- Email: [email protected]
Combustion and emissions characteristics of dual-channel double-vortex combustion for gas turbine engines
Applied Energy 130 (2014) 314–325
Contents lists available at ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
Combustion and emissions characteristics of dual-channel double-vortex combustion for gas turbine engines R.C. Zhang ⇑, W.J. Fan, Q. Shi, W.L. Tan School of Energy and Power Engineering, Beijing University of Aeronautics and Astronautics, Xueyuan Road, Beijing 100191, China
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
A novel double-vortex combustor
with a dual channel was designed. The preheating effect of the
evaporation tube is conducive to improving the combustion and emissions performance. The combustion organization method of the combustor is reasonable. The staged method significantly affects the performance of the combustor.
a r t i c l e
i n f o
Article history: Received 9 December 2013 Received in revised form 27 April 2014 Accepted 25 May 2014
Keywords: Vortex combustor Structure Performance Emission Experiment Aviation kerosene
a b s t r a c t A vortex combustor is a novel gas turbine combustor that uses staged combustion technology. Research examining the combustion organization method of the pilot combustion zone and the mainstream combustion zone is an important component of the design of the structure of a vortex combustor. In this paper, a new type of single-cavity vortex combustor fueled with aviation kerosene is presented. A doublevortex flow field structure and an evaporation tube for the fuel supply are used in the pilot zone. The flow-field structure of a double recirculation zone and a pneumatic atomization injector for the fuel supply are used in the mainstream combustion zone. The combustion experiment was performed under atmospheric pressure. The influence of the air-flow parameters, fuel parameters and staged method on the combustion performance and the characteristics of the pollutant emissions were studied in detail. Research indicates that the inlet temperature and the staged method primarily influence the ignition limit, lean blowout, combustion efficiency, temperature distribution of the outlet and pollutant emissions. The equivalence ratio primarily influences the temperature distribution of the wall and pollutant emissions. The inlet velocity influences the total pressure loss of the combustor. Ó 2014 Elsevier Ltd. All rights reserved.
Contents 1. 2.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 Combustor structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 2.1. Overall configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
⇑ Corresponding author. Tel.: +86 10 82317422. E-mail address: [email protected] (R.C. Zhang). http://dx.doi.org/10.1016/j.apenergy.2014.05.059 0306-2619/Ó 2014 Elsevier Ltd. All rights reserved.
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
3. 4.
5.
6.
7. 8.
2.2. Oil-supply device. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Test piece . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Experimental method. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Flow and combustion-performance analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Performance of ignition and lean blowout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Combustion efficiency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Total pressure loss. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Temperature-distribution analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Outlet-temperature distribution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Wall-temperature distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pollutant-emissions analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1. Carbon monoxide emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2. Nitrogen oxide emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3. Unburned hydrocarbon emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Numerical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
315
316 318 318 319 319 320 320 321 321 321 322 322 322 322 322 324 324 324
Nomenclature
g Standard d EI L K Ma Oh OTDF Q Re T V w
U
k diameter emission index stoichiometric ratio K number inlet Mach number Ohnesorge number outlet temperature distribution factor flow rate Reynolds number temperature velocity velocity normal to the surface equivalence ratio
1. Introduction With the development of aero gas turbine technology, the inlet temperature of the combustor and the temperature rise of the inlet and outlet of the combustor are gradually increased. Thus, control of the pollutant emissions in the combustor becomes more difficult. Studies show that the pollutant emissions of a gas turbine combustor can be effectively reduced using staged combustion technology [1–3] or flameless combustion technology [4–7]. The common staged combustion modes include lean premixed prevaporized (LPP) [8,9], variable-geometry combustor (VGC) [10], rich-burn, quick-mix, lean-burn (RQL) [11,12], twin annular premixing swirler (TAPS) [13] and trapped-vortex combustor (TVC) [14,15]. Among them, TVC has the advantages of a wide stable-combustion range and a compact structure [16,17]. Thus, vortex-combustion technology has potential in the field of aero gas turbines. A vortex combustor is comprised of a cavity pilot combustion zone and a mainstream combustion zone. Fuel and air are injected into both zones in a specific way, and the injection can strengthen the stability of the vortex, which will help to improve the combustion performance [18–20]. Under low-power conditions, only the pilot zone is in operation. The main pollutants are carbon monoxide (CO) and unburned hydrocarbon (UHC). Under high-power conditions, the two zones are in operation simultaneously, and
l q r C,
combustion efficiency surface tension viscosity density total pressure loss coefficient M, a, m, n, x, y constant
Sub and cavity exp inlet main num outlet wall
superscripts cavity combustion zone experimental inlet of the combustor mainstream combustion zone numerical outlet of the combustor wall of the cavity
the main pollutants are nitrogen oxides (NOx). By using an optimized air-inlet mode, fuel-inlet mode and fuel–air-mixing mode in the combustion zone, the pollutant emissions for both the low- and high-power conditions can be expected to be reduced [21–23]. These optimized modes should be able to reinforce the trapped vortex in the cavity zone, reduce the residence time in the cavity zone, and enhance the mass and energy transport between the cavity zone and the mainstream zone [2,16]. The traditional trapped-vortex combustor has a symmetrical, double-cavity structure. The cavity zone is the pilot zone. To further reduce the weight of the combustor, the number of pilot zones is reduced from two to one, thus forming an asymmetric structure with a unilateral cavity [24]. Compared with the double-cavity structure, the single-cavity structure is more compact. However, in this case, adjustment of the temperature-distribution field of the outlet of the combustor is more difficult. For an aero gas turbine combustor, there are several requirements for the performance, which sometimes conflict with each other [1,14,25–28]. In research on the structure of a combustor, the overall performance of a new type combustor must be studied in detail. Through repeated optimization, all technical indexes of the combustor can meet the design requirements. In this paper, a single-cavity double-vortex combustor is introduced. In the air supply of the cavity combustion zone, several rows of air jets are injected into the cavity zone to form a
316
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
Fig. 1. Schematic diagram of the double vortex in the cavity combustion zone.
double-vortex flow-field structure and cool the wall of the cavity. As shown in Fig. 1, the double vortex is generated by air jets from holes in the upper region of the fore-wall and the middle region of the after-wall of the cavity. The rotation directions of the two vortices are opposite. Additionally, the mainstream combustion zone adopts a double-channel inlet mode to form the structure of the double recirculation zone. In the fuel supply, the cavity zone uses an evaporation tube. The direction of the fuel injection is consistent with the air-flow direction. This can strengthen the stability of the vortex in the cavity zone. The mainstream zone uses a pneumatic atomization injector for the fuel supply. The fuel-injection point is located in the front of the air inlet channel, where the fuel can be fully mixed with the air. The influence of the air-flow parameters, the fuel parameters and the staged method on the performance of the combustor was studied. The results can be used for the subsequent structural design and optimization of single-cavity vortex combustors.
2. Combustor structure 2.1. Overall configuration A double-vortex combustor uses aviation kerosene as the fuel and has the structure shown in Figs. 2 and 3. The test piece is one-third of the whole combustor because of the machining cost, the experimental cost, and the necessity, as shown in Fig. 4. The combustor has a single cavity structure, which is beneficial to reduce the flow resistance and the weight of the combustor. Two rows of air-inlet holes are arranged staggeredly in the upper front wall of the cavity. One row of air-inlet holes is arranged in the center of the back wall. The cavity zone can form a double-vortex flow-field structure. The mainstream air flow enters into the mainstream combustion zone from the two symmetric channels, thus forming two vortexes with opposite directions in the mainstream zone. A row of cooling air inlet holes is arranged at the lower right of the cavity. The air film formed by the cooling air can effectively protect the corner zone. Two rows of mixing holes and three rows
Fig. 2. Two-dimensional schematic diagram of the double-vortex combustor with dual channels.
Fig. 3. Three-dimensional schematic diagram of the double-vortex combustor with dual channels.
of transpiration cooling holes are arranged in the mixing section of the wall, and the uniformity of the temperature distribution of the outlet can be improved at the same time as that at the cooling wall. The flowrate allocation has been calculated, and the flow field structure has been analyzed. Previous results show that the temperature of the combustion zone is important for a low-emission combustor. To achieve a high combustion efficiency and low pollutant emissions of CO and UHC, the equivalence ratios of the combustion zones should be approximately 0.8. To minimize NOx and smoke, the temperatures of the combustion zones should be kept below 1850 K [1]. Based on these design criteria and the numerical results of flow-field research, the ratios of the air flowrate of each zone to the total air flowrate are selected. The ratios are as follows: the air-inlet flowrate in the mainstream zone is 45%, the air-inlet flowrate in the cavity zone is 23%, the air used for cooling and the air inlet flowrate in the mixing zone is 32%. In addition, the area blockage ratio of the mainstream flame holder is 43%.
2.2. Oil-supply device The evaporation tube is selected as the fuel-supply device in the cavity zone. The combustion products in the cavity combustion zone preheat the evaporation tube, which is conducive to the atomization and evaporation of fuel to avoid creating a local fuel-rich zone. This can effectively improve the combustion efficiency of the combustor and also the uniformity of the temperature distribution. Due to the simple structure of the evaporation tube, the weight of the combustor is lighter than that of the
Fig. 4. Three-dimensional schematic diagram of the cavity.
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
Fig. 5. Evaporation tube (fuel supply device of the cavity zone).
Fig. 6. Evaporation performance of the evaporation tube [31].
conventional combustor with a centrifugal injector [29–31]. This is conducive to the improvement of the thrust-to-weight ratio of the aero gas turbine. The fuel supply of a double-vortex combustor involves four circumferentially and evenly arranged evaporation tubes. The structure is shown as Fig. 5. Part of the fuel is injected from the two holes in the front of the evaporation tube, from the upper left to the upper right of the cavity zone. The direction of injection is the same as the air-flow direction at the injection point. Thus, the vortex is strengthened. The remaining portion of the fuel is injected from the two holes on the side of the evaporation tube to achieve the effect of cross-flame. To avoid vapor lock, part of the evaporation tube is located outside the cavity zone, and the other part is located in the cavity zone. The injection point of fuel bar is arranged in the tube region that remains outside the cavity combustion zone, and the T-type tube used for evaporation is arranged at the fringe of the combustion zone. In the computational process of structural parameters, the jet velocity of the fuel–air mixture of the evaporation tube should be close to the air velocity of the zone, where the T-type tube is located. This study was conducted under the conditions of the maximum inlet temperature of the air, the maximum flow rate of the fuel, and complete evaporation of the fuel. Finally, the structural parameters, especially the inner diameter of the tube, are determined. An experimental study of the atomization and evaporation performance of the T-type tube was conducted by the author; some of these results are shown in Fig. 6 [31]. The results show that the air velocity is the major factor influencing the evaporation performance and that vapor lock is avoided. A pneumatic atomization injector is used as the fuel supply device in the mainstream zone. The injector has a total of six nozzles. The air and fuel for atomization are injected from the nozzle of the fuel injector in the same direction. The fuel–air mixture enters into the mainstream combustion zone from the two symmetric channels, with fuel injection in the downstream direction. There are six cross-flame plates downstream of the nozzle, as shown in Fig. 7. The fuel–air mixture is injected onto the plate [32]. The atomization quality of the fuel can be effectively improved through collision, and hence the combustion properties in the mainstream combustion zone are improved. There is no ignition spark plug in the mainstream zone. The fuel–air mixture of the mainstream zone is ignited by the combustion products from the cavity zone. Because the low-velocity zone is formed behind the cross-flame plate, the combustion products of the cavity zone can enter into the mainstream zone, which helps to improve the temperature in the recirculation zone. The cross-flame plates play the role of the flame holder in the mainstream zone. This enhances the mass and energy transport from the cavity zone to the mainstream zone and further improves the combustion performance of the combustor. Meanwhile, it improves the outlettemperature distribution of the combustor in the radial direction.
Fig. 7. Schematic diagram of the cross-flame plates.
Table 1 Droplet-wall collision conditions. Parameter
Range
q (Density) (kg/m3)
779 (10–60) 106 [31] 60–80 0.00115 0.02526 (0.4–3.2) 103 (33–82) 103 149–822
d (Diameter) (m) w (Velocity normal to the surface) (m/s) l (Viscosity) (Pas) k (Surface tension) (N/m) Re (Reynolds number) (=qdw/l) Oh (Ohnesorge number) (=l/(qkd)0.5) K (K number) (=OhRe1.25)
317
Fig. 8. The position of the ignition spark plug.
318
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
Fig. 11. The positions of the measuring points.
Fig. 9. Photo of the combustor (outside).
Based on previous research results on droplet-wall collision, the K number can be used to analyze the collisional processes. A K number exceeding 57.7 will lead to splashing of the droplet, and a K number less than 57.7 will lead to deposition of the droplet [33–35]. In our research, the droplet-wall collision conditions are shown in Table 1, and the diameters of the droplets were measured in the atomization experiment of the fuel injector of the mainstream zone by a Malvern laser particle size analyzer. The K numbers under all collision conditions exceed 57.7, and the droplets will not deposit onto the cross-flame plate. Therefore, the formation of char deposit can be avoided. An ignition spark plug is arranged in the cavity zone, with the installation position shown in Fig. 8. The local equivalence ratio at the end surface of the plug is relatively high, with a low air-flow velocity, which is favors ignition. In the circumferential direction, the plug is located between the two adjacent evaporation tubes.
Fig. 12. Ignition performance of the cavity zone.
in the inner wall of the mainstream zone. The cooling-air film formed can effectively protect the inner wall. On the side of the shell of the test piece, two quartz glass observation windows are arranged for observing combustion conditions in the combustor.
2.3. Test piece A photo of the test piece is shown in Fig. 9. To isolate the factors that affect the combustion properties for separate study, a set of air pipelines is used to provide the air needed for the atomization with the evaporation tube. The pressure of the air supply remains constant, and temperature is the atmospheric temperature. Multiple rows of inclined holes for the inlet of the cooling air are arranged
3. Experimental method The experiments were performed under atmospheric pressure. The experimental system was mainly composed of the test piece, the air-supply system, the fuel-supply system, the heating system for the mainstream and the measurement-control system, as
Fig. 10. Schematic diagram of the experimental system of the double-vortex combustor.
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
Fig. 13. Lean blowout performance of the cavity zone.
319
Fig. 15. Combustion efficiency of the vortex combustor (only the cavity zone was fueled).
shown in Fig. 10. Through data acquisition, the computer can display and save the data in real time. The positions of the temperature-measuring point, the pressure-measuring point, and the gas composition-measuring point are shown in Fig. 11. The equivalence ratio in the cavity Ucavity, the equivalence ratio of the ignition in the mainstream zone Uignition-main and the total equivalence ratio in the combustor Utotal were defined as follows, respectively:
Ucav ity ¼ Q cav ity-fuel =ð0:23Lfuel Q total-air Þ
ð1Þ
Uignition-main ¼ Q main-fuel =ðLfuel Q total-air Þ
ð2Þ
Utotal ¼ ðQ main-fuel þ Q cav ity-fuel Þ=ðLfuel Q total-air Þ:
ð3Þ
The measurement accuracies of the orifice flowmeter, the coriolis mass flowmeter, the pressure sensor, and the thermocouple are 0.5%, 0.2%, 0.2%, and 1%, respectively. The component measurement accuracies of CO, CO2, NO, NO2, and O2 are 2%, 1.5%, 2.8%, 1%, and 1.2%, respectively. According to the uncertainty-analysis theory, the accuracies of the total equivalence ratio Utotal, the total pressure-loss coefficient r, the combustion efficiency gtotal, and the outlet-temperaturedistribution coefficient OTDF are 0.5%, 3.6%, 7.8%, and 11.2%, respectively [32]. 4. Flow and combustion-performance analysis 4.1. Performance of ignition and lean blowout
reversed. The results are shown in Fig. 12, and the total equivalence ratio of the ignition was between 0.19 and 0.22. The influence of the inlet Mach number on the ignition performance in the cavity zone is smaller. This is because double vortexes are formed in the cavity zone in the double-vortex combustor. The end surface of the igniter plug is positioned above the vortex. The influence of the variation of the mainstream velocity on the flow-field structure is small. The influence of the inlet temperature on the ignition performance of the cavity zone is large. With an increase in the inlet temperature, the reaction rate increases, and the equivalence ratio of the ignition decreases. Thus, the ignition performance is improved. As shown in Fig. 13, the total equivalence ratio of the blowout was between 0.04 and 0.09, and the equivalence ratio of the blowout in the combustor decreases with an increased inlet Mach number. This is contrary to the conventional experimental results for a general combustor. When the inlet velocity of the vortex combustor increases, the air velocity in the cavity zone increases. The corresponding fuel supply increases as well. Because the cavity zone has a double-vortex structure, when the inlet velocity increases, the flow velocity of the upper vortex slightly increases, and the fuel supply quantity increases. This leads to a large local equivalence ratio in the upper cavity zone. Thus, the performance of the lean blowout is improved. In addition, the increase of the air temperature of the inlet is beneficial to the improvement of the blowout performance. The ignition performance of the mainstream zone is shown in Fig. 14, and the equivalence ratio of the ignition in the mainstream zone was between 0.11 and 0.15. When the inlet temperature increases or the equivalence ratio of the cavity zone increases,
In the combustor, the ignition plug is only arranged in the cavity zone. In the test, the cavity zone was ignited first, and then the fuel supply in the mainstream zone was gradually increased until the mainstream zone was ignited. For blowout, the sequence was
Fig. 14. Ignition performance of the mainstream zone.
Fig. 16. Combustion efficiency of the vortex combustor (the cavity zone and the mainstream zone were fueled).
320
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
the equivalence ratio of the ignition in the mainstream zone decreases. This is because the increase in the temperature improves the rate of the chemical reaction. When the inlet Mach number increases, the equivalence ratio of the ignition of the mainstream zone decreases. This is because, when the velocity increases, the combustion products of the cavity zone are mixed more intensively with those in the mainstream zone. As a result, more combustion products in the cavity zone enter into the mainstream zone. Compared with the equivalence ratio of the ignition in the cavity zone, the equivalence ratio of the ignition in the mainstream zone is small. This indicates that the combustion products in the cavity zone have a good preheating effect on the inlet air in the mainstream zone. In addition, the equivalence ratio of the blowout in the mainstream zone is very small and close to zero, which is also due to the preheating effect of the combustion products of the cavity zone. 4.2. Combustion efficiency The gas-composition method was used to measure the combustion efficiency using Eq. (4) [30,36]:
gtotal ¼ ð½CO2 þ 0:531½CO 0:397½H2 0:319½CH4 Þ=ð½CO2 þ ½CO þ ½UHCÞ:
ð4Þ
When only the cavity zone is in combustion (Fig. 15), the combustion efficiency of the combustor was between 91% and 99%. With an increase in the equivalence ratio, the combustion efficiency first increases significantly and then decreases slightly. This phenomenon is due to the increase in the chemical-reaction rate as well as the increase in the equivalence ratio. The temperature of the evaporation tube increases, the evaporation rate of the fuel increases, and the combustion efficiency is improved. When the equivalence ratio reaches a certain value, the cavity zone enters into the fuel-rich combustion state. The equivalence ratio increases, and the combustion efficiency decreases. When the inlet temperature is low (Tinlet = 450 K), the combustion efficiency is low. When the inlet temperature increases to 550 K, the combustion efficiency increases. At this time, by further improving the inlet temperature, the efficiency does not change. This is because as the inlet temperature increases, the evaporation rate in the evaporation tubes increases, making the combustion more complete, and the combustion efficiency increases. However, the increased amplitude of the evaporation rate is limited. When the fuel is supplied to the two zones (Fig. 16), the combustion efficiency of the combustor was between 94% and 99%. At Ucavity = 0.6, the combustion efficiency is the lowest. When the
Fig. 17. Influence of the total equivalence ratio and Ma on the total pressure-loss coefficient.
equivalence ratio of the cavity zone increases from 0.6 to 0.7, the temperature of the combustion products in the cavity zone increases, thereby raising the temperature of the recirculation zone in the mainstream zone. The combustion in the mainstream zone becomes more complete. When the equivalence ratio of the cavity zone further increases to 0.8, the combustion efficiency of the combustor slightly increases. Under the same equivalence ratio of the cavity zone, when the Qmain-fuel/Qtotal-fuel ratio is high, the combustion efficiency is high. This indicates that, under stablecombustion conditions, the equivalence ratio of the cavity zone should be increased, and the amount of air in the cavity zone should be reduced. In addition, when the fuel is supplied to both zones and the equivalence ratio of the cavity zone remains unchanged, this study shows several effects. (1) When the inlet temperature of the mainstream zone increases, the combustion efficiency increases. (2) When the total equivalence ratio is small, the combustion efficiency increases significantly when the inlet Mach number increases. This is because, when the equivalence ratio is small and the combustion temperature is low, an increase in the inlet Mach number is in favor of the mixing of the fuel injected by the injector with the air in the inlet channel. This is conducive to an improvement in the combustion efficiency. (3) When the total equivalence ratio is relatively large, the combustion efficiency for every Mach number is relatively high. The influence of the Mach number on the combustion efficiency is not obvious. This is because, when the equivalence ratio is large, the combustion temperature is high, and the combustion of fuel is sufficient. 4.3. Total pressure loss The total pressure loss is an important index of the combustor performance. Usually, the total pressure-loss coefficient, r, is used for analysis:
r ¼ ðPinlet;average Poutlet;average Þ=Pinlet;average
ð5Þ
Because the effects of the staged method and the inlet temperature on the total pressure loss are very small, the effects of the Mach number and the total equivalence ratio are mainly analyzed below. As shown in Fig. 17, r of the combustor was between 0.04 and 0.11, and it is in direct proportion to the square of the Mach number. This is because, when the velocity increases, the flow resistance increases as well. When in the combustion state, r is approximately 0.015 higher than that in the cold state. This is because the heat resistance increases when the temperature increases.
Fig. 18. OTDF of the vortex combustor (the cavity zone and the mainstream zone were fueled).
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
(a) Ma=0.15, Tinlet=550K, Φcavity =0.6, Φtotal=0.37 (OTDF=0.36)
(b) Ma=0.15, Tinlet=550K, Φcavity =0.7, Φtotal=0.37 (OTDF=0.23) Fig. 19. Outlet temperature-distribution diagrams (the cavity zone and the mainstream zone were fueled).
321
is supplied to both zones, the result is shown in Fig. 18. Under the same equivalence ratio of the cavity zone, the total equivalence ratio has little effect on the OTDF. This indicates that, when the temperature of the combustor increases, the highest temperature of the outlet of the combustor increases as well. When the equivalence ratio of the cavity zone increases, the OTDF decreases significantly. This is because the equivalence ratio of the cavity zone increases, and the temperature of the combustion products in the cavity zone increases as well. When the temperature in the mainstream zone increases, the distance required by the combustion of the fuel in the mainstream zone is shortened. The combustion products have a longer distance to be mixed. In addition, when the inlet temperature increases, the OTDF decreases. The reason is similar to that of the influence of the equivalence ratio of the cavity zone on the OTDF. When the inlet Mach number increases, the OTDF increases. This is because, when the two zones are in combustion at the same time, an increase in the Mach number is not conducive to the mixing of the combustion products in the two combustion zones. The temperature distribution of the outlet becomes nonuniform. The temperature distribution diagrams of the outlet are shown in Fig. 19. When the equivalence ratio of the cavity zone is small, the high-temperature zone is mainly located on the right side, and the area is small. When the equivalence ratio increases to 0.7, the temperature of the low-temperature zone of the lower left side increases. The highest temperature at the outlet decreases, and the temperature distribution is more uniform. Thus, with the increase of the equivalence ratio of the cavity zone, the temperature of the combustion products in the two lower recirculation zones of the mainstream zone can be raised, and the temperature-distribution uniformity is improved. 5.2. Wall-temperature distribution A type k thermocouple was used to measure the temperature of the outer wall of the cavity. The measuring points were arranged at the central cross-section of test piece, with the specific arrangement shown in Fig. 20. The test result is shown in Fig. 21. In each state, the temperature distributions of the wall of the combustor are essentially identical. The temperatures in the upper front wall (measuring point 3) and the upper back wall (measuring point 5) are relatively high. This indicates that the heat exchange between the combustion products in the cavity zone and the cavity wall is mainly concentrated in these two regions. When the equivalence ratio of the cavity zone increases, the temperature of the cavity wall increases on average, and the variation of the temperature of wall in the mixing section is relatively small. The combustion in the cavity zone of the
Fig. 20. Positions of the measuring points of the wall temperature.
Fig. 21. Wall-temperature distribution of the cavity.
5. Temperature-distribution analysis 5.1. Outlet-temperature distribution The temperature-distribution uniformity of the outlet is an important index of the performance of the combustor. Usually, the outlet temperature-distribution factor, OTDF, is used to reflect the uniformity of the temperature field:
OTDF ¼ ðT outlet;max T outlet;average Þ=ðT outlet;average T inlet;average Þ:
ð6Þ
When the fuel is only supplied in the cavity zone, the temperature rise is small, and the temperature-distribution uniformity of the outlet is not important. The temperature-distribution uniformity of the outlet is researched under a large load. When the fuel
Fig. 22. Emissions of carbon monoxide (only the cavity zone was fueled).
322
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
6.1. Carbon monoxide emissions
Fig. 23. Emissions of carbon monoxide (the cavity zone and the mainstream zone were fueled).
double-vortex combustor is mainly in the upper vortex. When the combustion products flow to the lower vortex, the combustion products begin to be mixed with the air inflow from the channel. Then, the temperatures of wall at the lower back wall and the mixing section are relatively low, with a small variation in the equivalence ratio of the cavity zone. For the same equivalence ratio of the cavity zone, when the fuel is supplied to both zones, the temperature of the cavity wall is higher than that when the fuel is only supplied to the cavity zone. The combustion state of the mainstream zone influences the temperature distribution of the cavity wall. This indicates that part of the combustion products of the mainstream zone are entrained into the cavity zone, resulting in a high temperature of the cavity wall.
6. Pollutant-emissions analysis According to the relevant industry standards [36], the emission index (EI) can be calculated by the following formula:
EISpecies ¼ ð½Speciesð1 þ aC1 Þ=ð½CO þ ½CO2 þ ½UHCÞÞ 3
ð10 MSpecies =MC þ n=mMH Þ
ð7Þ
where C1 = (2C2 (n/m))/(4(1 + n (aC2/2))), MCO = 28.011, MC = 12.011, MH = 1.008, MUHC = 16.043, MNO2 = 46.008, C2 = (2 [CO] ((2/x) (y/2x))[UHC]+[NO2])/([CO]+[CO2]+[UHC]), x = 1, y = 4, m = 10, n = 20, and a = 0.00032.
When the fuel is supplied only to the cavity zone, the result is shown in Fig. 22. With an increased equivalence ratio, EICO gradually decreases. This occurs because, when the temperature of the combustion products in the combustion zone increases, the combustion becomes complete. When the equivalence ratio is greater than 1, EICO increases slightly. When the fuel is supplied to both zones, as shown in Fig. 23, with an increased total equivalence ratio, EICO gradually decreases. The reason is the same as above. Compared with the situation in which the fuel is only supplied to the cavity zone, EICO for the situation in which the fuel is supplied to both zones is far smaller, given the same total equivalence ratio. This shows that, when the fuel is supplied to both zones, a large amount of CO produced by the combustion in the cavity zone becomes CO2 in the mainstream combustion zone due to oxidation. In addition, when the fuel is supplied to both zones, the following results are obtained for the constant equivalence ratio of the cavity zone. (1) With increased inlet temperature of the combustor, EICO decreases. (2) When the total equivalence ratio is small, the inlet Mach number increases, and EICO decreases significantly. (3) When the total equivalence ratio is relatively large, the EICO for each Mach number is small, and the influence of the Mach number is not obvious. The reason is similar to that for the variation of the combustion efficiency. 6.2. Nitrogen oxide emissions Because NOx is mainly produced under high load conditions, the situation in which the fuel is supplied to both zones is analyzed, as shown in Fig. 24. With an increase in the total equivalence ratio, EINOx increases, which is due to the increase of the flame temperature in the mainstream combustion zone. When the Mach number increases, EINOx increases. This is because the increase in the combustion efficiency leads to the increase in the flame temperature. 6.3. Unburned hydrocarbon emissions The EIUHC in the combustor is in the range of 1–30 g/(kg fuel). Its variation pattern is consistent with that of EICO but with a smaller value. This is because the generation rule of unburned hydrocarbons is similar to that of CO. However, the unburned hydrocarbons are more likely to be oxidized, and the values of the emissions are smaller. 7. Numerical analysis
Fig. 24. Emissions of nitrogen oxides (the cavity zone and the mainstream zone were fueled).
A 30° fan-shaped area was selected for modeling, and an unstructured mesh system consisting of 3.2 million cells (illustrated in Fig. 25) was used. Mesh refinement was specifically conducted in the regions of the air-inlet holes and the cavity zone. Under steady-state conditions, a numerical simulation was performed using the k–e turbulence model. The number of the fuel injection point and the injection direction of the fuel are consistent with the entity. The velocity distribution in the cross-section with the crossflame plate is shown in Fig. 26. A vortex exists in the upper right corner of the cavity zone, and a low-velocity region is formed in the rear of the cross-flame plate. The velocity distribution in the cross-section without the cross-flame plate is shown in Fig. 27. There are vortices in the upper right corner, the upper left corner and the lower right corner of the cavity zone. A recirculation zone is formed in the rear of the mainstream holder, and the zone is extended to the location of the air injection from the mixing holes.
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
323
Fig. 25. Schematic diagram of the unstructured mesh system.
Fig. 26. Velocity distribution in the cross-section with the cross-flame plate.
Fig. 29. Temperature distribution in the cross-section without the cross-flame plate.
Fig. 27. Velocity distribution in the cross-section without the cross-flame plate.
Fig. 30. Comparisons of r and the OTDF between the numerical simulations and the experiments.
Fig. 28. Temperature distribution in the cross-section with the cross-flame plate.
Temperature-distribution diagrams in the cross-section of combustor are shown in Figs. 28 and 29. In the cross-section with the cross-flame plate, there are high-temperature zones in the lower right corner of the cavity zone and the lower area of the recirculation zone in the mainstream zone. In the cross-section without the cross-flame plate, there are high-temperature zones in the upper
left corner of the cavity zone and the lower area of the mainstream zone, and the mainstream combustion zone is strongly separated from the cavity combustion zone by the air jet of mainstream. The cross-flame plates enhance the energy transport between the two combustion zones and improve the radial temperature distribution of the outlet. Comparisons of r, OTDF, EICO and EINOx between the numerical simulations and the experiments are shown in Figs. 30 and 31. The numerical results correspond to the experimental results. The total equivalence ratio of the combustor has a large influence on the EICO and EINOx values. Thus, it is important to control the temperature of the combustion zone in an appropriate range.
324
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
References
Fig. 31. Comparisons of EICO and EINOx between the numerical simulations and the experiments.
Therefore, the flowrate of the combustion zone should be exactly considered in the design process of the combustor, based on the operating conditions of the combustor, especially the total equivalence ratio.
8. Conclusions An experimental and numerical investigation of a single-cavity vortex combustor fueled with aviation kerosene is presented. A double-vortex combustion mode is adopted in the pilot combustion zone of the staged combustor, and the jets of the fuel–air mixture from the evaporation tube enhance the stability of the vortex in the pilot zone. The combustion mode of the double recirculation zone is adopted in the mainstream combustion zone, and the fuel injected by the pneumatic atomization injector is mixed with mainstream air before entering the mainstream zone. The cross-flame plates are located in the upper site of the flame holder to enhance the energy transport between the pilot and the mainstream zone. The results of the combustion experiment show that the combustion mode of the double vortex with a dual channel has good comprehensive performance. The combustion organization method in the combustion zone is reasonable. A stable double vortex can be formed in the pilot zone. At the same time, the preheating effect of the evaporation tube causes the fuel to have high-quality atomization. Due to the rapid fuel–air mixing and the high stability of the double-vortex, stable combustion has been demonstrated over a wide range of total equivalence ratios. Under typical operating conditions, high combustion performance with low pollutant emissions was obtained, and the outlet-temperature distribution of the combustor was uniform, which is very important for a single-cavity combustor with an asymmetric structure. The influence of each factor on the combustion performance and pollutant emissions is very complex. Among these factors, the inlet velocity has little effect on the performance of combustion. This is because the inlet velocity has little effect on the upper vortex in the cavity zone. The staged method has a significant effect on the multiple performance of the combustor, which is due to the complex mass and energy transport between the two combustion zones. Acknowledgement This work was supported by the Postdoctoral Science Foundation of China (2013M530514). The support is gratefully acknowledged.
[1] Lefebvre AH, Ballal DR. Gas turbine combustion. Taylor & Francis Group; 2010. [2] Roquemore WM, Shouse D, Burrus D, et al. Vortex combustor concept for gas turbine engines. In: 39th aerospace sciences meeting and exhibit, AIAA 20010483; 2001. http://dx.doi.org/10.2514/6.2001-483. [3] Ballester J, Sanz A, Gonzalez MA. Investigation of the characteristics and stability of air-staged flames. Exp Therm Fluid Sci 2008;32:776–90. http:// dx.doi.org/10.1016/j.expthermflusci.2007.09.007. [4] Khalil AEE, Gupta AK. Distributed swirl combustion for gas turbine application. Appl Energy 2011;88:4898–907. http://dx.doi.org/10.1016/j.apenergy.2011. 06.051. [5] Khalil AEE, Gupta AK. Fuel flexible distributed combustion for efficient and clean gas turbine engines. Appl Energy 2013;109:267–74. http://dx.doi.org/ 10.1016/j.apenergy.2013.04.052. [6] Arghode VK, Gupta AK. Role of thermal intensity on operational characteristics of ultra-low emission colorless distributed combustion. Appl Energy 2013;111:930–56. http://dx.doi.org/10.1016/j.apenergy.2013.06.039. [7] Arghode VK, Khalil AEE, Gupta AK. Fuel dilution and liquid fuel operational effects on ultra-high thermal intensity distributed combustor. Appl Energy 2012;95:132–8. http://dx.doi.org/10.1016/j.apenergy.2012.02.020. [8] Dhanuka SK, Temme JE, Driscoll JF, et al. Vortex-shedding and mixing layer effects on periodic flashback in a lean premixed prevaporized gas turbine combustor. Proc Combust Inst 2009;32:2901–8. http://dx.doi.org/10.1016/ j.proci.2008.06.155. [9] Bucher J, Edmonds RG, Steele RC, et al. The development of a lean-premixed trapped vortex combusto. In: Proceedings of ASME turbo expo; 2003. http:// dx.doi.org/10.1115/GT2003-38236. [10] Giuliani F, Woisetschläger J, Leitgeb T. Design and validation of a burner with variable geometry for extended combustion range. ASME Turbo Expo 2012. http://dx.doi.org/10.1115/GT2012-68236. [11] Straub DL, Casleton KH, Lewis RE, et al. Assessment of rich-burn, quick-mix, lean-burn trapped vortex combustor for stationary gas turbines. J Eng Gas Turb Power 2005;127:36–41. http://dx.doi.org/10.1115/1.1789152. [12] Samuelsen GS, Brouwer J, Vardakas MA, et al. Experimental and modeling investigation of the effect of air preheat on the formation of NOx in an RQL combustor. Heat Mass Transf 2013;49:219–31. http://dx.doi.org/10.1007/ s00231-012-1080-0. [13] Dodds W. Twin annular premixing swirler (TAPS) combustor. In: The roaring 20th aviation noise & air quality symposium; 2005. [14] Hsu KY, Goss LP, Roquemore WM. Characteristics of a trapped- vortex combustor. J Propul Power 1998;14:57–65. http://dx.doi.org/10.2514/2.5266. [15] Bruno C, Losurdo M. The trapped vortex combustor: an advanced combustion technology for aerospace and gas turbine applications. In: Syred N, Khalatov A, editors. Advanced combustion and aerothermal technologies. Springer; 2007. http://dx.doi.org/10.1007/978-1-4020-6515-6_28. [16] Ezhil Kumar PK, Mishra DP. Numerical simulation of cavity flow structure in an axisymmetric trapped vortex combustor. Aerosp. Sci. Technol. 2012;21:16–23. http://dx.doi.org/10.1016/j.ast.2011.04.007. [17] Ezhil Kumar PK, Mishra DP. Numerical investigation of the flow and flame structure in an axisymmetric trapped vortex combustor. Fuel 2012;102:78–84. http://dx.doi.org/10.1016/j.fuel.2012.06.056. [18] Vengadesan S, Sony C. Enhanced vortex stability in trapped vortex combustor. Aeronaut J 2010;114:333–7. [19] Agarwal KK, Krishna S, Ravikrishna RV. Mixing enhancement in a compact trapped vortex combustor. Combust Sci Technol 2013;185:363–78. http:// dx.doi.org/10.1080/00102202.2012.721034. [20] Szwaja S, Jamrozik A, Tutak W. A two-stage combustion system for burning lean gasoline mixtures in a stationary spark ignited engine. Appl Energy 2013;105:271–81. http://dx.doi.org/10.1016/j.apenergy.2012.12.080. [21] Ghenai C, Zbeeb K, Janajreh I. Combustion of alternative fuels in vortex trapped combustor. Energy Convers Manage 2013;65:819–28. http://dx.doi.org/ 10.1016/j.enconman.2012.03.012. [22] Jin Y, He X, Zhang J, et al. Experimental study on emission performance of an LPP/TVC. Chin J Aeronaut 2012;25:335–41. http://dx.doi.org/10.1016/S10009361(11)60394-4. [23] Fan WJ, Yan M, Yi Q, et al. Low NOx emission of rich-burn, quick-mix, leanburn trapped vortex combustor. J Propul Technol 2006;27:88–91. [24] Singhal A, Ravikrishna RV. Single cavity trapped vortex combustor dynamics – Part-1: Experiments. Int J Spray Combust Dynam 2011;3:23–44. http:// dx.doi.org/10.1260/1756-8277.3.1.23. [25] Nemitallah MA, Habib MA. Experimental and numerical investigations of an atmospheric diffusion oxy-combustion flame in a gas turbine model combustor. Appl Energy 2013;111:401–15. http://dx.doi.org/10.1016/j. apenergy.2013.05.027. [26] Payri R, Salvador FJ, Gimeno J, et al. Influence of injector technology on injection and combustion development–Part 2: Combustion analysis. Appl Energy 2011;88:1130–9. http://dx.doi.org/10.1016/j.apenergy.2010.10.004. [27] Khalil AEE, Gupta AK. Swirling distributed combustion for clean energy conversion in gas turbine applications. Appl Energy 2011;88:3685–93. http:// dx.doi.org/10.1016/j.apenergy.2011.03.048. [28] Syred N, Giles A, Lewis J, et al. Effect of inlet and outlet configurations on blowoff and flashback with premixed combustion for methane and a high hydrogen content fuel in a generic swirl burner. Appl Energy 2014;116:288–96. http:// dx.doi.org/10.1016/j.apenergy.2013.11.071.
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325 [29] Xing F, Wang P, Zhang S, et al. Experiment and simulation study on lean blowout of trapped vortex combustor with various aspect ratios. Aerosp Sci Technol 2012;18:48–55. http://dx.doi.org/10.1016/j.ast.2011.03.010. [30] Xing F, Zhang S, Wang P, et al. Experimental investigation of a single trappedvortex combustor with a slight temperature raise. Aerosp Sci Technol 2010;14:520–5. http://dx.doi.org/10.1016/j.ast.2010.04.007. [31] Zhang R, Fan W, Song S. Atomization and evaporation performance experiment of evaporating tube for trapped vortex combustor. J Aerosp Power 2011;26:2495–502. [32] Zhang R. Fundamental investigation of trapped vortex combustor with high temperature rise. PhD thesis, Beijing University of Aeronautics and Astronautics; 2012.
325
[33] Mundo CHR, Sommerfeld M, Tropea C. Droplet-wall collisions: experimental studies of the deformation and breakup process. Int J Multiphase Flow 1995;21:151–73. http://dx.doi.org/10.1016/0301-9322(94)00069-V. [34] Yarin AL. Drop impact dynamics: splashing, spreading, receding, bouncing. Annu Rev Fluid Mech 2006;38:159–92. http://dx.doi.org/10.1146/annurev. fluid.38.050304.092144. [35] Zhai X, Peng R, Fan W, et al. Numerical simulation on a certain trapped vortex combustor. J Aerosp Power 2013;5:1134–41. [36] Aviation industry department of China, Aero-engine gaseous pollutants sampling and measurement procedures 1987, Aviation industry standard HB 6117-87.
Contents lists available at ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
Combustion and emissions characteristics of dual-channel double-vortex combustion for gas turbine engines R.C. Zhang ⇑, W.J. Fan, Q. Shi, W.L. Tan School of Energy and Power Engineering, Beijing University of Aeronautics and Astronautics, Xueyuan Road, Beijing 100191, China
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
A novel double-vortex combustor
with a dual channel was designed. The preheating effect of the
evaporation tube is conducive to improving the combustion and emissions performance. The combustion organization method of the combustor is reasonable. The staged method significantly affects the performance of the combustor.
a r t i c l e
i n f o
Article history: Received 9 December 2013 Received in revised form 27 April 2014 Accepted 25 May 2014
Keywords: Vortex combustor Structure Performance Emission Experiment Aviation kerosene
a b s t r a c t A vortex combustor is a novel gas turbine combustor that uses staged combustion technology. Research examining the combustion organization method of the pilot combustion zone and the mainstream combustion zone is an important component of the design of the structure of a vortex combustor. In this paper, a new type of single-cavity vortex combustor fueled with aviation kerosene is presented. A doublevortex flow field structure and an evaporation tube for the fuel supply are used in the pilot zone. The flow-field structure of a double recirculation zone and a pneumatic atomization injector for the fuel supply are used in the mainstream combustion zone. The combustion experiment was performed under atmospheric pressure. The influence of the air-flow parameters, fuel parameters and staged method on the combustion performance and the characteristics of the pollutant emissions were studied in detail. Research indicates that the inlet temperature and the staged method primarily influence the ignition limit, lean blowout, combustion efficiency, temperature distribution of the outlet and pollutant emissions. The equivalence ratio primarily influences the temperature distribution of the wall and pollutant emissions. The inlet velocity influences the total pressure loss of the combustor. Ó 2014 Elsevier Ltd. All rights reserved.
Contents 1. 2.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 Combustor structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 2.1. Overall configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
⇑ Corresponding author. Tel.: +86 10 82317422. E-mail address: [email protected] (R.C. Zhang). http://dx.doi.org/10.1016/j.apenergy.2014.05.059 0306-2619/Ó 2014 Elsevier Ltd. All rights reserved.
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
3. 4.
5.
6.
7. 8.
2.2. Oil-supply device. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Test piece . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Experimental method. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Flow and combustion-performance analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Performance of ignition and lean blowout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Combustion efficiency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Total pressure loss. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Temperature-distribution analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Outlet-temperature distribution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Wall-temperature distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pollutant-emissions analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1. Carbon monoxide emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2. Nitrogen oxide emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3. Unburned hydrocarbon emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Numerical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
315
316 318 318 319 319 320 320 321 321 321 322 322 322 322 322 324 324 324
Nomenclature
g Standard d EI L K Ma Oh OTDF Q Re T V w
U
k diameter emission index stoichiometric ratio K number inlet Mach number Ohnesorge number outlet temperature distribution factor flow rate Reynolds number temperature velocity velocity normal to the surface equivalence ratio
1. Introduction With the development of aero gas turbine technology, the inlet temperature of the combustor and the temperature rise of the inlet and outlet of the combustor are gradually increased. Thus, control of the pollutant emissions in the combustor becomes more difficult. Studies show that the pollutant emissions of a gas turbine combustor can be effectively reduced using staged combustion technology [1–3] or flameless combustion technology [4–7]. The common staged combustion modes include lean premixed prevaporized (LPP) [8,9], variable-geometry combustor (VGC) [10], rich-burn, quick-mix, lean-burn (RQL) [11,12], twin annular premixing swirler (TAPS) [13] and trapped-vortex combustor (TVC) [14,15]. Among them, TVC has the advantages of a wide stable-combustion range and a compact structure [16,17]. Thus, vortex-combustion technology has potential in the field of aero gas turbines. A vortex combustor is comprised of a cavity pilot combustion zone and a mainstream combustion zone. Fuel and air are injected into both zones in a specific way, and the injection can strengthen the stability of the vortex, which will help to improve the combustion performance [18–20]. Under low-power conditions, only the pilot zone is in operation. The main pollutants are carbon monoxide (CO) and unburned hydrocarbon (UHC). Under high-power conditions, the two zones are in operation simultaneously, and
l q r C,
combustion efficiency surface tension viscosity density total pressure loss coefficient M, a, m, n, x, y constant
Sub and cavity exp inlet main num outlet wall
superscripts cavity combustion zone experimental inlet of the combustor mainstream combustion zone numerical outlet of the combustor wall of the cavity
the main pollutants are nitrogen oxides (NOx). By using an optimized air-inlet mode, fuel-inlet mode and fuel–air-mixing mode in the combustion zone, the pollutant emissions for both the low- and high-power conditions can be expected to be reduced [21–23]. These optimized modes should be able to reinforce the trapped vortex in the cavity zone, reduce the residence time in the cavity zone, and enhance the mass and energy transport between the cavity zone and the mainstream zone [2,16]. The traditional trapped-vortex combustor has a symmetrical, double-cavity structure. The cavity zone is the pilot zone. To further reduce the weight of the combustor, the number of pilot zones is reduced from two to one, thus forming an asymmetric structure with a unilateral cavity [24]. Compared with the double-cavity structure, the single-cavity structure is more compact. However, in this case, adjustment of the temperature-distribution field of the outlet of the combustor is more difficult. For an aero gas turbine combustor, there are several requirements for the performance, which sometimes conflict with each other [1,14,25–28]. In research on the structure of a combustor, the overall performance of a new type combustor must be studied in detail. Through repeated optimization, all technical indexes of the combustor can meet the design requirements. In this paper, a single-cavity double-vortex combustor is introduced. In the air supply of the cavity combustion zone, several rows of air jets are injected into the cavity zone to form a
316
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
Fig. 1. Schematic diagram of the double vortex in the cavity combustion zone.
double-vortex flow-field structure and cool the wall of the cavity. As shown in Fig. 1, the double vortex is generated by air jets from holes in the upper region of the fore-wall and the middle region of the after-wall of the cavity. The rotation directions of the two vortices are opposite. Additionally, the mainstream combustion zone adopts a double-channel inlet mode to form the structure of the double recirculation zone. In the fuel supply, the cavity zone uses an evaporation tube. The direction of the fuel injection is consistent with the air-flow direction. This can strengthen the stability of the vortex in the cavity zone. The mainstream zone uses a pneumatic atomization injector for the fuel supply. The fuel-injection point is located in the front of the air inlet channel, where the fuel can be fully mixed with the air. The influence of the air-flow parameters, the fuel parameters and the staged method on the performance of the combustor was studied. The results can be used for the subsequent structural design and optimization of single-cavity vortex combustors.
2. Combustor structure 2.1. Overall configuration A double-vortex combustor uses aviation kerosene as the fuel and has the structure shown in Figs. 2 and 3. The test piece is one-third of the whole combustor because of the machining cost, the experimental cost, and the necessity, as shown in Fig. 4. The combustor has a single cavity structure, which is beneficial to reduce the flow resistance and the weight of the combustor. Two rows of air-inlet holes are arranged staggeredly in the upper front wall of the cavity. One row of air-inlet holes is arranged in the center of the back wall. The cavity zone can form a double-vortex flow-field structure. The mainstream air flow enters into the mainstream combustion zone from the two symmetric channels, thus forming two vortexes with opposite directions in the mainstream zone. A row of cooling air inlet holes is arranged at the lower right of the cavity. The air film formed by the cooling air can effectively protect the corner zone. Two rows of mixing holes and three rows
Fig. 2. Two-dimensional schematic diagram of the double-vortex combustor with dual channels.
Fig. 3. Three-dimensional schematic diagram of the double-vortex combustor with dual channels.
of transpiration cooling holes are arranged in the mixing section of the wall, and the uniformity of the temperature distribution of the outlet can be improved at the same time as that at the cooling wall. The flowrate allocation has been calculated, and the flow field structure has been analyzed. Previous results show that the temperature of the combustion zone is important for a low-emission combustor. To achieve a high combustion efficiency and low pollutant emissions of CO and UHC, the equivalence ratios of the combustion zones should be approximately 0.8. To minimize NOx and smoke, the temperatures of the combustion zones should be kept below 1850 K [1]. Based on these design criteria and the numerical results of flow-field research, the ratios of the air flowrate of each zone to the total air flowrate are selected. The ratios are as follows: the air-inlet flowrate in the mainstream zone is 45%, the air-inlet flowrate in the cavity zone is 23%, the air used for cooling and the air inlet flowrate in the mixing zone is 32%. In addition, the area blockage ratio of the mainstream flame holder is 43%.
2.2. Oil-supply device The evaporation tube is selected as the fuel-supply device in the cavity zone. The combustion products in the cavity combustion zone preheat the evaporation tube, which is conducive to the atomization and evaporation of fuel to avoid creating a local fuel-rich zone. This can effectively improve the combustion efficiency of the combustor and also the uniformity of the temperature distribution. Due to the simple structure of the evaporation tube, the weight of the combustor is lighter than that of the
Fig. 4. Three-dimensional schematic diagram of the cavity.
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
Fig. 5. Evaporation tube (fuel supply device of the cavity zone).
Fig. 6. Evaporation performance of the evaporation tube [31].
conventional combustor with a centrifugal injector [29–31]. This is conducive to the improvement of the thrust-to-weight ratio of the aero gas turbine. The fuel supply of a double-vortex combustor involves four circumferentially and evenly arranged evaporation tubes. The structure is shown as Fig. 5. Part of the fuel is injected from the two holes in the front of the evaporation tube, from the upper left to the upper right of the cavity zone. The direction of injection is the same as the air-flow direction at the injection point. Thus, the vortex is strengthened. The remaining portion of the fuel is injected from the two holes on the side of the evaporation tube to achieve the effect of cross-flame. To avoid vapor lock, part of the evaporation tube is located outside the cavity zone, and the other part is located in the cavity zone. The injection point of fuel bar is arranged in the tube region that remains outside the cavity combustion zone, and the T-type tube used for evaporation is arranged at the fringe of the combustion zone. In the computational process of structural parameters, the jet velocity of the fuel–air mixture of the evaporation tube should be close to the air velocity of the zone, where the T-type tube is located. This study was conducted under the conditions of the maximum inlet temperature of the air, the maximum flow rate of the fuel, and complete evaporation of the fuel. Finally, the structural parameters, especially the inner diameter of the tube, are determined. An experimental study of the atomization and evaporation performance of the T-type tube was conducted by the author; some of these results are shown in Fig. 6 [31]. The results show that the air velocity is the major factor influencing the evaporation performance and that vapor lock is avoided. A pneumatic atomization injector is used as the fuel supply device in the mainstream zone. The injector has a total of six nozzles. The air and fuel for atomization are injected from the nozzle of the fuel injector in the same direction. The fuel–air mixture enters into the mainstream combustion zone from the two symmetric channels, with fuel injection in the downstream direction. There are six cross-flame plates downstream of the nozzle, as shown in Fig. 7. The fuel–air mixture is injected onto the plate [32]. The atomization quality of the fuel can be effectively improved through collision, and hence the combustion properties in the mainstream combustion zone are improved. There is no ignition spark plug in the mainstream zone. The fuel–air mixture of the mainstream zone is ignited by the combustion products from the cavity zone. Because the low-velocity zone is formed behind the cross-flame plate, the combustion products of the cavity zone can enter into the mainstream zone, which helps to improve the temperature in the recirculation zone. The cross-flame plates play the role of the flame holder in the mainstream zone. This enhances the mass and energy transport from the cavity zone to the mainstream zone and further improves the combustion performance of the combustor. Meanwhile, it improves the outlettemperature distribution of the combustor in the radial direction.
Fig. 7. Schematic diagram of the cross-flame plates.
Table 1 Droplet-wall collision conditions. Parameter
Range
q (Density) (kg/m3)
779 (10–60) 106 [31] 60–80 0.00115 0.02526 (0.4–3.2) 103 (33–82) 103 149–822
d (Diameter) (m) w (Velocity normal to the surface) (m/s) l (Viscosity) (Pas) k (Surface tension) (N/m) Re (Reynolds number) (=qdw/l) Oh (Ohnesorge number) (=l/(qkd)0.5) K (K number) (=OhRe1.25)
317
Fig. 8. The position of the ignition spark plug.
318
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
Fig. 11. The positions of the measuring points.
Fig. 9. Photo of the combustor (outside).
Based on previous research results on droplet-wall collision, the K number can be used to analyze the collisional processes. A K number exceeding 57.7 will lead to splashing of the droplet, and a K number less than 57.7 will lead to deposition of the droplet [33–35]. In our research, the droplet-wall collision conditions are shown in Table 1, and the diameters of the droplets were measured in the atomization experiment of the fuel injector of the mainstream zone by a Malvern laser particle size analyzer. The K numbers under all collision conditions exceed 57.7, and the droplets will not deposit onto the cross-flame plate. Therefore, the formation of char deposit can be avoided. An ignition spark plug is arranged in the cavity zone, with the installation position shown in Fig. 8. The local equivalence ratio at the end surface of the plug is relatively high, with a low air-flow velocity, which is favors ignition. In the circumferential direction, the plug is located between the two adjacent evaporation tubes.
Fig. 12. Ignition performance of the cavity zone.
in the inner wall of the mainstream zone. The cooling-air film formed can effectively protect the inner wall. On the side of the shell of the test piece, two quartz glass observation windows are arranged for observing combustion conditions in the combustor.
2.3. Test piece A photo of the test piece is shown in Fig. 9. To isolate the factors that affect the combustion properties for separate study, a set of air pipelines is used to provide the air needed for the atomization with the evaporation tube. The pressure of the air supply remains constant, and temperature is the atmospheric temperature. Multiple rows of inclined holes for the inlet of the cooling air are arranged
3. Experimental method The experiments were performed under atmospheric pressure. The experimental system was mainly composed of the test piece, the air-supply system, the fuel-supply system, the heating system for the mainstream and the measurement-control system, as
Fig. 10. Schematic diagram of the experimental system of the double-vortex combustor.
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
Fig. 13. Lean blowout performance of the cavity zone.
319
Fig. 15. Combustion efficiency of the vortex combustor (only the cavity zone was fueled).
shown in Fig. 10. Through data acquisition, the computer can display and save the data in real time. The positions of the temperature-measuring point, the pressure-measuring point, and the gas composition-measuring point are shown in Fig. 11. The equivalence ratio in the cavity Ucavity, the equivalence ratio of the ignition in the mainstream zone Uignition-main and the total equivalence ratio in the combustor Utotal were defined as follows, respectively:
Ucav ity ¼ Q cav ity-fuel =ð0:23Lfuel Q total-air Þ
ð1Þ
Uignition-main ¼ Q main-fuel =ðLfuel Q total-air Þ
ð2Þ
Utotal ¼ ðQ main-fuel þ Q cav ity-fuel Þ=ðLfuel Q total-air Þ:
ð3Þ
The measurement accuracies of the orifice flowmeter, the coriolis mass flowmeter, the pressure sensor, and the thermocouple are 0.5%, 0.2%, 0.2%, and 1%, respectively. The component measurement accuracies of CO, CO2, NO, NO2, and O2 are 2%, 1.5%, 2.8%, 1%, and 1.2%, respectively. According to the uncertainty-analysis theory, the accuracies of the total equivalence ratio Utotal, the total pressure-loss coefficient r, the combustion efficiency gtotal, and the outlet-temperaturedistribution coefficient OTDF are 0.5%, 3.6%, 7.8%, and 11.2%, respectively [32]. 4. Flow and combustion-performance analysis 4.1. Performance of ignition and lean blowout
reversed. The results are shown in Fig. 12, and the total equivalence ratio of the ignition was between 0.19 and 0.22. The influence of the inlet Mach number on the ignition performance in the cavity zone is smaller. This is because double vortexes are formed in the cavity zone in the double-vortex combustor. The end surface of the igniter plug is positioned above the vortex. The influence of the variation of the mainstream velocity on the flow-field structure is small. The influence of the inlet temperature on the ignition performance of the cavity zone is large. With an increase in the inlet temperature, the reaction rate increases, and the equivalence ratio of the ignition decreases. Thus, the ignition performance is improved. As shown in Fig. 13, the total equivalence ratio of the blowout was between 0.04 and 0.09, and the equivalence ratio of the blowout in the combustor decreases with an increased inlet Mach number. This is contrary to the conventional experimental results for a general combustor. When the inlet velocity of the vortex combustor increases, the air velocity in the cavity zone increases. The corresponding fuel supply increases as well. Because the cavity zone has a double-vortex structure, when the inlet velocity increases, the flow velocity of the upper vortex slightly increases, and the fuel supply quantity increases. This leads to a large local equivalence ratio in the upper cavity zone. Thus, the performance of the lean blowout is improved. In addition, the increase of the air temperature of the inlet is beneficial to the improvement of the blowout performance. The ignition performance of the mainstream zone is shown in Fig. 14, and the equivalence ratio of the ignition in the mainstream zone was between 0.11 and 0.15. When the inlet temperature increases or the equivalence ratio of the cavity zone increases,
In the combustor, the ignition plug is only arranged in the cavity zone. In the test, the cavity zone was ignited first, and then the fuel supply in the mainstream zone was gradually increased until the mainstream zone was ignited. For blowout, the sequence was
Fig. 14. Ignition performance of the mainstream zone.
Fig. 16. Combustion efficiency of the vortex combustor (the cavity zone and the mainstream zone were fueled).
320
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
the equivalence ratio of the ignition in the mainstream zone decreases. This is because the increase in the temperature improves the rate of the chemical reaction. When the inlet Mach number increases, the equivalence ratio of the ignition of the mainstream zone decreases. This is because, when the velocity increases, the combustion products of the cavity zone are mixed more intensively with those in the mainstream zone. As a result, more combustion products in the cavity zone enter into the mainstream zone. Compared with the equivalence ratio of the ignition in the cavity zone, the equivalence ratio of the ignition in the mainstream zone is small. This indicates that the combustion products in the cavity zone have a good preheating effect on the inlet air in the mainstream zone. In addition, the equivalence ratio of the blowout in the mainstream zone is very small and close to zero, which is also due to the preheating effect of the combustion products of the cavity zone. 4.2. Combustion efficiency The gas-composition method was used to measure the combustion efficiency using Eq. (4) [30,36]:
gtotal ¼ ð½CO2 þ 0:531½CO 0:397½H2 0:319½CH4 Þ=ð½CO2 þ ½CO þ ½UHCÞ:
ð4Þ
When only the cavity zone is in combustion (Fig. 15), the combustion efficiency of the combustor was between 91% and 99%. With an increase in the equivalence ratio, the combustion efficiency first increases significantly and then decreases slightly. This phenomenon is due to the increase in the chemical-reaction rate as well as the increase in the equivalence ratio. The temperature of the evaporation tube increases, the evaporation rate of the fuel increases, and the combustion efficiency is improved. When the equivalence ratio reaches a certain value, the cavity zone enters into the fuel-rich combustion state. The equivalence ratio increases, and the combustion efficiency decreases. When the inlet temperature is low (Tinlet = 450 K), the combustion efficiency is low. When the inlet temperature increases to 550 K, the combustion efficiency increases. At this time, by further improving the inlet temperature, the efficiency does not change. This is because as the inlet temperature increases, the evaporation rate in the evaporation tubes increases, making the combustion more complete, and the combustion efficiency increases. However, the increased amplitude of the evaporation rate is limited. When the fuel is supplied to the two zones (Fig. 16), the combustion efficiency of the combustor was between 94% and 99%. At Ucavity = 0.6, the combustion efficiency is the lowest. When the
Fig. 17. Influence of the total equivalence ratio and Ma on the total pressure-loss coefficient.
equivalence ratio of the cavity zone increases from 0.6 to 0.7, the temperature of the combustion products in the cavity zone increases, thereby raising the temperature of the recirculation zone in the mainstream zone. The combustion in the mainstream zone becomes more complete. When the equivalence ratio of the cavity zone further increases to 0.8, the combustion efficiency of the combustor slightly increases. Under the same equivalence ratio of the cavity zone, when the Qmain-fuel/Qtotal-fuel ratio is high, the combustion efficiency is high. This indicates that, under stablecombustion conditions, the equivalence ratio of the cavity zone should be increased, and the amount of air in the cavity zone should be reduced. In addition, when the fuel is supplied to both zones and the equivalence ratio of the cavity zone remains unchanged, this study shows several effects. (1) When the inlet temperature of the mainstream zone increases, the combustion efficiency increases. (2) When the total equivalence ratio is small, the combustion efficiency increases significantly when the inlet Mach number increases. This is because, when the equivalence ratio is small and the combustion temperature is low, an increase in the inlet Mach number is in favor of the mixing of the fuel injected by the injector with the air in the inlet channel. This is conducive to an improvement in the combustion efficiency. (3) When the total equivalence ratio is relatively large, the combustion efficiency for every Mach number is relatively high. The influence of the Mach number on the combustion efficiency is not obvious. This is because, when the equivalence ratio is large, the combustion temperature is high, and the combustion of fuel is sufficient. 4.3. Total pressure loss The total pressure loss is an important index of the combustor performance. Usually, the total pressure-loss coefficient, r, is used for analysis:
r ¼ ðPinlet;average Poutlet;average Þ=Pinlet;average
ð5Þ
Because the effects of the staged method and the inlet temperature on the total pressure loss are very small, the effects of the Mach number and the total equivalence ratio are mainly analyzed below. As shown in Fig. 17, r of the combustor was between 0.04 and 0.11, and it is in direct proportion to the square of the Mach number. This is because, when the velocity increases, the flow resistance increases as well. When in the combustion state, r is approximately 0.015 higher than that in the cold state. This is because the heat resistance increases when the temperature increases.
Fig. 18. OTDF of the vortex combustor (the cavity zone and the mainstream zone were fueled).
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
(a) Ma=0.15, Tinlet=550K, Φcavity =0.6, Φtotal=0.37 (OTDF=0.36)
(b) Ma=0.15, Tinlet=550K, Φcavity =0.7, Φtotal=0.37 (OTDF=0.23) Fig. 19. Outlet temperature-distribution diagrams (the cavity zone and the mainstream zone were fueled).
321
is supplied to both zones, the result is shown in Fig. 18. Under the same equivalence ratio of the cavity zone, the total equivalence ratio has little effect on the OTDF. This indicates that, when the temperature of the combustor increases, the highest temperature of the outlet of the combustor increases as well. When the equivalence ratio of the cavity zone increases, the OTDF decreases significantly. This is because the equivalence ratio of the cavity zone increases, and the temperature of the combustion products in the cavity zone increases as well. When the temperature in the mainstream zone increases, the distance required by the combustion of the fuel in the mainstream zone is shortened. The combustion products have a longer distance to be mixed. In addition, when the inlet temperature increases, the OTDF decreases. The reason is similar to that of the influence of the equivalence ratio of the cavity zone on the OTDF. When the inlet Mach number increases, the OTDF increases. This is because, when the two zones are in combustion at the same time, an increase in the Mach number is not conducive to the mixing of the combustion products in the two combustion zones. The temperature distribution of the outlet becomes nonuniform. The temperature distribution diagrams of the outlet are shown in Fig. 19. When the equivalence ratio of the cavity zone is small, the high-temperature zone is mainly located on the right side, and the area is small. When the equivalence ratio increases to 0.7, the temperature of the low-temperature zone of the lower left side increases. The highest temperature at the outlet decreases, and the temperature distribution is more uniform. Thus, with the increase of the equivalence ratio of the cavity zone, the temperature of the combustion products in the two lower recirculation zones of the mainstream zone can be raised, and the temperature-distribution uniformity is improved. 5.2. Wall-temperature distribution A type k thermocouple was used to measure the temperature of the outer wall of the cavity. The measuring points were arranged at the central cross-section of test piece, with the specific arrangement shown in Fig. 20. The test result is shown in Fig. 21. In each state, the temperature distributions of the wall of the combustor are essentially identical. The temperatures in the upper front wall (measuring point 3) and the upper back wall (measuring point 5) are relatively high. This indicates that the heat exchange between the combustion products in the cavity zone and the cavity wall is mainly concentrated in these two regions. When the equivalence ratio of the cavity zone increases, the temperature of the cavity wall increases on average, and the variation of the temperature of wall in the mixing section is relatively small. The combustion in the cavity zone of the
Fig. 20. Positions of the measuring points of the wall temperature.
Fig. 21. Wall-temperature distribution of the cavity.
5. Temperature-distribution analysis 5.1. Outlet-temperature distribution The temperature-distribution uniformity of the outlet is an important index of the performance of the combustor. Usually, the outlet temperature-distribution factor, OTDF, is used to reflect the uniformity of the temperature field:
OTDF ¼ ðT outlet;max T outlet;average Þ=ðT outlet;average T inlet;average Þ:
ð6Þ
When the fuel is only supplied in the cavity zone, the temperature rise is small, and the temperature-distribution uniformity of the outlet is not important. The temperature-distribution uniformity of the outlet is researched under a large load. When the fuel
Fig. 22. Emissions of carbon monoxide (only the cavity zone was fueled).
322
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
6.1. Carbon monoxide emissions
Fig. 23. Emissions of carbon monoxide (the cavity zone and the mainstream zone were fueled).
double-vortex combustor is mainly in the upper vortex. When the combustion products flow to the lower vortex, the combustion products begin to be mixed with the air inflow from the channel. Then, the temperatures of wall at the lower back wall and the mixing section are relatively low, with a small variation in the equivalence ratio of the cavity zone. For the same equivalence ratio of the cavity zone, when the fuel is supplied to both zones, the temperature of the cavity wall is higher than that when the fuel is only supplied to the cavity zone. The combustion state of the mainstream zone influences the temperature distribution of the cavity wall. This indicates that part of the combustion products of the mainstream zone are entrained into the cavity zone, resulting in a high temperature of the cavity wall.
6. Pollutant-emissions analysis According to the relevant industry standards [36], the emission index (EI) can be calculated by the following formula:
EISpecies ¼ ð½Speciesð1 þ aC1 Þ=ð½CO þ ½CO2 þ ½UHCÞÞ 3
ð10 MSpecies =MC þ n=mMH Þ
ð7Þ
where C1 = (2C2 (n/m))/(4(1 + n (aC2/2))), MCO = 28.011, MC = 12.011, MH = 1.008, MUHC = 16.043, MNO2 = 46.008, C2 = (2 [CO] ((2/x) (y/2x))[UHC]+[NO2])/([CO]+[CO2]+[UHC]), x = 1, y = 4, m = 10, n = 20, and a = 0.00032.
When the fuel is supplied only to the cavity zone, the result is shown in Fig. 22. With an increased equivalence ratio, EICO gradually decreases. This occurs because, when the temperature of the combustion products in the combustion zone increases, the combustion becomes complete. When the equivalence ratio is greater than 1, EICO increases slightly. When the fuel is supplied to both zones, as shown in Fig. 23, with an increased total equivalence ratio, EICO gradually decreases. The reason is the same as above. Compared with the situation in which the fuel is only supplied to the cavity zone, EICO for the situation in which the fuel is supplied to both zones is far smaller, given the same total equivalence ratio. This shows that, when the fuel is supplied to both zones, a large amount of CO produced by the combustion in the cavity zone becomes CO2 in the mainstream combustion zone due to oxidation. In addition, when the fuel is supplied to both zones, the following results are obtained for the constant equivalence ratio of the cavity zone. (1) With increased inlet temperature of the combustor, EICO decreases. (2) When the total equivalence ratio is small, the inlet Mach number increases, and EICO decreases significantly. (3) When the total equivalence ratio is relatively large, the EICO for each Mach number is small, and the influence of the Mach number is not obvious. The reason is similar to that for the variation of the combustion efficiency. 6.2. Nitrogen oxide emissions Because NOx is mainly produced under high load conditions, the situation in which the fuel is supplied to both zones is analyzed, as shown in Fig. 24. With an increase in the total equivalence ratio, EINOx increases, which is due to the increase of the flame temperature in the mainstream combustion zone. When the Mach number increases, EINOx increases. This is because the increase in the combustion efficiency leads to the increase in the flame temperature. 6.3. Unburned hydrocarbon emissions The EIUHC in the combustor is in the range of 1–30 g/(kg fuel). Its variation pattern is consistent with that of EICO but with a smaller value. This is because the generation rule of unburned hydrocarbons is similar to that of CO. However, the unburned hydrocarbons are more likely to be oxidized, and the values of the emissions are smaller. 7. Numerical analysis
Fig. 24. Emissions of nitrogen oxides (the cavity zone and the mainstream zone were fueled).
A 30° fan-shaped area was selected for modeling, and an unstructured mesh system consisting of 3.2 million cells (illustrated in Fig. 25) was used. Mesh refinement was specifically conducted in the regions of the air-inlet holes and the cavity zone. Under steady-state conditions, a numerical simulation was performed using the k–e turbulence model. The number of the fuel injection point and the injection direction of the fuel are consistent with the entity. The velocity distribution in the cross-section with the crossflame plate is shown in Fig. 26. A vortex exists in the upper right corner of the cavity zone, and a low-velocity region is formed in the rear of the cross-flame plate. The velocity distribution in the cross-section without the cross-flame plate is shown in Fig. 27. There are vortices in the upper right corner, the upper left corner and the lower right corner of the cavity zone. A recirculation zone is formed in the rear of the mainstream holder, and the zone is extended to the location of the air injection from the mixing holes.
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
323
Fig. 25. Schematic diagram of the unstructured mesh system.
Fig. 26. Velocity distribution in the cross-section with the cross-flame plate.
Fig. 29. Temperature distribution in the cross-section without the cross-flame plate.
Fig. 27. Velocity distribution in the cross-section without the cross-flame plate.
Fig. 30. Comparisons of r and the OTDF between the numerical simulations and the experiments.
Fig. 28. Temperature distribution in the cross-section with the cross-flame plate.
Temperature-distribution diagrams in the cross-section of combustor are shown in Figs. 28 and 29. In the cross-section with the cross-flame plate, there are high-temperature zones in the lower right corner of the cavity zone and the lower area of the recirculation zone in the mainstream zone. In the cross-section without the cross-flame plate, there are high-temperature zones in the upper
left corner of the cavity zone and the lower area of the mainstream zone, and the mainstream combustion zone is strongly separated from the cavity combustion zone by the air jet of mainstream. The cross-flame plates enhance the energy transport between the two combustion zones and improve the radial temperature distribution of the outlet. Comparisons of r, OTDF, EICO and EINOx between the numerical simulations and the experiments are shown in Figs. 30 and 31. The numerical results correspond to the experimental results. The total equivalence ratio of the combustor has a large influence on the EICO and EINOx values. Thus, it is important to control the temperature of the combustion zone in an appropriate range.
324
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325
References
Fig. 31. Comparisons of EICO and EINOx between the numerical simulations and the experiments.
Therefore, the flowrate of the combustion zone should be exactly considered in the design process of the combustor, based on the operating conditions of the combustor, especially the total equivalence ratio.
8. Conclusions An experimental and numerical investigation of a single-cavity vortex combustor fueled with aviation kerosene is presented. A double-vortex combustion mode is adopted in the pilot combustion zone of the staged combustor, and the jets of the fuel–air mixture from the evaporation tube enhance the stability of the vortex in the pilot zone. The combustion mode of the double recirculation zone is adopted in the mainstream combustion zone, and the fuel injected by the pneumatic atomization injector is mixed with mainstream air before entering the mainstream zone. The cross-flame plates are located in the upper site of the flame holder to enhance the energy transport between the pilot and the mainstream zone. The results of the combustion experiment show that the combustion mode of the double vortex with a dual channel has good comprehensive performance. The combustion organization method in the combustion zone is reasonable. A stable double vortex can be formed in the pilot zone. At the same time, the preheating effect of the evaporation tube causes the fuel to have high-quality atomization. Due to the rapid fuel–air mixing and the high stability of the double-vortex, stable combustion has been demonstrated over a wide range of total equivalence ratios. Under typical operating conditions, high combustion performance with low pollutant emissions was obtained, and the outlet-temperature distribution of the combustor was uniform, which is very important for a single-cavity combustor with an asymmetric structure. The influence of each factor on the combustion performance and pollutant emissions is very complex. Among these factors, the inlet velocity has little effect on the performance of combustion. This is because the inlet velocity has little effect on the upper vortex in the cavity zone. The staged method has a significant effect on the multiple performance of the combustor, which is due to the complex mass and energy transport between the two combustion zones. Acknowledgement This work was supported by the Postdoctoral Science Foundation of China (2013M530514). The support is gratefully acknowledged.
[1] Lefebvre AH, Ballal DR. Gas turbine combustion. Taylor & Francis Group; 2010. [2] Roquemore WM, Shouse D, Burrus D, et al. Vortex combustor concept for gas turbine engines. In: 39th aerospace sciences meeting and exhibit, AIAA 20010483; 2001. http://dx.doi.org/10.2514/6.2001-483. [3] Ballester J, Sanz A, Gonzalez MA. Investigation of the characteristics and stability of air-staged flames. Exp Therm Fluid Sci 2008;32:776–90. http:// dx.doi.org/10.1016/j.expthermflusci.2007.09.007. [4] Khalil AEE, Gupta AK. Distributed swirl combustion for gas turbine application. Appl Energy 2011;88:4898–907. http://dx.doi.org/10.1016/j.apenergy.2011. 06.051. [5] Khalil AEE, Gupta AK. Fuel flexible distributed combustion for efficient and clean gas turbine engines. Appl Energy 2013;109:267–74. http://dx.doi.org/ 10.1016/j.apenergy.2013.04.052. [6] Arghode VK, Gupta AK. Role of thermal intensity on operational characteristics of ultra-low emission colorless distributed combustion. Appl Energy 2013;111:930–56. http://dx.doi.org/10.1016/j.apenergy.2013.06.039. [7] Arghode VK, Khalil AEE, Gupta AK. Fuel dilution and liquid fuel operational effects on ultra-high thermal intensity distributed combustor. Appl Energy 2012;95:132–8. http://dx.doi.org/10.1016/j.apenergy.2012.02.020. [8] Dhanuka SK, Temme JE, Driscoll JF, et al. Vortex-shedding and mixing layer effects on periodic flashback in a lean premixed prevaporized gas turbine combustor. Proc Combust Inst 2009;32:2901–8. http://dx.doi.org/10.1016/ j.proci.2008.06.155. [9] Bucher J, Edmonds RG, Steele RC, et al. The development of a lean-premixed trapped vortex combusto. In: Proceedings of ASME turbo expo; 2003. http:// dx.doi.org/10.1115/GT2003-38236. [10] Giuliani F, Woisetschläger J, Leitgeb T. Design and validation of a burner with variable geometry for extended combustion range. ASME Turbo Expo 2012. http://dx.doi.org/10.1115/GT2012-68236. [11] Straub DL, Casleton KH, Lewis RE, et al. Assessment of rich-burn, quick-mix, lean-burn trapped vortex combustor for stationary gas turbines. J Eng Gas Turb Power 2005;127:36–41. http://dx.doi.org/10.1115/1.1789152. [12] Samuelsen GS, Brouwer J, Vardakas MA, et al. Experimental and modeling investigation of the effect of air preheat on the formation of NOx in an RQL combustor. Heat Mass Transf 2013;49:219–31. http://dx.doi.org/10.1007/ s00231-012-1080-0. [13] Dodds W. Twin annular premixing swirler (TAPS) combustor. In: The roaring 20th aviation noise & air quality symposium; 2005. [14] Hsu KY, Goss LP, Roquemore WM. Characteristics of a trapped- vortex combustor. J Propul Power 1998;14:57–65. http://dx.doi.org/10.2514/2.5266. [15] Bruno C, Losurdo M. The trapped vortex combustor: an advanced combustion technology for aerospace and gas turbine applications. In: Syred N, Khalatov A, editors. Advanced combustion and aerothermal technologies. Springer; 2007. http://dx.doi.org/10.1007/978-1-4020-6515-6_28. [16] Ezhil Kumar PK, Mishra DP. Numerical simulation of cavity flow structure in an axisymmetric trapped vortex combustor. Aerosp. Sci. Technol. 2012;21:16–23. http://dx.doi.org/10.1016/j.ast.2011.04.007. [17] Ezhil Kumar PK, Mishra DP. Numerical investigation of the flow and flame structure in an axisymmetric trapped vortex combustor. Fuel 2012;102:78–84. http://dx.doi.org/10.1016/j.fuel.2012.06.056. [18] Vengadesan S, Sony C. Enhanced vortex stability in trapped vortex combustor. Aeronaut J 2010;114:333–7. [19] Agarwal KK, Krishna S, Ravikrishna RV. Mixing enhancement in a compact trapped vortex combustor. Combust Sci Technol 2013;185:363–78. http:// dx.doi.org/10.1080/00102202.2012.721034. [20] Szwaja S, Jamrozik A, Tutak W. A two-stage combustion system for burning lean gasoline mixtures in a stationary spark ignited engine. Appl Energy 2013;105:271–81. http://dx.doi.org/10.1016/j.apenergy.2012.12.080. [21] Ghenai C, Zbeeb K, Janajreh I. Combustion of alternative fuels in vortex trapped combustor. Energy Convers Manage 2013;65:819–28. http://dx.doi.org/ 10.1016/j.enconman.2012.03.012. [22] Jin Y, He X, Zhang J, et al. Experimental study on emission performance of an LPP/TVC. Chin J Aeronaut 2012;25:335–41. http://dx.doi.org/10.1016/S10009361(11)60394-4. [23] Fan WJ, Yan M, Yi Q, et al. Low NOx emission of rich-burn, quick-mix, leanburn trapped vortex combustor. J Propul Technol 2006;27:88–91. [24] Singhal A, Ravikrishna RV. Single cavity trapped vortex combustor dynamics – Part-1: Experiments. Int J Spray Combust Dynam 2011;3:23–44. http:// dx.doi.org/10.1260/1756-8277.3.1.23. [25] Nemitallah MA, Habib MA. Experimental and numerical investigations of an atmospheric diffusion oxy-combustion flame in a gas turbine model combustor. Appl Energy 2013;111:401–15. http://dx.doi.org/10.1016/j. apenergy.2013.05.027. [26] Payri R, Salvador FJ, Gimeno J, et al. Influence of injector technology on injection and combustion development–Part 2: Combustion analysis. Appl Energy 2011;88:1130–9. http://dx.doi.org/10.1016/j.apenergy.2010.10.004. [27] Khalil AEE, Gupta AK. Swirling distributed combustion for clean energy conversion in gas turbine applications. Appl Energy 2011;88:3685–93. http:// dx.doi.org/10.1016/j.apenergy.2011.03.048. [28] Syred N, Giles A, Lewis J, et al. Effect of inlet and outlet configurations on blowoff and flashback with premixed combustion for methane and a high hydrogen content fuel in a generic swirl burner. Appl Energy 2014;116:288–96. http:// dx.doi.org/10.1016/j.apenergy.2013.11.071.
R.C. Zhang et al. / Applied Energy 130 (2014) 314–325 [29] Xing F, Wang P, Zhang S, et al. Experiment and simulation study on lean blowout of trapped vortex combustor with various aspect ratios. Aerosp Sci Technol 2012;18:48–55. http://dx.doi.org/10.1016/j.ast.2011.03.010. [30] Xing F, Zhang S, Wang P, et al. Experimental investigation of a single trappedvortex combustor with a slight temperature raise. Aerosp Sci Technol 2010;14:520–5. http://dx.doi.org/10.1016/j.ast.2010.04.007. [31] Zhang R, Fan W, Song S. Atomization and evaporation performance experiment of evaporating tube for trapped vortex combustor. J Aerosp Power 2011;26:2495–502. [32] Zhang R. Fundamental investigation of trapped vortex combustor with high temperature rise. PhD thesis, Beijing University of Aeronautics and Astronautics; 2012.
325
[33] Mundo CHR, Sommerfeld M, Tropea C. Droplet-wall collisions: experimental studies of the deformation and breakup process. Int J Multiphase Flow 1995;21:151–73. http://dx.doi.org/10.1016/0301-9322(94)00069-V. [34] Yarin AL. Drop impact dynamics: splashing, spreading, receding, bouncing. Annu Rev Fluid Mech 2006;38:159–92. http://dx.doi.org/10.1146/annurev. fluid.38.050304.092144. [35] Zhai X, Peng R, Fan W, et al. Numerical simulation on a certain trapped vortex combustor. J Aerosp Power 2013;5:1134–41. [36] Aviation industry department of China, Aero-engine gaseous pollutants sampling and measurement procedures 1987, Aviation industry standard HB 6117-87.