous after the modification, and it is beneficial for reducing NOx concentration. Meanwhile, the high temperature area enlarged from the PCZ to the RZ. Before modification, high temperature mainly concentrated in the PCZ, and the maximum temperature can reach 2100K, which results in a high NOx formation.

Figure 5 Average temperature distribution of cross-section along the furnace height

Numerical results and test results are compared in the figure 6. The employed model in the present study is appropriate to capture the information about the combustion condition in the

furnace. The maximum value of CO occurred in the D layer in the PCZ, numerical and test results agreed well in the whole range. CO content in the PCZ and RZ can reach as high as 800μL/L from the test results, more attention should be paid to high temperature corrosion in the ordinary operating condition. In order to decrease CO, the damper opening of AB and CD should be turned up.

Figure 6 Average CO content distribution compared with numerical and test results

Fig.7 presents the numerical and test results of NO content distribution along furnace height. Numerical results are lower than test results in the PCZ and OFA, but the tendency is consistent. NOx concentration is controlled within 150mg/Nm3, the deep air staging combustion technology is efficient to control the NOx formation, where its content decreases evidently in the RZ. With the air addition from the SOFA nozzles, NOx is ascended later.

Figure 7 Average NO content distribution compared with numerical and test results

In order to investigate the effects of oxygen content and SOFA distribution mode on NOx formation, Fig.8 and Fig.9 present the O2 content with NO along furnace height and three

0 10 20 30 40 50 60 700

0.5

1

1.5

2

2.5x 10-3

CO

cont

ent /

%

Furnace height / m

PCZ RZ OFA

0 10 20 30 40 50 60 700

200

400

600

800

1000

CO

conc

entra

tion

/ uL/

L

Numerical results Test results

0 10 20 30 40 50 60 700

2

x 10-3

NO

cont

ent /

%

Furnace height / m

PCZ RZ OFA

0 10 20 30 40 50 60 70

50

100

150

NO

x / m

g/m

3

Numerical results Test results

different SOFA horizontal angles cases, respectively. From figure 8, it is obvious that the NO content distribution have the same shape with oxygen distribution, which means the dominant factors on NOx formation are operating oxygen after modification. Due to the combustion is divided into PCZ and OFA, the combustion share of two zones can be adjusted through changing the bellows differential pressure. With the opening of SOFA damper, the combustion share in OFA increased which is helpful to improve the steam temperature. Meanwhile, oxygen content in PCZ should not be controlled less than 1.2%, due to the high CO content. Figure 9 gives three SOFA horizontal angle cases. The horizontal angle is zero in case1, case2 means the SOFA has an angle of 4 degree with the primary air, and the angle in corner 2 is zero in case3 compared with case2.

Figure 8 Average Oxygen and NO content distribution along the furnace height

Figure 9 NO distribution of three different SOFA horizontal angle case

From figure 9, it is easy to find that the NO mass fraction distribution is similar in three different cases. The horizontal angle of SOFA has little impact on NOx formation.

Fig.10 shows the numerical results of main species distribution in the furnace. Volatiles were

0 10 20 30 40 50 600

1

2

3x 10-3

NO

mas

s fra

ctio

n / %

Furnace height / m

PCZ RZ OFA

0 10 20 30 40 50 600

0.02

0.04

0.06

O 2 m

ass

frac

tion

/ %

NO O

2

0 10 20 30 40 50 600

0.5

1

1.5

2

2.5

3x 10-3

Furnace height / m

NO

mas

s fra

ctio

n / %

PCZ RZ OFA

case1case2case3

mainly concentrated in the PCZ and RZ, and it consumed quickly with the auxiliary air injection after coal devolatilization. Concentration of CO2 was oscillated because of coal and air injection alternatively in the PCZ, and it decreased from the RZ. The main combustion share is still maintained in the PCZ after furnace modification.

Figure 10 Main species distribution along the furnace height

3.2 Eliminating Steam Temperature Deviation Steam temperature deviation often occurred in the low NOx retrofitted boiler. Figure 11 shows

the superheater platen temperature between the left side and right side. There was nearly 8 degree centigrade deviation between the two sides.

Figure 11 Temperature distribution of superheater platen

In order to improve the boiler operation safety, measures to diminish temperature deviation of two sides is presented in figure 12. Through changing the horizontal angle of SOFA, it is helpful to move the position of ‘center fire ball’ in the furnace. The dashed arrows indicate the adjustment direction. The temperature deviation becomes smaller after the adjustment showed in figure 13.

0 10 20 30 40 50 600

1

2

3

4

5

O2 -

Vol

atile

s mas

s fra

ctio

n / %

Furnace height / m

PCZ RZ OFA

0 10 20 30 40 50 6014

15

16

17

18

19

CO

2mas

s fra

ctio

n / % O

2

CO2

Volatiles

0 2 4 6 8 10 12572

574

576

578

580

582

584

Gauging point

Wal

l Tem

pera

ture

/

℃

Left superheater platen Right superheater platen

The deviation can be controlled within 2 degree centigrade in the full load condition after the adjustment.

Figure 12 Schematic of SOFA horizontal angle adjustment

Figure 13 Variation of superheater temperature and temperature deviation

3.3 Combustion Optimization measures In order to get the general adjustment formulation for low NOx retrofitted boiler, a lot of on-site

experiments have been completed. Table 2 gives the optimum distribution mode of SOFA in the whole load range. With the increase of load, the damper opening of SOFA is opened from the bottom layer to the top layer, and it is also necessary to keep the bellows differential pressure to an appropriate value to ensure the validity of air distribution mode. The reheat steam temperature is adjusted through the vertical angle of PCZ and SOFA. Inlet NOx concentration of SCR is controlled by deep air-staging combustion, and operation oxygen is the dominant factor.

Table 2 Optimum distribution mode of SOFA in the whole load range

Load（MW） 400 450 500 540 600 SOFA-V（%） 0 0 20 50 70 SOFA-IV（%） 0 0 60 70 80 SOFA-III（%） 30 40 90 90 90 SOFA-II（%） 50 70 90 100 100

400 450 500 550 600 650 700550

555

560

565

570

575

Sup

erhe

ater

ste

am te

mpe

ratu

re /

℃

Load / MW

400 450 500 550 600 650 7000

2

4

6

8

10

Ste

am te

mpe

ratu

re d

evia

tion

/

T T deviation

SOFA-I（%） 90 90 90 100 100 COFA-II（%） 10 10 30 30 40 COFA-I（%） 10 10 30 30 40

SOFA Angle（o） 75 75 80 80 75 Burner Angle（o） 65 65 70 65 65

O2（%） 4.5 4.4 4.3 4.0 3.0 Bellows differential

pressure（Pa） 550 600 650 650 680

SCR inlet NOX

（mg/m3） 150 160 150 120 135

4. Conclusions The characteristics of temperature, species concentration and NOx emissions in a 600MW low

NOx retrofitted boiler have been numerically and experimentally investigated. The good agreement of numerical results and test results are obtained.

Comparison of the test results and numerical results indicate that CO content in PCZ and RZ maintained a high level, which means the risk of high temperature corrosion rising after boiler modification. NOx concentration decreased sharply in the OFA, the SOFA makes a major contribution to the NOx reduction. The horizontal angle of SOFA can eliminate the temperature deviation of superheater platen. A general combustion optimization direction is pointed out for low NOx retrofitted boiler.

The results improve the understanding of combustion, NOx formation and combustion optimization measures for low NOx retrofitted boilers.

References

[1] Huang Wenjing, Miao Zhengqing, Wang Cicheng, Li Jiangtao. Numerical Simulation Study of Low NOx Emissions Combustion in a 300MW Tangentially Fired Boiler. Boiler Technology, 2014, 45(3): 39-43.

[2] Liu Liping. Numerical Simulation of Combustion Process and Air Distribution of Tangentially Pulverized Coal-fired Boiler: [Master Dissertation]. Dalian University of Technology, 2008.

[3] Meng Fanjing. Numerical Simulation of Combustion and NOx Exhaust in a Tangential Fired Boiler: [Master Dissertation]. Dalian University of Technology, 2008.

[4] Wang Chunbo, Wei Jianguo, Sheng Jingui, Li Yanqi. Numerical Simulation of Combustion Characteristics of a 300MW Blast Furnace Gas/Pulverized Coal Combined Combustion Boiler. Proceedings of the CSEE, 2012, 32(14): 31-33.

[5] Fan Xianzhen, Guo Liejin, Gao Hui, Nie Jianping. Numerical Simulation of Flow and Combustion Process in the Tangentially Fired Furnace of a 200MW Pulverized Coal Boiler. Journal of Xi’an Jiaotong University, 2002, 36(3): 241-245.

[6] Bai Tao, Sun Baomin, Kang Zhizhong, etc. Numerical Investigation on the Characteristic of Low NOx Combustion in a 600MW Tangential Boiler. Boiler Technology, 2014, 45(2): 35-40.

[7] Choeng Ryul Choi, Chang Nyung Kimb. Numerical investigation on the flow, combustion and NOx emission characteristics in a 500 MWe tangentially fired pulverized-coal boiler. Fuel, 2009, 88(1): 1720-1731.

[8] Chungen Yin, Sébastien Caillat, Jean-Luc Harion, Bernard Baudoin, Everest Perez. Investigation of the flow, combustion, heat-transfer and emissions from a 609 MW utility tangentially fired pulverized-coal boiler. Fuel, 2002, 81(1): 997-1006.

[9] H.Y.Park, M.Faulkner, M.D.Turrell, P.J.Stopford, D.S.Kang. Coupled fluid dynamics and whole plant simulation of coal combustion in a tangentially-fired boiler. Fuel, 2010, 89(8): 2001-2010.

[10] Xiaohui Zhang, Jue Zhou, Shaozeng Sun, etc. Numerical Investigation of low NOx Combustion Strategies in tangentially-fired coal boilers. Fuel, 2015, 142, 215-221.

[11] Jun Li, Radosław Jankowski, Michał Kotecki, etc. Numerical analysis of loads effect on combustion performance and NOx emissions of a 220 MW pulverized coal boiler. Cleaner Combustion and Sustainable World, Springer Berlin Heidelberg, 2013, Part VI, 1019-1029.

[12] Zhi Zhang, Zhenshan Li, Ningsheng Cai. Formation of reductive and corrosive gases during air-staged combustion of blends of anthracite/sub-bituminous coals. Energy & Fuels, 2016, 30: 4353-4362.

[13] Wenjing Sun, Wenqi Zhong, Aibing Yu, etc. Numerical investigation on the flow, combustion and NOx emission characteristics in a 660 MWe tangential firing ultra-supercritical boiler. Advances in Mechanical Engineering, 2016, 8(2): 1-13.

[14] Joo-Hyang Yang, Jung-Eun A. Kim, Jaehyun Hong, etc. Effects of detailed operating parameters on combustion in 500-MWe coal-fired boilers of an identical design. Fuel, 2015, 144(15): 145-156.

[15] Zhang Yongsheng, Lv Xuyang, Li Chunxi, Yan Huibo. Analysis on Wind Volume Optimization Adjusting after Low NOx Burner Reformation of 600MW Unit Boiler. Hebei Electric Power, 2014, 33(3): 3-6.

[16] Guan Fengyi, Xu Youning, Zhang Qian. Experimental Investigation on Low NOx Burner Transformation of 300 MW Boiler. Journal of Shenyang Institute of Engineering (Natural Science), 2014, 10(4): 321-324.

[17] Sun Yipeng, Cheng Liang, Zhu Xianran, Zhang Qingfeng. NOx Emission Reduction Experiments Research by Combustion Adjustment on Boiler Burning Low Ash Fusion Point Coal. Power System Engineering, 2013, 29(4): 29-31.

[18] Gu Lijing, Li Yonghua, Li Lu. Hybrid Model Prediction of Utility Boiler Combustion Optimization. Proceedings of the CSEE, 2015, 35(9): 2231-2237.

[19] Xiao Haiping, Zhang Qian, Wang Lei. Sun Baomin. Effect of Combustion Adjustment on NOx Emission and Boiler Efficiency. Proceedings of the CSEE, 2011, 31(8): 1-6.

[20] Deng Junci. Analysis of Optimal Combustion Test of 600 MW Subcritical Boiler after Low NOx Modification. Guangxi Electric Power, 2014, 137(3): 82-85.

[21] Gao Jilu, Zhang Yong, Jiang Chong. Study on Combustion Optimization Adjustment Test for 600 MW Supercritical Boiler. Northeast Electric Power Technology, 2011, 32(12): 7-10.

[22] Pan Guoqing, Xu Xiaoqiong, Ying Mingliang, Cai Jiecong. Combustion Optimization for 1000MW Ultra-Supercritical Tower Boiler. Zhejiang Electric Power, 2011, 7: 30-33.

[23] Sun Yipeng, Cheng Liang, Yu Yang, Zhang Qingfeng. The Impact of Shenhua Coal Changing Combustion on Coal Pulverizing System and Boiler Operation. Proceedings of the CSEE, 2013, 33(S1): 122-127.

[24] Zhao Zhenning. Compressive Optimization of Primary Air Parameters of Medium-speed Pulverizing System. Proceedings of the CSEE, 2010, 30(S1): 124-130.

[25] Zheng Chu, Liu Zhao, Duan Xue, etc. Numerical and experimental investigations on the performance of a 300MW pulverized coal furnace. Proceedings of the Combustion Institute, 2002, 29:811-8.

[26] Fang Qingyan, Musa A.B. Musa, etc. Numerical simulation of multi-fuel combustion in a 200MW tangentially fired utility boiler. Energy Fuels, 2011, 26: 313-23.

[27] Jingzhang Liu, Sheng Chen, Zhaohui Liu, etc. Mathematical modeling of air- and oxy-coal confined swirling flames on two extended eddy-dissipation models. Industrial & Engineering Chemistry Research, 2011, 51: 691-703.