Energy Fuels 2010, 24, 1603–1610 Published on Web 02/23/2010
: DOI:10.1021/ef901286r
Influence of the Staged-Air Declination Angle on Flow-Field Deflection in a Down-Fired Pulverized-Coal 300 MWe Utility Boiler with Direct-Flow Split Burners Min Kuang, Zhengqi Li,* Yunfeng Han, Lianjie Yang, Qunyi Zhu, Jia Zhang, and Shanping Shen School of Energy Science and Engineering, Harbin Institute of Technology, 92 West Dazhi Street, Harbin 150001, People’s Republic of China Received November 2, 2009. Revised Manuscript Received February 8, 2010
Cold airflow experiments were conducted within a small-scale furnace of a down-fired pulverized-coal 300 MWe utility boiler. With focus on the large combustion difference between the zones near the front and rear walls in down-fired pulverized-coal boilers, we investigate the aerodynamic field at different staged-air declination angles of 0°, 15°, 30°, 45°, and 55°. For declination angles of 0°, 15°, and 30°, a deflected flow field appeared in the lower furnace, with downward airflow velocities near the rear wall decaying more rapidly than velocities near the front wall. In addition to the downward airflow reach into the lower furnace, the turbulence intensity and longitudinal-velocity components at certain cross sections were lower near the rear wall than near the front wall. Through an increase of the declination angle from 0° to 30°, the flow-field deflection diminished, which was accompanied by a slower decay in the downward airflow near the rear wall and an increase in the reach (as measured by the dimensionless depth) of the downward airflow near the rear wall as well as longitudinal-velocity components within the associated cross section. Those near the front wall changed only slightly. For larger angles of 45° and 55°, the deflected flow field disappeared. Turbulence intensities in the staged-air zones near the front and rear walls increased steadily as the declination angle increased from 0° to 55°. The optimal setting for staged air would necessitate a declination angle of 45°.
boilers, while the ratios for FW boilers are about 30% on arches and 70% below arches. Aerodynamic characteristics are of great importance in pulverized-coal combustion. Performing experiments with small-scale models is a widely popular means to obtaining approximate aerodynamic characteristics with boilers.1-5 Li et al. have studied aerodynamic characteristics in a small-scale model furnace for a FW down-fired boiler.6,7 Zhou et al. have investigated the effect of the swirl number of secondary air on the aerodynamic characteristics of a B&W down-fired boiler.8 Che et al. have focused on the influences of momentum ratios among different tiers of secondary air and burner’s angle on the whole furnace flow field for a Stein-designed down-fired boiler.9,10 However, little research has been reported on MBEL down-fired boilers, which suffer similarly from problems of large differences in combustion between the zones near the front and rear walls, high carbon content in the fly
1. Introduction Down-fired boilers are designed to burn anthracite and lean coal. Four types of down-fired utility boilers are manufactured: the Foster Wheeler (FW), the Babcock & Wilcox (B&W), the Stein, and the Mitsui Babcock Energy Limited (MBEL) down-fired boilers. The major differences between them are the type and arrangement of the burner and air distribution. FW boilers are equipped with direct-flow pulverized-coal burners enriched by double cyclones and circular primary air nozzles, while B&W boilers employ swirl pulverized-coal burners. Stein boilers utilize direct-flow split pulverized-coal burners without enrichment of the primary air/fuel mixture flow. MBEL boilers also use direct-flow split pulverized-coal burners but with the primary air/fuel mixture flow enriched by cyclones. With respect to the arrangement of the burners and air distribution, MBEL boilers have an uneven grouping on arches, but the other three types have burners uniformly arranged on arches along the furnace width. FW boilers have fuel-rich and flow-lean flows arranged on arches and secondary air injected from the front and rear walls into the furnace. For both B&W and Stein boilers, the primary air/ fuel mixture flow and secondary air are arranged on arches, while the vent air and staged air are sent into the furnace from the front and rear walls. However, fuel-rich and flow-lean flows and secondary air for MBEL boilers are all arranged on arches, and only staged air is sent into the furnace from the front and rear walls. Air mass flux ratios are about 80% on arches and 20% below arches for B&W, Stein, and MBEL
(1) Zhou, Y. G.; Xu, T. M.; Hui, S. E.; Zhang, M. C. Appl. Therm. Eng. 2009, 29, 732–739. (2) Li, Y. Q.; Zhou, H. C. Flow Meas. Instrum. 2006, 17, 113–122. (3) He, B. S.; Chen, M. Q.; Liu, S. M.; Fan, L. J.; Xu, J. Y.; Pan, W. P. Exp. Therm. Fluid Sci. 2005, 29, 537–554. (4) Zhou, Y. G.; Zhang, M. C.; Xu, T. M.; Hui, S. E. Energy Fuels 2009, 23, 5375–5382. (5) Lin, Z. C.; Fan, W. D.; Li, Y. Y.; Li, Y. H.; Zhang, M. C. Energy Fuels 2009, 23, 744–753. (6) Ren, F.; Li, Z. Q.; Chen, Z. C.; Wang, J. J.; Chen, Z. Energy Fuels 2009, 23, 2437–2443. (7) Ren, F.; Li, Z. Q.; Zhang, Y. B.; Sun, S. Z.; Zhang, X. H.; Chen, Z. C. Energy Fuels 2007, 21, 668–676. (8) Zhou, Z. J.; Zhu, Z. L.; Zhao, X.; Yao, Q.; Cao, X. Y.; Cen, K. F. Power Eng. 1999, 19, 33-37 (in Chinese). (9) Che, G.; He, L. M.; Hui, S. E.; Xu, T. M. Xi’an Jiaotong Daxue Xuebao 2000, 34, 38-43 (in Chinese). (10) Che, G.; He, L. M.; Xu, T. M.; Hui, S. E. Power Eng. 2001, 21, 1132-1136 (in Chinese).
*To whom correspondence should be addressed. Tel.: þ86 451 8641 8854. Fax: þ86 451 8641 2528. E-mail:
[email protected]. r 2010 American Chemical Society
1603
pubs.acs.org/EF
Energy Fuels 2010, 24, 1603–1610
: DOI:10.1021/ef901286r
Kuang et al.
Figure 1. Vertical and horizontal cross sections through the furnace and combustion system of the down-fired boiler (dimensions in millimeters).
ash, and serious slagging in the dry bottom hopper. Li et al. have conducted industrial tests on a 300 MWe MBEL boiler and reported large differences in gas temperatures, heat fluxes, and gas species concentrations between the regions near the front and rear walls: (1) Gas temperatures in the region near the front wall are significantly lower than those near the rear wall, resulting in high carbon content in the fly ash due to poor burnout in the region near the front wall and serious slagging in the dry bottom hopper near the rear wall. (2) The O2 (respectively NOx) content is much lower (respectively higher) in the region near the rear wall than in that near the front wall.11 In a MBEL down-fired boiler, staged air not only reduces NOx emissions but also adjusts the penetration depth of primary air. This has a great impact on the flow and combustion characteristics in the lower furnace. Li et al. have investigated the effect of the staged-air ratio on the aerodynamic characteristics of a 300 MWe MBEL boiler and reported that, when staged air is injected horizontally into the furnace with a large transverse-momentum component, a high staged-air ratio is one major factor favoring the formation of flow-field deflection.11 The transverse-momentum component decreases when staged air was injected into the furnace at a certain declination angle. With focus on the above
problems, cold airflow experiments with a small-scale furnace of a MBEL down-fired pulverized-coal 300 MWe utility boiler were conducted to investigate the aerodynamic field at different staged-air declination angles. 2. Experimental Facility Setup Vertical and horizontal cross sections of the furnace showing the layout of the combustion system are presented in Figure 1. The arches separate the furnace into two regions, the rectangular upper furnace and the octagonal lower furnace with four wing walls. W1 and W2 denotes the widths of the lower and upper furnace, respectively. L1 is the height of the upper furnace. A total of 16 cyclones symmetrically arranged on the arches divide the primary air/fuel mixture into fuel-rich and fuel-lean flows needed to regulate the fuel-bias combustion. The fuel-rich flow is channeled through nozzles centered over the furnace, while the fuellean flow is injected through nozzles near the front and rear walls. There are eight burners lining the front and rear arches. Four fuelrich-flow nozzles, four fuel-lean-flow nozzles, and eight secondary-air ports fed each burner. The air box is partitioned into two parts, one on and the other below the arches. Each part has eight small boxes, with each box on the arches corresponding to one of the burners and each box below the arches feeding a group of staged-air ports. Near-wall air is partitioned air fed to the air box on the arches. The distribution of secondary air and staged air is adjustable by damper openings associated with each box. The experimental system is shown in Figure 2. It consists of an induced-draft fan, a small-scale furnace model, and an IFA300
(11) Li, Z. Q.; Kuang, M.; Zhang, J.; Han, Y. F.; Zhu, Q. Y.; Yang, L. J.; Kong, W. G. Environ. Sci. Technol. 2010, 44, 1130–1136.
1604
Energy Fuels 2010, 24, 1603–1610
: DOI:10.1021/ef901286r
Kuang et al.
Figure 2. Small-scale cold experiment system and layout of the measurement points (dimensions in millimeters).
constant-temperature anemometer. The small-scale model furnace is a 1:15-scaled version of the original. X0 is the horizontal distance between the front and rear walls in the lower furnace. H0 is the vertical distance between the outlet of the fuel-rich-flow nozzle and the upper edge of the dry bottom hopper, while H is the vertical distance from the measurement point to the outlet of the fuel-rich-flow nozzle. All airflow rates into the small-scale furnace were measured by Venturi tube flowmeters. Measurement errors were less than 10%. An IFA300 constant-temperature anemometer system was used to measure the air velocity at various locations within the furnace. The constant-temperature anemometer comprises a bridge and amplifier circuit that controls a thin-film sensor at constant temperature. As the fluid flow passes over the heated sensor, the amplifier senses the bridge offbalance and adjusts the voltage to the top of the bridge, keeping the bridge in balance. The voltage on the top of the bridge can therefore be related to the flow velocity. The bridge voltage is sensitive to the temperature as well as the velocity, and so the built-in thermocouple circuit is attached to a thermocouple to measure the fluid temperature. This temperature reading is then used to correct the results, minimizing the effect of temperature. Here, a 1240-type two-dimensional probe with two hot-film sensors was used, giving a velocity measurement error of less than 5%. Although the flow field in the lower furnace in a downfired boiler is very complex and three-dimensional, the velocity along the furnace width direction is relatively small. Here, cold airflow experiments focus on the W-shaped flow field along the vertical cross section through the lower furnace. Thus, a twodimensional probe was employed. The velocity component calculation follows the parallelogram law of vector addition. The flows in the model furnace are self-modeling, and the momentum ratios among the airflows of the small-scale furnace are the same as those of the full-scale furnace. The influence of the staged-air declination angle on airflow fields is investigated under conditions of constant momentum ratios between airflows at a room temperature of 20 °C. Here, the staged-air declination angle is the angle between the staged-air and horizontal directions. Airflow velocity parameters are listed in Table 1. At these
Table 1. Air Velocities for the Boiler Operation and the Small-Scale Airflow Experiments air velocity (m/s)
fuel-rich flow fuel-lean flow secondary air near-wall air staged air
boiler operation
small-scale airflow experiments
9.50 20.00 28.00 15.00 30.00
12.80 20.58 24.98 12.58 28.11
velocities, the staged-air declination angle was adjusted in turn to 0°, 15°, 30°, 45°, and 55°.
3. Results and Discussion Cold aerodynamic field velocities were measured along the longitudinal cross section intersecting the vertical center line of one of the fuel-rich-flow nozzles. In support of our claim of good reproducibility, we cite the example of the longitudinalvelocity component along the cross section H/H0 = 0.16 when readings were taken at different times at a staged-air declination angle of 0°. Here Vy is the velocity component along the y direction. V0 signifies the outlet velocity of the fuel-rich-flow nozzle along the y direction. X is the distance from the furnace center to the measurement points along the x direction (see Figure 2). As shown in Figure 3, excellent reproducibility is demonstrated. In a down-fired boiler, the lower furnace constitutes the combustion zone and the upper furnace is the burnout zone. The pulverized-coal combustion process mainly depends on the flow field in the lower furnace because pulverized-coal combustion is almost complete in the lower furnace. Thus, only the cold aerodynamic flow in the lower furnace is shown. As shown in Figure 4, for declination angles of 0°, 15°, and 30°, a deflected flow field appears in the lower furnace. After 1605
Energy Fuels 2010, 24, 1603–1610
: DOI:10.1021/ef901286r
Kuang et al.
Figure 5. Longitudinal-velocity component profiles along different cross sections for various staged-air declination angles. Figure 3. Longitudinal-velocity component along the cross-section H/H0 = 0.16 for a staged-air declination angle of 0° taken at different times.
declination angles. The cross section at H/H0 = 0.160 below the arch is away from any staged air, while that at H/H0 = 0.394 lies in the middle of the airflow zone of the staged-air ports (see Figure 2). As shown in Figure 5a, from the front and rear walls to the furnace center, the dimensionless longitudinal-velocity components Vy/V0 at the cross section H/H0 = 0.160 increase at a high rate initially for all five declination angles. However, they decrease quickly after reaching maxima at X/X0 = -0.42 to -0.38 near the front wall and X/X0 = 0.38 near the rear wall. Relatively little change occurs in the region X/X0 = -0.3 to -0.2. Longitudinal-velocity components are relatively symmetric near the front and rear walls in the regions X/X0 = -0.5 to -0.35 and þ0.35-0.5. For declination angles of 0°, 15°, and 30°, airflows near the front and rear walls begin to deflect upward in the zone near X/X0 = -0.3 and 0.2, respectively, with a poor symmetrical distribution of the longitudinal-velocity component between the zones near the front and rear walls. This is because upward airflow deflects toward the front wall and a large recirculation zone exists below the rear arch in a deflected flow field (see Figure 4a-c). With no airflow field deflection at angles of 45° and 55°, airflows near the front and rear walls turn upward near cross sections X/X0 = -0.3 and þ0.3, respectively, with a more symmetrical distribution of the longitudinal-velocity components between the zones near the front and rear walls. As shown in Figure 5b, the longitudinal-velocity components from the front and rear walls to the furnace center at the cross section H/H0 = 0.394 decrease rapidly for all declination angles, finally attaining negative values from which they change little. The exception is the zone near the rear wall where Vy/V0 is close to 0 in the range X/X0 = 0.2-0.5 with the staged-air declination angle set at 0°. For declination angles of 0°, 15°, and 30°, airflows near the front and rear walls begin to deflect upward in the zone near X/X0 = -0.3 and þ0.2. Velocities Vy/V0 near the rear wall are lower than those near the front wall and as the declination angle increases and the difference decreases because of an increase in the former and little variation in the latter. This occurs because of the formation of a deflected airflow field. In such an eventuality, the downward airflow near the front wall is forced closer to and penetrates further into the dry bottom hopper while slightly changing its longitudinal-velocity component. Meanwhile, the airflow near the rear wall rises as the declination angle increases. For larger angles of 45° and 55°, the distribution of Vy/V0 is essentially symmetric along the furnace center in the zones near the front and rear walls. A distinct difference can be seen in the profiles in Figure 5a, b with the appearance of a dual peak in the former but not in the latter. The reason is not to be found in the boundary layers near the front and rear walls but in the different distances from
Figure 4. Aerodynamic fields for cases with different staged-air declination angles.
penetration of the upper part of the dry bottom hopper region, the airflow near the rear wall deflects up toward the zone near the front wall and occupies the central area in the lower furnace. Deflections suppress the downward airflow near the front wall, forcing it closer toward and penetrating further into the dry bottom hopper. A large recirculation zone appears in the lower furnace below the rear arch. Repetition of measurements for one setting demonstrated that the flowfield deflection is steady. With the staged-air deflection angle set at 0°, the downward airflow near the rear wall deflects upward toward the zone near the front wall after reaching the region above the staged air. Moreover, a recirculation zone appears in the region between the staged-air zone and the outlet of the lower furnace. As the angle increases to 15° and 30°, the flow-field deflection weakens to some extent and increases the reach of the downward airflow near the rear wall. In addition, the recirculation zone shrinks and draws near the upper region of the dry bottom hopper. For larger angles of 45° and 55°, the deflected flow field disappears from the lower furnace. Figure 5 presents the longitudinal-velocity component profiles along different cross sections for different staged-air 1606
Energy Fuels 2010, 24, 1603–1610
: DOI:10.1021/ef901286r
Kuang et al.
Figure 7. Sketch of the directions of mixed jet and declined angles of mixed air near the staged-air ports.
mixing is strong in the staged-air zone near the front wall and the turbulence intensities are high because of suppression of the deflected upward flow (see Figure 4a-c). With an increase in the declination angle, the reach of the downward airflow near the rear wall increases, enhancing mixing in the staged-air zone. For declination angles of 45° and 55°, maxima in the turbulence intensities are found at X/X0 = -0.35 and þ0.38, respectively, and minima at X/X0 = 0.05 near the furnace center, with only a small difference between the zones near the front and rear walls. Thus, at declination angles of 45° and 55°, the distribution of the turbulence intensity between these two zones at the cross section H/H0 = 0.394 is much more symmetric than that at smaller angles of 0°, 15°, and 30°. The staged-air and downward coal/airflow mix intensely and produce violent fluctuations that enhance heat and mass transfer in the combustion process, providing efficient burnout. Figure 6 also shows that the turbulence intensities rise as the declination angles increase. This is because a small declination angle corresponds to relatively large transversemomentum components of staged air in the upper half of the staged-air zone that removes most of the momentum from the downward airflow. Thus, mixing and accompanying fluctuations are less intense at this cross section, leading to relatively weak turbulence intensities. Conservation of momentum is applicable in calculating the final direction of the mixed airflow given two impinging airflows. If we suppose β is the angle between the main and mixed airflows, then the following relation can be obtained from the parallelogram law of vector addition: sin R β ¼ arctan M1 ð2Þ þ cos R M2
Figure 6. Turbulence intensity distribution in the cross section H/H0 = 0.394 for various staged-air declination angles.
the two cross sections to the fuel-rich-flow nozzle outlet. In the regions near the front and rear walls in the small-scale furnace, the minimum distance between the wall and the measurement point is 34 mm (see Figure 2), which is far greater than that of the boundary layers near the front and rear walls. Thus, the longitudinal-velocity component should not be influenced by the boundary layers. At small distances away from the fuelrich-flow nozzle outlet, near the burner outlet, the downward airflow expands insufficiently at the cross section H/H0 = 0.160 in Figure 5a. The downward velocities are high in the region below the secondary-air port outlet and low in the region near the wall and furnace center. Thus, velocities at this cross section exhibit a dual peak distribution. However, at the cross section H/H0 = 0.394, which is far away from the fuelrich-flow nozzle outlet in Figure 5b, primary air and secondary air mixes well and the downward airflow expands sufficiently. Affected by the declined staged-air ejection, the longitudinal-velocity component is high in the region near the wall and low near the furnace center, without a dual peak distribution. Figure 6 presents the turbulence intensity distribution at the cross section H/H0 = 0.394 along the x direction for the five angular settings. Turbulence intensity T is defined as sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi V 0x 2 þ V 0y 2 2 ð1Þ T ¼ Vm where Vx0 and Vy0 are velocity fluctuations in the x and y directions at a measurement point, respectively.6 Vm is the mean velocity at the measurement point. With angle settings of 0°, 15°, and 30°, turbulence intensities near the front wall are observed to be higher than those near the rear wall. With an increase in the declination angle, differences in the turbulence intensities in the zones near the front and rear walls decrease. From the front and rear walls to the furnace center, the turbulence intensities increase at a high rate initially but then decrease rapidly after attaining maxima at X/X0 = -0.3 to -0.25 near the front wall and X/X0 = 0.25-0.3 near the rear wall. A minimum value is located at X/X0 = 0.15, which is nearing the rear wall. This is also a consequence of the presence of the deflected airflow field in the furnace. The downward airflow near the rear wall is blocked from penetrating further after reaching the staged-air zone, creating a weak mixing zone between the downward flow and staged air with a relatively low turbulence intensity. Simultaneously,
where R is the angle between the main and impact airflows having momentum M1 and M2, respectively (see Figure 7a). Here the downward airflow meets with staged air near the front and rear walls to form the mixed airflow. The main airflow is the downward airflow consisting of primary air and secondary air. Before impingement with staged air, the direction of the downward airflow is thought to be vertical because of both primary air and secondary air are vertically ejected downward into the furnace. Thus, the angle R is the complement of the staged-air declination angle. M1 is the momentum flow ratio of the downward airflow reaching the cross section H/H0 = 0.394, obtained through the following expressions: ð3Þ M1 ¼ mv v ¼ m=Fs 1607
Energy Fuels 2010, 24, 1603–1610
: DOI:10.1021/ef901286r
Kuang et al.
Figure 8. Decay curves for longitudinal-velocity components of the fuel-rich flow at different staged-air declination angles.
where m is the total mass flow ratio of primary air and secondary air. v denotes the mean velocity of the downward airflow in the regions near the front and rear walls at the cross section. s is the valid flow area for the downward airflow at the cross section, and F is the air density at 20 °C. Because of the large distance between the cross section and the burner nozzle outlet, primary air and secondary air mix evenly and the downward airflow carries relatively high longitudinal-velocity components in the region from the wall to the fourth measurement point (see Figures 4 and 5). Thus, this region at the cross section is thought to be the valid flow area for the downward airflow, which can be calculated with the size of the small-scale furnace in Figure 2. M2 is the momentum flow ratio of staged air, which is obtained from the product of the mass flow ratio and the velocity of staged air. The angle between the mixed airflow direction near the staged-air ports and the horizontal is defined as the declined angle, which can be calculated according to the above expressions (2) and (3). Figure 7b presents the calculated and measured declined angles of the mixed airflow in the staged-air zone. With an increase in the staged-air declination angle, the calculated declined angles show a linear increase. The measured declined angle near the rear wall shows a similar trend and increases more rapidly than the calculated value in the range of 0-30°, but thereafter, there was a slight decline from 30° to 55°. When the staged-air declination angle is set at 0°, the measured declined angle near the rear wall is seen to be lower than the calculated value. This is because the downward airflow near the rear wall cannot penetrate into the staged air (see Figure 4a), which holds the largest transverse momentum at this angle. Measured declined angles near the rear wall are higher than the calculated values at 15° and 30°. This is because the ability of downward airflow penetrating into the staged-air zone is enhanced (see Figure 4b,c), and because no air inlets are configured in the region below the airflow zone of staged air, the air pressure is low in this part of the furnace. Thus, mixed air tends to be angled downward more than theoretically expected.6 Measured declined angles near the rear wall are lower than the calculated values at 45° and 55°. This is because when staged air is angled to 30°, the measured declined angle near the rear wall increased to a maximum value of 50°. This value is very near 55°, which is the angle between the dry bottom hopper wall and the horizontal direction (see Figure 1). With a further increase in the staged-air declination angle above 30°, the dry bottom hopper wall prevents the mixed airflow from changing angle at the original rate. The measured declined angles near the front wall are constant at 45° with slight
fluctuations. This is because a deflected flow field forms when the declination angle ranges between 0 and 30°, with upward airflow deflecting toward the zone near the front wall and forcing the mixed airflow near the front wall closer into the dry bottom hopper. With the declination angle increased to 45° and 55°, this impulse disappears as the flow-field deflection vanishes, but unfortunately the dry bottom hopper wall still obstructs the mixed airflow from angling downward. Figure 8 presents a comparison of the decay curves of the fuel-rich flow with respect to different staged-air declination angles. Here (Vy)max is the largest longitudinal-velocity component of the fuel-rich flow at fixed measurement point depths below the front and rear arches. After the fuel-rich flow leaves the nozzle outlets, the values of (Vy)max/V0 in the zone below the front and rear arches are greater than unity. This is to say that flow velocities are greater than their outlet velocities. Thereafter, velocities increase at a high rate initially but then decrease fast after reaching maxima at H/H0 = 0.08-0.12. The peak here occurs because the secondary-air velocity is far greater than that of the fuel-rich flow. The secondary-air expansion makes the fuel-rich-flow velocity increase initially at a high rate after leaving the nozzle outlet upon mixing with secondary air. In matching the secondary-air velocity, the fuel-rich-flow velocity reaches its peak but then decreases rapidly with the decaying downward airflow. After a rise at H/H0 = 0.4 for all five cases but for the downward flow near the rear wall at 0° angle, flow velocities near the front and rear walls continually slow down until H/H0 = 0.5 and then change only slightly afterward. Another velocity peak at H/ H0 = 0.4 is attributed to the staged-air ejection. The center line of the staged-air ports is located at the cross section H/H0 = 0.394, which is sufficiently far away from the fuelrich-flow outlet. The velocity is small in the region above the staged-air ports because of the large decay in the downward airflow. For staged-air ejection, the downward velocity in the staged-air region initially increases rapidly and reaches another peak because of staged-air mixing and then decreases as the mixed airflow penetrates further. The downward airflow, consisting of primary air and secondary air, flows downward into the regions near the front and rear walls with continually decaying velocities (see Figure 4). For small fuel-rich-flow outlet velocities (V0), it is thought that the fuel-rich flow will not effectively penetrate further after the downward airflow has reached points at which the largest dimensionless longitudinal-velocity component has decayed to (Vy)max/V0 = 0.4. Thus, the longitudinal distance between the fuel-rich-flow nozzle outlet and the point at which (Vy)max/V0 = 0.4 is defined as the depth that the 1608
Energy Fuels 2010, 24, 1603–1610
: DOI:10.1021/ef901286r
Kuang et al.
becomes asymmetric. Velocities are small in the region near the rear wall and large in the region near the front wall. Because of the short upper furnace, the asymmetric airflow velocity in the upper furnace inevitably impacts the flow field in the lower furnace, causing the downward airflow near the rear wall to deflect upward toward the region near the front wall and causing it near the front wall to penetrate further into the dry bottom hopper under extrusion of the upward airflow from the region near the rear wall in the lower furnace. Thus, a deflected flow field forms in the lower furnace. The region between the downward airflow near the rear wall and the upward airflow deflected toward the front wall is a lowpressure zone, and a large recirculation zone appears. For this reason, the recirculation zone is always toward the rear wall and not the front wall. In addition, some factors that favor the formation of deflection flow are as follows.11 First, before the two upward airflows near the front and rear walls flow through the entrance of the upper furnace, interactional extrusion between the two upward airflows is strong because the ratio between the width at the entrance of the upper furnace (W2) and the width of the lower furnace (W1) is small (see Figure 1). In association with their cooperating flow extrusion, a deflected flow field forms easily because some differences exist between the two airflows. For 300 MWe down-fired boilers, W2/W1 values are 0.54, 0.54, and 0.51 for FW, B&W, and Stein boilers, respectively, with their large values corresponding to weak extrusion and their flow fields all being symmetric within the furnaces.6,7,10,12 Here W2/W1 is smaller at 0.47, yielding stronger extrusion between the two airflows, which counts as one factor favoring a ready formation of a deflected flow field within this boiler. Second, boiler noise may be the second factor that imposes some effect on the upward flow near the rear wall, causing some differences from the upward flow near the front wall. Third, depending on the angle and ratio of the staged air when injected into furnace, i.e., either horizontally or at a small declination angle (the angle between the staged-air and horizontal directions), staged air has a large transverse-momentum component, providing a strong impetus on the two downward airflows. This may be contributing toward the ready formation of a deflected flow field. In addition, the large span of the secondary-air port, located on the center of an arch and covering almost half of the arch along the W1 direction (see Figure 1), and the high ratio of secondary air (about 70% of the total air mass flux) may be factors enhancing extrusion between the two upward airflows, leading to the formation of a deflected flow field. The above analysis represents a preliminary study of this phenomenon, and further investigation will be forthcoming via experiments and numerical simulation. At declination angles of 0°, 15°, and 30°, a deflected flow field forms in the lower furnace, with a large recirculation region appearing below the rear arch. Cross sections of the longitudinal-velocity components and turbulence intensities, and declined angles of the mixed airflow in the staged-air zone, and downward airflow reach all show asymmetric patterns in the zones near the front and rear walls. When the boiler is operating with a deflected flow field, high-temperature recirculation gas in the zone near the rear wall will mix with the downward flow of pulverized coal/air to enhance combustion, which counts as one favorable factor in good burnout near the rear wall. The flow will move into the high-temperature
Figure 9. Dimensionless depth that the downward flow reaches in the lower furnace with different staged-air angles.
downward airflow reaches into the lower furnace. The ratio between the depth and H0 is the dimensionless depth that the downward airflow reaches into the lower furnace. Figure 9 presents the dimensionless downward flow reach depth versus the staged-air declination angle with the exception of declination angle 55°. This is because, at this angle, the decay curves have no intersection with the dashed line at which (Vy)max/ V0 = 0.4 (see Figure 8). Apparently, the H/H0 values near the front and rear walls are greater than 0.65 but are unsupported by concrete values. As seen from Figure 9, at declination angles of 0°, 15°, and 30°, the reach of the downward flow near the rear wall is only 0.34, 0.48, and 0.51, respectively, depths that are not very far from the nozzle outlet. The fuel-rich flow then reverses and deflects upward at shallower depths. The unburnt carbon content is likely to be high in this situation because residence times for reactions to take place in the furnace are short. Flow reach near the rear wall is lower than that near the front wall. Moreover, as the declination angle increases, the depth near the rear wall increases and that near the front wall changes little. This is because, in a deflected flow field as described above, there is, with an increase in the staged-air declination angle, an increase in the longitudinal momentum component of the mixed airflow in the staged-air zone near the rear wall but little change for that near the front wall. When the declination angle is set at either 45° or 55°, the difference in the dimensionless depths near the front and rear walls is small; depths at 45° are slightly lower than those at 55°. At a 55° angle, the dimensionless depths near the front and rear walls are greater than 0.65. The downward airflow penetrates the middle and lower parts of the dry bottom hopper and washes the hopper walls with high-velocity particulate, which causes slagging. Comprehensive consideration of the appropriate flow depths suggests that the optimal declination angle of staged air is 45°, particularly if slagging is to be avoided in the dry bottom hopper. Although the lower furnace has a symmetrical configuration, the upper furnace is asymmetric, with the boiler nose and upper furnace outlet located on the rear wall. Thus, the direct cause for the formation of the deflection flow field may be the boiler’s short upper furnace. The ratio between the height of the upper furnace (L1) and the equivalent diameter of the straight section of the upper furnace (de) is small at 1.58 (see Figure 1). The influence of the boiler’s nose and upper furnace outlet on the airflow in the straight section of the upper furnace cannot be ignored. With an asymmetric configuration in the upper furnace, the airflow velocity in the straight section
(12) Fan, J. R.; Liang, X. H.; Xu, Q. S.; Zhang, X. Y.; Cen, K. F. Energy 1997, 22, 847–857.
1609
Energy Fuels 2010, 24, 1603–1610
: DOI:10.1021/ef901286r
Kuang et al.
central area in the lower furnace after deflection toward the zone near the front wall. This contributes another factor favoring good burnout near the rear wall. However, in the absence of a recirculation zone such as that occurring below the rear arch and with the pulverized coal/air flow moving along the low-temperature zone near the front wall and dry bottom hopper wall, combustion will be poor and the unburnt carbon content high in the zone near the front wall. That is to say, the deflected flow field results in lean combustion conditions in the regions near the front and rear walls. Low temperatures and poor burnout in the region near the front wall leads to high carbon content in the fly ash throughout the whole boiler. If the staged-air declination angle is set at 45° or 55°, a symmetric flow field forms in the lower furnace instead of the deflected flow field. The difference in the airflow reach near the front and rear walls is small, and the depth at 45° is slightly lower than that at 55°. For a declination angle of 55°, dimensionless depths near the front and rear walls are greater than 0.65. In this case, the airflow reaches the middle and lower parts of the dry bottom hopper and washes the hopper walls, causing slagging. Again, thorough consideration of the airflow reach indicates that the optimal declination angle of staged air is 45° if slagging is to be avoided in the dry bottom hopper.
below the rear arch. Cross sections of longitudinal-velocity components and turbulence intensities, declined angles of the mixed airflow in the staged-air zone, and downward airflow reach are all asymmetric in zones near the front and rear walls. With an increase in the declination angle, the flow-field deflection weakens while the recirculation zone shrinks, drawing nearer the upper region of the dry bottom hopper. Differences in the longitudinal-velocity components, turbulence intensities at certain cross sections, declined angles of the mixed airflow in the staged-air zone, and depths reached by the downward airflows near the front and rear walls all decreased. At staged-air angles of 45° and 55°, flow-field deflection disappears in the furnace. After exiting the nozzle outlets into the furnace, fuel-richflow velocities were greater than their outlet velocities. Thereafter, velocities increased at a high rate initially but then decreased quickly after reaching maxima at H/H0 = 0.08-0.12. After a rise at H/H0 = 0.4, flow velocities near the front and rear walls steadily decayed. At declination angles of 0°, 15°, and 30°, depths reached by the downward airflows near the rear wall were lower than those near the front wall. Moreover, depths near the rear wall increased while those near the front wall changed little as the declination angle increased. For declination angles of 45° and 55°, the difference in the dimensionless depths near the front and rear walls was small and the depths at 45° were slightly lower than those at 55°. In considering an appropriate airflow reach, particularly if slagging is to be avoided in the dry bottom hopper, an optimal declination angle of staged air was found at 45°.
4. Conclusion Cold airflow experiments were conducted within a smallscale furnace of a MBEL down-fired pulverized-coal 300 MWe utility boiler to investigate the aerodynamic field. Effects of the staged-air declination angle on flow field deflection in the furnace were determined. At declination angles of 0°, 15°, and 30°, a deflected flow field had formed, coexisting with a large recirculation region
Acknowledgment. This work was sponsored by the Hi-Tech Research and Development Program of China (863 program; Contract 2006AA05Z321).
1610
