- Email: [email protected]
Experimental study of the effect of a radiant tube on the temperature distribution in a horizontal heating furnace
Applied Thermal Engineering 113 (2017) 1–7
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Research Paper
Experimental study of the effect of a radiant tube on the temperature distribution in a horizontal heating furnace H.T. Xu a, X.W. Liao b, Z.G. Qu c,⇑, Y.Z. Li b, J. Chen a a
School of Energy and Power Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China China Special Equipment Inspection and Research Institute, Beijing 100013, China c MOE Key Laboratory of Thermo-Fluid Science and Engineering, School of Energy and Power Engineering, Xi’an Jiaotong University, Xi’an 710049, China b
h i g h l i g h t s 1:1 scale experimental study of the temperature distribution by a radiant tube. Two burners are investigated for the heat transfer comparisons. The non-uniformity coefficient
a r t i c l e
e is adopted to quantify temperature deviation.
i n f o
Article history: Received 16 May 2016 Revised 9 October 2016 Accepted 15 October 2016 Available online 3 November 2016 Keywords: Experimental study Heating furnace Radiant tube Non-uniformity coefficient Burner
a b s t r a c t This article reports the experimental study of the heat transfer characteristics in a horizontal heating furnace that is extensively used in the petroleum industry of China. To avoid furnace burnout caused by an uneven furnace wall temperature, a straight radiant tube is proposed to be installed in the furnace, which transfers heat to the furnace wall mainly via a radiation pattern. Seven groups of measure points were selected to record the temperature data; each group consisted of two testing points at the top wall and middle wall. Two burners were investigated to compare the temperature distribution along the horizontal furnace wall. A non-uniformity coefficient e of the wall temperature was used to quantify the temperature deviation. The experimental results show that the radiant tube can significantly reduce the temperature deviation on the furnace wall. The furnace wall with the SR100 burner has a generally larger e than that with the HQ05 burner without the radiant tube. The SR100 burner has a larger overall decrease in e than the HQ05 burner after the radiant tube is installed. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Horizontal heating furnaces are widely used in the petroleum industry of China. Because of direct contact with the combustion flame, the internal-wall temperature of the heating furnace is non-uniformly distributed. At a component that has a high heating load, the internal wall can easily be destroyed by combustion [1– 6]. Although sufficient measures, such as material selection, have been considered, failures of the heating furnace cannot be thoroughly avoided and can cause major safety accidents, which are dangerous during the operation in the petroleum industry. In the last decade, some numerical and experimental studies have been performed to inspect the heat and mass transfer behaviors of the heating furnace. Ahanj et al. [7] presented a new methodology to simulate a three-dimensional reheating furnace ⇑ Corresponding author. E-mail address: [email protected] (Z.G. Qu). http://dx.doi.org/10.1016/j.applthermaleng.2016.10.208 1359-4311/Ó 2016 Elsevier Ltd. All rights reserved.
of steel billets with natural gas burners. The energy transport equation was modified to convert the transient movement. The effects of excess air and inlet air on the heater efficiency were also studied. Tsioumanis et al. [8] studied the combustion process in an industrial burner with a radiant tube. The effects of certain radianttube boiler design features were analyzed and the heat transfer processes in the burner were reported. Casal et al. [9] presented a new modelling methodology for a three-dimensional simulation of a reheating furnace with natural-gas burners. The results were consistent with the data obtained in an actual facility, which implies that the proposed methodology was adequate for the simulation of that system. Tu et al. [10] simulated the effects of the furnace chamber shape on the moderate combustion characteristics of natural gas. The results showed that the angle between the furnace roof and sidewalls significantly affected the combustion status. Saario et al. [11] simulated the reacting flow in a heavy fuel oil fired laboratory furnace. They found that the standard k-e model did not satisfactorily predict the highly swirling flow field
2
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
in the furnace. The RSM method was shown to improve the flow field prediction. Irfan and Chapman [12,13] analyzed the thermal stresses in radiant tubes. The analysis was verified using a finiteelement model. The axial temperature gradients were not a source of thermal stresses as long as the temperature distribution was linear. However, spikes in the axial temperature gradient were sources of high thermal stresses. Hu et al. [14] simulated the flow combustion and radiative heat transfer in the furnace. The results showed that the design of the radiation outlet caused asymmetric flue gas temperatures, concentrations and velocity profiles. There were large recirculation zones near the reactor tubes, which made the temperature distribution in the middle of the furnace more uniform. Oliveira et al. [15] conducted a thermal simulation to investigate whether it was possible to reduce fuel consumption and CO2 emission rates by controlling the combustion temperature. Yang et al. [16] investigated the temperature deviation in pulverized-coal boiler furnaces using simulation. They concluded that the nonlinear flow characteristics were key for the velocity and temperature deviations. In experimental studies, Cha et al. [17] investigated the effect of a non-uniform boundary–velocity gradient along the rim of a circular-nozzle burner. They found that the flames in the U-bend tubes had larger velocities than the case with straight tubes. Lou et al. [18] experimentally tested the temperature distribution and radiative properties in an oil-fired tunnel furnace using radiation analysis. The results showed that the temperature was higher in the center and lower near the refractory wall surface and decreased along the length of the tunnel furnace. Liu et al. [19] experimentally investigated the performance of a W-shaped regenerative radiant-tube burner. They found that the tube wall temperature was symmetrically distributed along the length of the tube, with an M-shaped profile and good temperature uniformity. Scribano et al. [20] studied a self-recuperative radiant-tube burner to find the best operating conditions in terms of the optimal equivalence ratio, thermal power and lower pollutant emissions, which resulted in the development of a new burner configuration. Abu-Qudais [21] conducted a test using a cylindrical water-cooled furnace to compare different fuels based on the combustion efficiency, heat transfer to the water jacket and gaseous and particulate emissions in a wide range of air-to-fuel (A/F) ratios at different energy input levels. Pan et al. [22] experimentally studied a cold-state model with a 1:10 scale for a regenerative heating annular furnace to measure the interior velocity distribution. Temperature control of heating furnaces is another key issue. Shi et al. [23] designed a self-adapting PID controller to regulate the temperature of heating furnaces. Dou et al. [24] found that traditional PID control could not effectively overcome the effects of the interference, load change and parameter change of the system and other factors. They improved the controller with a variable slope, which significantly enhanced the robustness of the system. Feng et al. [25] designed a new device that was able to remove most of the sand before it flowed into the heating furnace, which improved the safety and reliability of the heating furnace. The above literature review shows that there are few reported studies regarding the horizontal heating furnace (Fig. 1) in the petroleum industry. According to an investigation of horizontalheating-furnace failure in the Daqing petroleum industry of China, the horizontal furnace wall is often burned because of serious uneven heating (Fig. 2), which introduces safety issue and severely affects the daily production of crude oil. Accordingly, a radiant tube is proposed to be added to the furnace to reduce the temperature deviation on the internal wall by changing the heat transfer mode to the surrounding furnace wall to dominant radiation heat transfer. Prior to the tentative application, detailed experimental investigations of the heat transfer characteristics must be performed in a 1:1 scale of the horizontal heating furnace in Daqing petroleum
Fig. 1. Horizontal heating furnace in the Daqing petroleum industry of China.
industry. This article reports this experimental work and provides valuable suggestions for industrial application. The two most common burners (SR100 (passive) and HQ05 (active)) were used to compare the temperature distributions along the horizontal furnace wall before and after the installation of a radiant tube.
2. Experimental description 2.1. Brief of the experimental system The heating furnace in the experiment was composed of boiler steel. The length was 8000 mm, the inner diameter was 800 mm, and the wall thickness was 4 mm. A schematic diagram of the testing system is shown in Fig. 3. Natural gas was first prepared, and the main component was methane, with a concentration of 94%. The calorific value of the gas was 36.27–40.24 MJ/N m3; the gas pressure was 0.174 MPa. The internal wall temperature of the heating furnace was measured using K-type thermocouples with an inherent accuracy of ±2.5 °C. Seven groups of measuring points were selected along the horizontal direction; each group consisted of two testing points at the top wall and middle wall. A picture of the site testing system is presented in Fig. 4(a). The testing point arrangement on the heating furnace wall is schematically shown in Fig. 5. The temperature was recorded using the data acquisition unit TM-902C, which has an inherent accuracy of ±0.5%. The exhaust gas components NO/NO2 and CO/CO2 were recorded using two gas analyzers, CLD822S and ULTRAMAT23, respectively. Their accuracy was ±1 ppm. The flow rate of the cooling water in the furnace water jacket was measured using an ultrasonic flowmeter, FSC S10C1. The temperature of the inlet and outlet water was measured using Pt100, which has an accuracy of ±0.1%FS. The detailed testing apparatus is summarized in Table 1. Two different burners (Fig. 6) were installed at the left entrance of the horizontal heating furnace. The passive-type burner SR100 introduces natural gas into the heating furnace via the chimney effect. The SR100 burner has three channels: one for natural gas and the other two for air flow. The kinetic energy induced by high-speed natural gas produces a pressure difference, which brings air into the burner. The air volume can be adjusted using the guide wind plate. The active-type burner HQ05 uses a powerful fan to blow natural gas into the combustion chamber. Air is accelerated through the air channel before it reaches a circular disc, which is full of small holes for flame stability. The premixed gas passes through a swirling blade before it enters the burning chamber. The blade is used to generate a swirling flow to increase the combustion efficiency. Table 2 shows the detailed specifications of these two burners.
3
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
Fig. 2. Pictures of the burned furnace wall.
Gas
Gas flowmeter Pressure gauge
Chimney
Pressure relief valve
Shut-off valve
Gas tank
Measurement points
Burner Flue gas
Ball valve Pump
Inlet Water tank
Outlet
Ultrasonic flowmeter Fig. 3. Schematic diagram of the experimental test system.
A radiate tube (Fig. 4(b)) was proposed to be built in the heating furnace to mitigate the temperature deviation along the internal furnace wall. The tube was 4000 mm long and 4 mm thick, with an inner diameter of 500 mm. It was made of SUS304 (06Cr19Ni10) stainless steel with a density of 7.93 kg/m3 and a specific heat of 0.502 kJ/(kg °C) at 20 °C. Its melting point was approximately 1400 °C. First, gas was burnt in the radiate tube, and then, the heated radiant tube transferred heat to the surrounding wall mainly via radiation. In the first stage in 2012, the effects of different parameters (radiant-tube dimension, distance between the tube and the burner, etc.) on the interior combustion of the horizontal heating furnace were experimentally compared and reported in an interior research report. This article only uses the recommended data for the radiant tube and will not discuss their effects. 2.2. Experimental test procedure
Fig. 4. Site pictures of the experiment location and radiant tube (HQ05 burner): (a) experiment site; (b) radiant tube installation.
For experiment comparisons, first, the temperature distributions along the horizontal furnace wall were compared without the radiant tube for these two different burners. Then, the temperature comparisons were analyzed after the radiant tube installation for each burner. Before the test, the apparatus and water loop were checked and the combustible gas detector XP-3140 was used to ensure that there was no gas leakage from the pipe connection. Then, the burner was ignited and the power was slowly adjusted to 400 kW. Ktype thermocouples were also calibrated before the experiment. When the temperature fluctuation was stable for at least thirty
4
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
Fig. 5. Schematic diagram of the testing point arrangement.
Table 1 Apparatus used in the experimental study. No.
Apparatus
Model
Range
Accuracy
1 2 3 4 5 6 7
Thermocouple Data acquisition unit Gas flow meter Pressure gauge Digital thermometer Ultrasonic flowmeter Combustible gas detector
K Type TM-902C IRM-AG650 Y-60 Pt100 FSC S10C1 XP-3140
0–1300 °C 50 to 300 °C 6–1000 m3/h 0–0.16 MPa 50 to 200 °C – 0–100 LEF
±2.5 °C ±0.5%FS 0.5 ±1.5%FS ±0.1%FS Class 1 ±1.5 LEF
Fig. 6. Pictures of two different burners: (a) SR100; (b) HQ05.
Table 2 Specifications of the two burners. No.
Burner
Type
Driven force
Mixing mode
1 2
SR100 HQ05
Passive Active
Chimney effect Supply fan
Partially premixed Premixed
minutes, the data acquisition unit TM-902C was used to record the data for the analysis. The power was adjusted to 700 kW, and the above procedures were repeated.
After the wall temperature was measured, the gas supply pipe was closed and cool air was blown into the furnace for five minutes to ensure that there was no remaining combustible gas in the fur-
5
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
The purpose of the radiant tube is to mitigate the temperature deviation by radiation heat transfer to the furnace wall. To compare the effects of the radiant tube on the temperature distributions along the furnace internal wall, the temperatures were recorded before and after installing the radiant tube for two different burners. Figs. 7 and 8 show seven groups of measured temperatures on the top and middle walls when the SR100 and HQ05 burners were installed, respectively. The effects of the radiant tube on the temperature distributions were compared for each burner. We observed that the internal wall temperature dramatically decreased in the case with the radiant tube. Basically, the temperature first increased and subsequently decreased. At 400 kW, the peak temperatures occurred at point E for the top wall and point F for the middle wall, i.e., the flame center was approximately from point D to point F without the radiant tube. After the radiant tube was installed, the temperature distributions were more uniform 300 Without radiant tube, Top
With radiant tube, Top
Without radiant tube, Middle
With radiant tube, Middle
Temperature (oC)
250
Without radiant tube, Top
With radiant tube, Top
Without radiant tube, Middle
With radiant tube, Middle
200
Temperature (oC)
3. Results and discussion
250
150
100
50
A
B
C
D
E
F
G
Measurement points
(a) 350
Without radiant tube, Top Without radiant tube, Middle
With radiant tube, Top With radiant tube, Middle
300
Temperature (oC)
nace. After the burner and furnace were completely cooled, the burner was removed and the radiant tube was installed using a folk truck. The distance between the radiant tube and furnace entrance was 5 cm. After the apparatus and water loop were checked, the above procedures were repeated.
250
200
150
200
100 150
50
A
B
C
D
E
F
G
Measurement points
100
(b) 50
A
B
C
D
E
F
G
Measurement points
Fig. 8. Temperature comparisons before and after the installation of the radiant tube for the HQ05 burner: (a) 400 kW; (b) 700 kW.
(a) 350
Without radiant tube, Top Without radiant tube, Middle
With radiant tube, Top With radiant tube, Middle
Temperature (oC)
300
250
200
150
100
50
A
B
C
D
E
F
G
Measurement points
(b) Fig. 7. Temperature comparisons before and after the installation of the radiant tube for the SR100 burner: (a) 400 kW; (b) 700 kW.
and the temperature deviation was mitigated. At 700 kW, the overall temperature increased and a similar trend was obtained. The dramatic temperature decrease and uniformity occurred because the dominant heat transfer pattern to the furnace wall was radiation by the radiant tube and the combustion flame could not directly contact the furnace wall. The temperature distributions of two different burners after the radiant tube was installed are compared in Fig. 9. We observed that the SR100 burner generated a higher temperature than the HQ05 burner at 400 kW because the flame rigidity from HQ05 was stronger than that from SR100. At point E, the top-wall temperature from the SR100 burner was 4.0% higher than that from the HQ05 burner. At point F, the middle-wall temperature from SR100 burner was 7.3% higher than that from the HQ05 burner. The overall temperature difference was less than 10%. At 700 kW, the HQ05 burner provided a slightly higher temperature than the SR100 burner, which is different from the result at 400 kW. A possible reason for this difference is that HQ05 is a premixed combustion burner and the passive partially premixed SR100 burner cannot bring sufficient air when the power increases. A detailed inspection of the internal combustion characteristics and distributions by these two different burners should be performed using the
6
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
150
SR100, Top SR100, Middle
Table 4 CO and NOX emission comparisons for the two different burners after installing the radiant tube.
HQ05, Top HQ05, Middle
Power (kW)
Burner
Exhaust temperature (°C)
CO
NOX
400
SR100 HQ05
471.1 474.4
64 22.5
42.8 47
700
SR100 HQ05
653.3 583.28
59.4 3.5
49.2 49.5
Temperature (oC)
125
100
Concentration (ppm)
75
50
A
B
C
D
E
F
G
F
G
Measurement points
(a) 180
SR100, Top SR100, Middle
HQ05, Top HQ05, Middle
Temperature (oC)
160
140
120
100
80
60
A
B
C
D
E
Measurement points
(b) Fig. 9. Temperature comparisons between two burners after the installation of the radiant tube: (a) 400 kW; (b) 700 kW.
Table 3 Non-uniformity temperature difference coefficients before and after installing the radiant tube. Burner
Power (kW)
Radiant tube
e (top)
Difference
e (middle)
Difference
SR100 SR100 SR100 SR100 HQ05 HQ05 HQ05 HQ05
400 400 700 700 400 400 700 700
Without With Without With Without With Without With
0.681 0.378 0.686 0.417 0.624 0.406 0.636 0.427
44.5%;
0.637 0.473 0.662 0.494 0.633 0.532 0.627 0.505
25.7%;
39.2%; 34.9%; 32.9%;
them; e is an index of the temperature deviation. A larger e indicates a larger temperature deviation. Table 3 shows the non-uniformity coefficients of the wall temperature distribution from points A to G. In Table 3, the furnace wall with the SR100 burner had a generally larger e than that with the HQ05 burner without the radiant tube. For example, the e of the top wall was 0.681 for the SR100 burner and 0.624 for the HQ05 burner at 400 kW. Without the radiant tube, the overall non-uniformity coefficient of the top- and middle-wall temperature was approximately 0.627–0.686. After installation of the radiant tube, the e of the furnace wall with the SR100 burner was generally smaller than that with the HQ05 burner. We observed that the non-uniformity coefficient significantly decreased when the radiant tube was used. The e of the top wall decreased from 44.5% to 39.2% and 34.9% to 32.9% when the combustion power increased from 400 kW to 700 kW for the SR100 and HQ05 burners, respectively. Because hot flue gas flowed upward after combustion, the decrease in the e of the middle wall was much less than that of the top wall and in the range of 16.0–25.7%. The results also indicate that the SR100 burner caused a larger overall decrease of the non-uniformity coefficient e than the HQ05 burner after the radiant tube was installed. During the experiment, emissions of CO and NOx were also analyzed after radiant-tube installation. Table 4 shows the results for the SR100 and HQ05 burners at different combustion powers. The CO concentration from the HQ05 burner was significantly lower than that from the SR100 burner. Thus, the premixed burner HR05 has a higher combustion efficiency than the partially premixed burner SR100, particularly at a larger combustion power (700 kW). The emitted NOX concentrations were approximately identical. The experiment also shows that the exhaust gas temperature increases, which decreases the overall furnace efficiency. The trade-off between less burnout (save cost) because of a relatively uniform temperature distribution and efficiency decrease (waste energy) will be further analyzed by collecting detailed operation data in the Daqing petroleum industry in the future.
25.4%;
4. Conclusions 16.0%; 19.5%;
Computational Fluid Dynamics (CFD) technology at different combustion powers. At point E, the middle-wall temperature difference was approximately 8.2%. To quantify the temperature deviation from point A to point G, a non-uniformity coefficient e of the wall temperature distribution was defined as: e ¼ Dt=t mhboxmax ¼ 1 t min =t mhboxmax , where t min and tmhboxmax denote the minimum and maximum wall temperatures, respectively; Dt is the temperature difference between
This article presented the experimental results of the proposed design concept of a radiant tube installed in a horizontal heating furnace. The non-uniformity coefficient e of the wall temperature distribution was proposed to quantify the relative temperature deviation along the horizontal furnace wall. Regardless of the type of burner, the overall temperature decreases, and the temperature distribution is more uniform after a radiant tube is installed in the heating furnace because heat is mainly transferred to the furnace wall by radiation from the radiant tube. Compared with the premixed burner HQ05, the SR100 burner is partially premixed and cannot bring sufficient air when the power increases. Thus, the SR100 burner has a lower furnace wall temperature than the HQ05 burner at 700 kW. However, the overall temperature
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
difference between the two burners is less than 10% after a radiant tube is installed. With a radiant tube, the e of the top-wall temperature difference significantly decreases from 44.5% to 39.2% and 34.9% to 32.9% for the SR100 and HQ05 burners, respectively, when the combustion power increases from 400 kW to 700 kW. Compared with the top wall, the middle-wall temperature difference has a smaller e than the middle wall after the installation of a radiant tube because the hot flue gas flows upward in the radiant tube. This study indicates the possibility of using a radiant tube in a horizontal heating furnace. The radiant tube can reduce the burnout frequency of the heating furnace to save cost and avoid safety accidents during petroleum industry operations. Acknowledgements This work was financially supported by the National Natural Science Foundation of China (No. 51276117) and National Program for Support of Top-notch Young Professionals. The authors thank China Special Equipment Inspection and Research Institute for providing a place for the experiment testing. References [1] G. Dini, S.M. Monir Vaghefi, M. Lotfiani, M. Jafari, M. Safaei-Rad, M. Navabi, S. Abbasi, Computational and experimental failure analysis of continuousannealing furnace radiant tubes exposed to excessive temperature, Eng. Fail. Anal. 15 (2008) 445–457. [2] V. Mertinger, M. Benke, S. Szabó, O. Bánhidi, B. Bollo, Á. Kovács, Examination of a failure detected in the convection zone of a cracking furnace, Eng. Fail. Anal. 18 (2011) 1675–1682. [3] Z.C. Zhu, C.Q. Cheng, J. Zhao, L. Wang, High temperature corrosion and microstructure deterioration of KHR35H radiant tubes in continuous annealing furnace, Eng. Fail. Anal. 21 (2012) 59–66. [4] H.M. Shalaby, N. Al-Sebaii, W.T. Riad, P.K. Mukhopadhyay, Cracking of radiant tube of heater coil carring heavy oil in a hydrocracking unit, Eng. Fail. Anal. 31 (2013) 281–289. [5] W.H. Wang, K.W. Liang, C.Y. Wang, Q.S. Wang, Comparative analysis of failure probability for ethylene cracking furnace tube using Monte Carlo and API RBI technology, Eng. Fail. Anal. 45 (2014) 278–282. [6] M. Santos, M. Guedes, R. Baptista, V. Infante, R.A. Cláudio, Effect of severe operation conditions on the degradation state of radiant coils in pyrolysis furnaces, Eng. Fail. Anal. 56 (2015) 194–203. [7] M.D. Ahanj, M. Rahimi, A.A. Alsairafi, CFD modeling of a radiant tube heater, Int. Commun. Heat Mass 39 (2012) 432–438.
7
[8] N. Tsioumanis, J.G. Brammer, J. Hubert, Flow processes in a radiant tube burner: combusting flow, Energy Convers. Manage. 52 (2011) 2667–2675. [9] J.M. Casal, J. Porteiro, J.L. Míguez, A. Vázquez, New methodology for CFD threedimensional simulation of a walking beam type reheating furnace in steady state, Appl. Therm. Eng. 86 (2015) 69–80. [10] Y.J. Tu, H. Liu, S. Chen, Z.H. Liu, H.B. Zhao, C.G. Zheng, Effects of furnace chamber shape on the MILD combustion of natural gas, Appl. Therm. Eng. 76 (2015) 64–75. [11] A. Saario, A. Rebola, P.J. Coelho, M. Costa, A. Oksanen, Heavy fuel oil combustion in a cylindrical laboratory furnace: measurements and modeling, Fuel 84 (2005) 359–369. [12] M.A. Irfan, W. Chapman, Thermal stresses in radiant tubes due to axial, circumferential and radial temperature distributions, Appl. Therm. Eng. 29 (2009) 1913–1920. [13] M.A. Irfan, W. Chapman, Thermal stresses in radiant tubes: a comparison between recuperative and regenerative system, Appl. Therm. Eng. 30 (2010) 196–200. [14] G.H. Hu, H.G. Wang, F. Qian, K.M. Van Geem, C.M. Schietekat, G.B. Marin, Coupled simulation of an industrial naphtha cracking furnace equipped with long-flame and radiation burners, Comput. Chem. Eng. 38 (2012) 24–34. [15] F.A.D. Oliveira, J.A. Carvalho, P.M. Sobrinho, A.D. Castro, Analysis of oxy-fuel combustion as an alternative to combustion with air in metal reheating furnaces, Energy 78 (2014) 290–297. [16] M. Yang, Y.Y. Shen, H.T. Xu, M. Zhao, S.W. Shen, K. Huang, Numerical investigation of the nonlinear flow characteristics in an ultra-supercritical utility boiler furnace, Appl. Therm. Eng. 88 (2015) 237–247. [17] M.S. Cha, H.G. Kim, S.H. Chung, Boundary-velocity gradient and premixed flame blowoff in U-bend tubes with secondary flow, Combust. Flame 132 (2003) 601–609. [18] C. Lou, W.H. Li, H.C. Zhou, C.T. Salinas, Experimental investigation on simultaneous measurement of temperature distributions and radiative properties in an oil-fired tunnel furnace by radiation analysis, Int. J. Heat Mass Transf. 54 (2011) 1–8. [19] X.L. Liu, Y. Tian, Y. Yu, Z. Wen, D.M. Zhang, Z. Li, X.H. Feng, Experimental studies on the heating performance and emission characteristics of a Wshaped regenerative radiant tube burner, Fuel 135 (2014) 262–268. [20] G. Scribano, G. Solero, A. Coghe, Pollutant emissions reduction and performance optimization of an industrial radiant tube burner, Exp. Therm. Fluid Sci. 30 (2006) 605–612. [21] M. Abu-Qudais, Performance and emissions characteristics of a cylindrical water cooled furnace using non-petroleum fuel, Energy Convers. Manage. 43 (2002) 683–691. [22] L.M. Pan, H.C. Ji, S.M. Cheng, C.B. Wu, H.Q. Yong, An experimental investigation for cold-state flow field of regenerative heating annular furnace, Appl. Therm. Eng. 29 (2009) 3426–3430. [23] D.Q. Shi, G.L. Gao, Z.W. Gao, P. Xiao, Application of expert fuzzy PID method for temperature control of heating furnace, Proc. Eng. 29 (2012) 257–261. [24] Z.H. Dou, L.Y. Sun, Design of temperature controller for heating furnace in oil field, Phys. Proc. 24 (Part C) (2012) 2083–2088. [25] F. Ding, R. Chao, X.H. Liu, P. Wang, H. Zhang, The research of heating furnace with heavy oil desander, Energy Proc. 14 (2012) 674–680.
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Research Paper
Experimental study of the effect of a radiant tube on the temperature distribution in a horizontal heating furnace H.T. Xu a, X.W. Liao b, Z.G. Qu c,⇑, Y.Z. Li b, J. Chen a a
School of Energy and Power Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China China Special Equipment Inspection and Research Institute, Beijing 100013, China c MOE Key Laboratory of Thermo-Fluid Science and Engineering, School of Energy and Power Engineering, Xi’an Jiaotong University, Xi’an 710049, China b
h i g h l i g h t s 1:1 scale experimental study of the temperature distribution by a radiant tube. Two burners are investigated for the heat transfer comparisons. The non-uniformity coefficient
a r t i c l e
e is adopted to quantify temperature deviation.
i n f o
Article history: Received 16 May 2016 Revised 9 October 2016 Accepted 15 October 2016 Available online 3 November 2016 Keywords: Experimental study Heating furnace Radiant tube Non-uniformity coefficient Burner
a b s t r a c t This article reports the experimental study of the heat transfer characteristics in a horizontal heating furnace that is extensively used in the petroleum industry of China. To avoid furnace burnout caused by an uneven furnace wall temperature, a straight radiant tube is proposed to be installed in the furnace, which transfers heat to the furnace wall mainly via a radiation pattern. Seven groups of measure points were selected to record the temperature data; each group consisted of two testing points at the top wall and middle wall. Two burners were investigated to compare the temperature distribution along the horizontal furnace wall. A non-uniformity coefficient e of the wall temperature was used to quantify the temperature deviation. The experimental results show that the radiant tube can significantly reduce the temperature deviation on the furnace wall. The furnace wall with the SR100 burner has a generally larger e than that with the HQ05 burner without the radiant tube. The SR100 burner has a larger overall decrease in e than the HQ05 burner after the radiant tube is installed. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Horizontal heating furnaces are widely used in the petroleum industry of China. Because of direct contact with the combustion flame, the internal-wall temperature of the heating furnace is non-uniformly distributed. At a component that has a high heating load, the internal wall can easily be destroyed by combustion [1– 6]. Although sufficient measures, such as material selection, have been considered, failures of the heating furnace cannot be thoroughly avoided and can cause major safety accidents, which are dangerous during the operation in the petroleum industry. In the last decade, some numerical and experimental studies have been performed to inspect the heat and mass transfer behaviors of the heating furnace. Ahanj et al. [7] presented a new methodology to simulate a three-dimensional reheating furnace ⇑ Corresponding author. E-mail address: [email protected] (Z.G. Qu). http://dx.doi.org/10.1016/j.applthermaleng.2016.10.208 1359-4311/Ó 2016 Elsevier Ltd. All rights reserved.
of steel billets with natural gas burners. The energy transport equation was modified to convert the transient movement. The effects of excess air and inlet air on the heater efficiency were also studied. Tsioumanis et al. [8] studied the combustion process in an industrial burner with a radiant tube. The effects of certain radianttube boiler design features were analyzed and the heat transfer processes in the burner were reported. Casal et al. [9] presented a new modelling methodology for a three-dimensional simulation of a reheating furnace with natural-gas burners. The results were consistent with the data obtained in an actual facility, which implies that the proposed methodology was adequate for the simulation of that system. Tu et al. [10] simulated the effects of the furnace chamber shape on the moderate combustion characteristics of natural gas. The results showed that the angle between the furnace roof and sidewalls significantly affected the combustion status. Saario et al. [11] simulated the reacting flow in a heavy fuel oil fired laboratory furnace. They found that the standard k-e model did not satisfactorily predict the highly swirling flow field
2
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
in the furnace. The RSM method was shown to improve the flow field prediction. Irfan and Chapman [12,13] analyzed the thermal stresses in radiant tubes. The analysis was verified using a finiteelement model. The axial temperature gradients were not a source of thermal stresses as long as the temperature distribution was linear. However, spikes in the axial temperature gradient were sources of high thermal stresses. Hu et al. [14] simulated the flow combustion and radiative heat transfer in the furnace. The results showed that the design of the radiation outlet caused asymmetric flue gas temperatures, concentrations and velocity profiles. There were large recirculation zones near the reactor tubes, which made the temperature distribution in the middle of the furnace more uniform. Oliveira et al. [15] conducted a thermal simulation to investigate whether it was possible to reduce fuel consumption and CO2 emission rates by controlling the combustion temperature. Yang et al. [16] investigated the temperature deviation in pulverized-coal boiler furnaces using simulation. They concluded that the nonlinear flow characteristics were key for the velocity and temperature deviations. In experimental studies, Cha et al. [17] investigated the effect of a non-uniform boundary–velocity gradient along the rim of a circular-nozzle burner. They found that the flames in the U-bend tubes had larger velocities than the case with straight tubes. Lou et al. [18] experimentally tested the temperature distribution and radiative properties in an oil-fired tunnel furnace using radiation analysis. The results showed that the temperature was higher in the center and lower near the refractory wall surface and decreased along the length of the tunnel furnace. Liu et al. [19] experimentally investigated the performance of a W-shaped regenerative radiant-tube burner. They found that the tube wall temperature was symmetrically distributed along the length of the tube, with an M-shaped profile and good temperature uniformity. Scribano et al. [20] studied a self-recuperative radiant-tube burner to find the best operating conditions in terms of the optimal equivalence ratio, thermal power and lower pollutant emissions, which resulted in the development of a new burner configuration. Abu-Qudais [21] conducted a test using a cylindrical water-cooled furnace to compare different fuels based on the combustion efficiency, heat transfer to the water jacket and gaseous and particulate emissions in a wide range of air-to-fuel (A/F) ratios at different energy input levels. Pan et al. [22] experimentally studied a cold-state model with a 1:10 scale for a regenerative heating annular furnace to measure the interior velocity distribution. Temperature control of heating furnaces is another key issue. Shi et al. [23] designed a self-adapting PID controller to regulate the temperature of heating furnaces. Dou et al. [24] found that traditional PID control could not effectively overcome the effects of the interference, load change and parameter change of the system and other factors. They improved the controller with a variable slope, which significantly enhanced the robustness of the system. Feng et al. [25] designed a new device that was able to remove most of the sand before it flowed into the heating furnace, which improved the safety and reliability of the heating furnace. The above literature review shows that there are few reported studies regarding the horizontal heating furnace (Fig. 1) in the petroleum industry. According to an investigation of horizontalheating-furnace failure in the Daqing petroleum industry of China, the horizontal furnace wall is often burned because of serious uneven heating (Fig. 2), which introduces safety issue and severely affects the daily production of crude oil. Accordingly, a radiant tube is proposed to be added to the furnace to reduce the temperature deviation on the internal wall by changing the heat transfer mode to the surrounding furnace wall to dominant radiation heat transfer. Prior to the tentative application, detailed experimental investigations of the heat transfer characteristics must be performed in a 1:1 scale of the horizontal heating furnace in Daqing petroleum
Fig. 1. Horizontal heating furnace in the Daqing petroleum industry of China.
industry. This article reports this experimental work and provides valuable suggestions for industrial application. The two most common burners (SR100 (passive) and HQ05 (active)) were used to compare the temperature distributions along the horizontal furnace wall before and after the installation of a radiant tube.
2. Experimental description 2.1. Brief of the experimental system The heating furnace in the experiment was composed of boiler steel. The length was 8000 mm, the inner diameter was 800 mm, and the wall thickness was 4 mm. A schematic diagram of the testing system is shown in Fig. 3. Natural gas was first prepared, and the main component was methane, with a concentration of 94%. The calorific value of the gas was 36.27–40.24 MJ/N m3; the gas pressure was 0.174 MPa. The internal wall temperature of the heating furnace was measured using K-type thermocouples with an inherent accuracy of ±2.5 °C. Seven groups of measuring points were selected along the horizontal direction; each group consisted of two testing points at the top wall and middle wall. A picture of the site testing system is presented in Fig. 4(a). The testing point arrangement on the heating furnace wall is schematically shown in Fig. 5. The temperature was recorded using the data acquisition unit TM-902C, which has an inherent accuracy of ±0.5%. The exhaust gas components NO/NO2 and CO/CO2 were recorded using two gas analyzers, CLD822S and ULTRAMAT23, respectively. Their accuracy was ±1 ppm. The flow rate of the cooling water in the furnace water jacket was measured using an ultrasonic flowmeter, FSC S10C1. The temperature of the inlet and outlet water was measured using Pt100, which has an accuracy of ±0.1%FS. The detailed testing apparatus is summarized in Table 1. Two different burners (Fig. 6) were installed at the left entrance of the horizontal heating furnace. The passive-type burner SR100 introduces natural gas into the heating furnace via the chimney effect. The SR100 burner has three channels: one for natural gas and the other two for air flow. The kinetic energy induced by high-speed natural gas produces a pressure difference, which brings air into the burner. The air volume can be adjusted using the guide wind plate. The active-type burner HQ05 uses a powerful fan to blow natural gas into the combustion chamber. Air is accelerated through the air channel before it reaches a circular disc, which is full of small holes for flame stability. The premixed gas passes through a swirling blade before it enters the burning chamber. The blade is used to generate a swirling flow to increase the combustion efficiency. Table 2 shows the detailed specifications of these two burners.
3
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
Fig. 2. Pictures of the burned furnace wall.
Gas
Gas flowmeter Pressure gauge
Chimney
Pressure relief valve
Shut-off valve
Gas tank
Measurement points
Burner Flue gas
Ball valve Pump
Inlet Water tank
Outlet
Ultrasonic flowmeter Fig. 3. Schematic diagram of the experimental test system.
A radiate tube (Fig. 4(b)) was proposed to be built in the heating furnace to mitigate the temperature deviation along the internal furnace wall. The tube was 4000 mm long and 4 mm thick, with an inner diameter of 500 mm. It was made of SUS304 (06Cr19Ni10) stainless steel with a density of 7.93 kg/m3 and a specific heat of 0.502 kJ/(kg °C) at 20 °C. Its melting point was approximately 1400 °C. First, gas was burnt in the radiate tube, and then, the heated radiant tube transferred heat to the surrounding wall mainly via radiation. In the first stage in 2012, the effects of different parameters (radiant-tube dimension, distance between the tube and the burner, etc.) on the interior combustion of the horizontal heating furnace were experimentally compared and reported in an interior research report. This article only uses the recommended data for the radiant tube and will not discuss their effects. 2.2. Experimental test procedure
Fig. 4. Site pictures of the experiment location and radiant tube (HQ05 burner): (a) experiment site; (b) radiant tube installation.
For experiment comparisons, first, the temperature distributions along the horizontal furnace wall were compared without the radiant tube for these two different burners. Then, the temperature comparisons were analyzed after the radiant tube installation for each burner. Before the test, the apparatus and water loop were checked and the combustible gas detector XP-3140 was used to ensure that there was no gas leakage from the pipe connection. Then, the burner was ignited and the power was slowly adjusted to 400 kW. Ktype thermocouples were also calibrated before the experiment. When the temperature fluctuation was stable for at least thirty
4
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
Fig. 5. Schematic diagram of the testing point arrangement.
Table 1 Apparatus used in the experimental study. No.
Apparatus
Model
Range
Accuracy
1 2 3 4 5 6 7
Thermocouple Data acquisition unit Gas flow meter Pressure gauge Digital thermometer Ultrasonic flowmeter Combustible gas detector
K Type TM-902C IRM-AG650 Y-60 Pt100 FSC S10C1 XP-3140
0–1300 °C 50 to 300 °C 6–1000 m3/h 0–0.16 MPa 50 to 200 °C – 0–100 LEF
±2.5 °C ±0.5%FS 0.5 ±1.5%FS ±0.1%FS Class 1 ±1.5 LEF
Fig. 6. Pictures of two different burners: (a) SR100; (b) HQ05.
Table 2 Specifications of the two burners. No.
Burner
Type
Driven force
Mixing mode
1 2
SR100 HQ05
Passive Active
Chimney effect Supply fan
Partially premixed Premixed
minutes, the data acquisition unit TM-902C was used to record the data for the analysis. The power was adjusted to 700 kW, and the above procedures were repeated.
After the wall temperature was measured, the gas supply pipe was closed and cool air was blown into the furnace for five minutes to ensure that there was no remaining combustible gas in the fur-
5
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
The purpose of the radiant tube is to mitigate the temperature deviation by radiation heat transfer to the furnace wall. To compare the effects of the radiant tube on the temperature distributions along the furnace internal wall, the temperatures were recorded before and after installing the radiant tube for two different burners. Figs. 7 and 8 show seven groups of measured temperatures on the top and middle walls when the SR100 and HQ05 burners were installed, respectively. The effects of the radiant tube on the temperature distributions were compared for each burner. We observed that the internal wall temperature dramatically decreased in the case with the radiant tube. Basically, the temperature first increased and subsequently decreased. At 400 kW, the peak temperatures occurred at point E for the top wall and point F for the middle wall, i.e., the flame center was approximately from point D to point F without the radiant tube. After the radiant tube was installed, the temperature distributions were more uniform 300 Without radiant tube, Top
With radiant tube, Top
Without radiant tube, Middle
With radiant tube, Middle
Temperature (oC)
250
Without radiant tube, Top
With radiant tube, Top
Without radiant tube, Middle
With radiant tube, Middle
200
Temperature (oC)
3. Results and discussion
250
150
100
50
A
B
C
D
E
F
G
Measurement points
(a) 350
Without radiant tube, Top Without radiant tube, Middle
With radiant tube, Top With radiant tube, Middle
300
Temperature (oC)
nace. After the burner and furnace were completely cooled, the burner was removed and the radiant tube was installed using a folk truck. The distance between the radiant tube and furnace entrance was 5 cm. After the apparatus and water loop were checked, the above procedures were repeated.
250
200
150
200
100 150
50
A
B
C
D
E
F
G
Measurement points
100
(b) 50
A
B
C
D
E
F
G
Measurement points
Fig. 8. Temperature comparisons before and after the installation of the radiant tube for the HQ05 burner: (a) 400 kW; (b) 700 kW.
(a) 350
Without radiant tube, Top Without radiant tube, Middle
With radiant tube, Top With radiant tube, Middle
Temperature (oC)
300
250
200
150
100
50
A
B
C
D
E
F
G
Measurement points
(b) Fig. 7. Temperature comparisons before and after the installation of the radiant tube for the SR100 burner: (a) 400 kW; (b) 700 kW.
and the temperature deviation was mitigated. At 700 kW, the overall temperature increased and a similar trend was obtained. The dramatic temperature decrease and uniformity occurred because the dominant heat transfer pattern to the furnace wall was radiation by the radiant tube and the combustion flame could not directly contact the furnace wall. The temperature distributions of two different burners after the radiant tube was installed are compared in Fig. 9. We observed that the SR100 burner generated a higher temperature than the HQ05 burner at 400 kW because the flame rigidity from HQ05 was stronger than that from SR100. At point E, the top-wall temperature from the SR100 burner was 4.0% higher than that from the HQ05 burner. At point F, the middle-wall temperature from SR100 burner was 7.3% higher than that from the HQ05 burner. The overall temperature difference was less than 10%. At 700 kW, the HQ05 burner provided a slightly higher temperature than the SR100 burner, which is different from the result at 400 kW. A possible reason for this difference is that HQ05 is a premixed combustion burner and the passive partially premixed SR100 burner cannot bring sufficient air when the power increases. A detailed inspection of the internal combustion characteristics and distributions by these two different burners should be performed using the
6
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
150
SR100, Top SR100, Middle
Table 4 CO and NOX emission comparisons for the two different burners after installing the radiant tube.
HQ05, Top HQ05, Middle
Power (kW)
Burner
Exhaust temperature (°C)
CO
NOX
400
SR100 HQ05
471.1 474.4
64 22.5
42.8 47
700
SR100 HQ05
653.3 583.28
59.4 3.5
49.2 49.5
Temperature (oC)
125
100
Concentration (ppm)
75
50
A
B
C
D
E
F
G
F
G
Measurement points
(a) 180
SR100, Top SR100, Middle
HQ05, Top HQ05, Middle
Temperature (oC)
160
140
120
100
80
60
A
B
C
D
E
Measurement points
(b) Fig. 9. Temperature comparisons between two burners after the installation of the radiant tube: (a) 400 kW; (b) 700 kW.
Table 3 Non-uniformity temperature difference coefficients before and after installing the radiant tube. Burner
Power (kW)
Radiant tube
e (top)
Difference
e (middle)
Difference
SR100 SR100 SR100 SR100 HQ05 HQ05 HQ05 HQ05
400 400 700 700 400 400 700 700
Without With Without With Without With Without With
0.681 0.378 0.686 0.417 0.624 0.406 0.636 0.427
44.5%;
0.637 0.473 0.662 0.494 0.633 0.532 0.627 0.505
25.7%;
39.2%; 34.9%; 32.9%;
them; e is an index of the temperature deviation. A larger e indicates a larger temperature deviation. Table 3 shows the non-uniformity coefficients of the wall temperature distribution from points A to G. In Table 3, the furnace wall with the SR100 burner had a generally larger e than that with the HQ05 burner without the radiant tube. For example, the e of the top wall was 0.681 for the SR100 burner and 0.624 for the HQ05 burner at 400 kW. Without the radiant tube, the overall non-uniformity coefficient of the top- and middle-wall temperature was approximately 0.627–0.686. After installation of the radiant tube, the e of the furnace wall with the SR100 burner was generally smaller than that with the HQ05 burner. We observed that the non-uniformity coefficient significantly decreased when the radiant tube was used. The e of the top wall decreased from 44.5% to 39.2% and 34.9% to 32.9% when the combustion power increased from 400 kW to 700 kW for the SR100 and HQ05 burners, respectively. Because hot flue gas flowed upward after combustion, the decrease in the e of the middle wall was much less than that of the top wall and in the range of 16.0–25.7%. The results also indicate that the SR100 burner caused a larger overall decrease of the non-uniformity coefficient e than the HQ05 burner after the radiant tube was installed. During the experiment, emissions of CO and NOx were also analyzed after radiant-tube installation. Table 4 shows the results for the SR100 and HQ05 burners at different combustion powers. The CO concentration from the HQ05 burner was significantly lower than that from the SR100 burner. Thus, the premixed burner HR05 has a higher combustion efficiency than the partially premixed burner SR100, particularly at a larger combustion power (700 kW). The emitted NOX concentrations were approximately identical. The experiment also shows that the exhaust gas temperature increases, which decreases the overall furnace efficiency. The trade-off between less burnout (save cost) because of a relatively uniform temperature distribution and efficiency decrease (waste energy) will be further analyzed by collecting detailed operation data in the Daqing petroleum industry in the future.
25.4%;
4. Conclusions 16.0%; 19.5%;
Computational Fluid Dynamics (CFD) technology at different combustion powers. At point E, the middle-wall temperature difference was approximately 8.2%. To quantify the temperature deviation from point A to point G, a non-uniformity coefficient e of the wall temperature distribution was defined as: e ¼ Dt=t mhboxmax ¼ 1 t min =t mhboxmax , where t min and tmhboxmax denote the minimum and maximum wall temperatures, respectively; Dt is the temperature difference between
This article presented the experimental results of the proposed design concept of a radiant tube installed in a horizontal heating furnace. The non-uniformity coefficient e of the wall temperature distribution was proposed to quantify the relative temperature deviation along the horizontal furnace wall. Regardless of the type of burner, the overall temperature decreases, and the temperature distribution is more uniform after a radiant tube is installed in the heating furnace because heat is mainly transferred to the furnace wall by radiation from the radiant tube. Compared with the premixed burner HQ05, the SR100 burner is partially premixed and cannot bring sufficient air when the power increases. Thus, the SR100 burner has a lower furnace wall temperature than the HQ05 burner at 700 kW. However, the overall temperature
H.T. Xu et al. / Applied Thermal Engineering 113 (2017) 1–7
difference between the two burners is less than 10% after a radiant tube is installed. With a radiant tube, the e of the top-wall temperature difference significantly decreases from 44.5% to 39.2% and 34.9% to 32.9% for the SR100 and HQ05 burners, respectively, when the combustion power increases from 400 kW to 700 kW. Compared with the top wall, the middle-wall temperature difference has a smaller e than the middle wall after the installation of a radiant tube because the hot flue gas flows upward in the radiant tube. This study indicates the possibility of using a radiant tube in a horizontal heating furnace. The radiant tube can reduce the burnout frequency of the heating furnace to save cost and avoid safety accidents during petroleum industry operations. Acknowledgements This work was financially supported by the National Natural Science Foundation of China (No. 51276117) and National Program for Support of Top-notch Young Professionals. The authors thank China Special Equipment Inspection and Research Institute for providing a place for the experiment testing. References [1] G. Dini, S.M. Monir Vaghefi, M. Lotfiani, M. Jafari, M. Safaei-Rad, M. Navabi, S. Abbasi, Computational and experimental failure analysis of continuousannealing furnace radiant tubes exposed to excessive temperature, Eng. Fail. Anal. 15 (2008) 445–457. [2] V. Mertinger, M. Benke, S. Szabó, O. Bánhidi, B. Bollo, Á. Kovács, Examination of a failure detected in the convection zone of a cracking furnace, Eng. Fail. Anal. 18 (2011) 1675–1682. [3] Z.C. Zhu, C.Q. Cheng, J. Zhao, L. Wang, High temperature corrosion and microstructure deterioration of KHR35H radiant tubes in continuous annealing furnace, Eng. Fail. Anal. 21 (2012) 59–66. [4] H.M. Shalaby, N. Al-Sebaii, W.T. Riad, P.K. Mukhopadhyay, Cracking of radiant tube of heater coil carring heavy oil in a hydrocracking unit, Eng. Fail. Anal. 31 (2013) 281–289. [5] W.H. Wang, K.W. Liang, C.Y. Wang, Q.S. Wang, Comparative analysis of failure probability for ethylene cracking furnace tube using Monte Carlo and API RBI technology, Eng. Fail. Anal. 45 (2014) 278–282. [6] M. Santos, M. Guedes, R. Baptista, V. Infante, R.A. Cláudio, Effect of severe operation conditions on the degradation state of radiant coils in pyrolysis furnaces, Eng. Fail. Anal. 56 (2015) 194–203. [7] M.D. Ahanj, M. Rahimi, A.A. Alsairafi, CFD modeling of a radiant tube heater, Int. Commun. Heat Mass 39 (2012) 432–438.
7
[8] N. Tsioumanis, J.G. Brammer, J. Hubert, Flow processes in a radiant tube burner: combusting flow, Energy Convers. Manage. 52 (2011) 2667–2675. [9] J.M. Casal, J. Porteiro, J.L. Míguez, A. Vázquez, New methodology for CFD threedimensional simulation of a walking beam type reheating furnace in steady state, Appl. Therm. Eng. 86 (2015) 69–80. [10] Y.J. Tu, H. Liu, S. Chen, Z.H. Liu, H.B. Zhao, C.G. Zheng, Effects of furnace chamber shape on the MILD combustion of natural gas, Appl. Therm. Eng. 76 (2015) 64–75. [11] A. Saario, A. Rebola, P.J. Coelho, M. Costa, A. Oksanen, Heavy fuel oil combustion in a cylindrical laboratory furnace: measurements and modeling, Fuel 84 (2005) 359–369. [12] M.A. Irfan, W. Chapman, Thermal stresses in radiant tubes due to axial, circumferential and radial temperature distributions, Appl. Therm. Eng. 29 (2009) 1913–1920. [13] M.A. Irfan, W. Chapman, Thermal stresses in radiant tubes: a comparison between recuperative and regenerative system, Appl. Therm. Eng. 30 (2010) 196–200. [14] G.H. Hu, H.G. Wang, F. Qian, K.M. Van Geem, C.M. Schietekat, G.B. Marin, Coupled simulation of an industrial naphtha cracking furnace equipped with long-flame and radiation burners, Comput. Chem. Eng. 38 (2012) 24–34. [15] F.A.D. Oliveira, J.A. Carvalho, P.M. Sobrinho, A.D. Castro, Analysis of oxy-fuel combustion as an alternative to combustion with air in metal reheating furnaces, Energy 78 (2014) 290–297. [16] M. Yang, Y.Y. Shen, H.T. Xu, M. Zhao, S.W. Shen, K. Huang, Numerical investigation of the nonlinear flow characteristics in an ultra-supercritical utility boiler furnace, Appl. Therm. Eng. 88 (2015) 237–247. [17] M.S. Cha, H.G. Kim, S.H. Chung, Boundary-velocity gradient and premixed flame blowoff in U-bend tubes with secondary flow, Combust. Flame 132 (2003) 601–609. [18] C. Lou, W.H. Li, H.C. Zhou, C.T. Salinas, Experimental investigation on simultaneous measurement of temperature distributions and radiative properties in an oil-fired tunnel furnace by radiation analysis, Int. J. Heat Mass Transf. 54 (2011) 1–8. [19] X.L. Liu, Y. Tian, Y. Yu, Z. Wen, D.M. Zhang, Z. Li, X.H. Feng, Experimental studies on the heating performance and emission characteristics of a Wshaped regenerative radiant tube burner, Fuel 135 (2014) 262–268. [20] G. Scribano, G. Solero, A. Coghe, Pollutant emissions reduction and performance optimization of an industrial radiant tube burner, Exp. Therm. Fluid Sci. 30 (2006) 605–612. [21] M. Abu-Qudais, Performance and emissions characteristics of a cylindrical water cooled furnace using non-petroleum fuel, Energy Convers. Manage. 43 (2002) 683–691. [22] L.M. Pan, H.C. Ji, S.M. Cheng, C.B. Wu, H.Q. Yong, An experimental investigation for cold-state flow field of regenerative heating annular furnace, Appl. Therm. Eng. 29 (2009) 3426–3430. [23] D.Q. Shi, G.L. Gao, Z.W. Gao, P. Xiao, Application of expert fuzzy PID method for temperature control of heating furnace, Proc. Eng. 29 (2012) 257–261. [24] Z.H. Dou, L.Y. Sun, Design of temperature controller for heating furnace in oil field, Phys. Proc. 24 (Part C) (2012) 2083–2088. [25] F. Ding, R. Chao, X.H. Liu, P. Wang, H. Zhang, The research of heating furnace with heavy oil desander, Energy Proc. 14 (2012) 674–680.