OCR Text |
Show grate. In the B F B peat combustion, the typical air distribution is in the order of 30 to 4 0 % through the bed, as the rest is fed as O F A from the secondary and tertiary air nozzles. The results of the final design can be seen in Appendices from 1 to 3 where the velocity, temperature and oxygen fields are presented. A s a result of the modification, combustion is slightly shifted upwards as compared to the grate firing where practically all air was fed through the grate. According to Appendices 3 and 4, where the temperature and oxygen fields are presented, the major combustion takes place just above the secondary air level. The appropriate nozzle velocity and placement of secondary air assure good mixing with air factor about one and the resulting temperature field is reasonably uniform. T w o tertiary air orifices are placed in the same line with the fuel droptubes and approximately two metres above the secondary air level to ensure the final burn out of pyrolysed gases. Due to the shortness of the furnace, the air staging is a compromise between the NO-reduction and the amount of unburned fuel. For a better N O reduction, the O F A feeding should have been weighted more to the tertiary air level. Further, the distance between the secondary and the tertiary air levels should have been greater to ensure sufficient residence times for reburning to take place. However, as it can be seen in Table 1, the amount of unbumed already limits the NO-reduction that can be achieved by means of air staging. Fortunately, the reasonably uniform and moderate temperatures efficiently reduce thermal N O . The above estimations were later confirmed by measurements as can be seen in Table 1. The measured flue gas mass flow rate as well as the 02-content (Table 1) are clearly above the designed values. This is due to uncertainties associated with the O2- and Kurz-rype flow measurements in addition to the omitted leakage-air in simulations. A slight disappointment was the over estimation of the C O level, which was in practice significantly lower than what had been computed. This is probably due to different conditions of B F B as compared to the pulverised combustion, for which the models of A R D E M U S were originally developed and selected, not to forget the inaccuracies related to the CO-measurement. The measured N O x emission level was typically between 100 and 200 mg/MJ, which is a relatively good value for such a short furnace. The existing steam turbine plant was intended to be utilised also after the modification and therefore the total fuel power and the output steam features were preserved. The design steam characteristics at full load are 400 °C and 3.0 M P a for the superheated steam temperature and pressure, respectively. As can be seen in the measured data in Table 1, the steam values are at the upper limit or as much as exceed the design values even in the grate firing with coal. Unfortunately, the boiler is not equipped with a feed water spray system to the superheaters, but there are flue gas recirculation nozzles just below the superheaters for the control of superheated steam values. However, the effect of the latter has been found to be unsatisfactory. It was also found that the economiser tubes were somewhat twisted which indicates evaporation in the economiser. In conclusion, in the case of coal grate firing there was no extra capacity available in the evaporation and radiative heat transfer surfaces and one should be careful in the design not to exceed the steam temperature of 400 °C. The high moisture content of peat increases flue gas energy due to the increased flue gas flow and its specific heat, that will further decrease the flue gas temperature and radiation. Hence, the above mentioned problems with convective section at the furnace were expected to be worse. Table 1. Primary process values ofVanaja furnace at full load before and after the modification Flue gas flow rate N 0 2 (dry gas) o2 CO Tbed T before curtain T before I s.h. T before II s.h T after s.h. T before airpreh. Tafter airpreh. T s.h. steam Steam in econ.2) Boiler efficiency2) ! kgs"1 rngMJ-' vdry-% ppm 0 C 0 0 c 0 c 0 c 0 c c 0 0 c wc-% % Grate meas. 32.2 7.9 150 - 1050 9002) 7702) 5902) 250 180 410 5.1 89.9 ] calc. 30.1 - 3.9 1500 850 1120 930 810 660 270 190 390 1.6 89.9 BFB meas. 36.812 200 6.5 170 870 9902) 7602) 6702) 5802) 220 155 386 6.0 88.7 1) Estimated including 4.5 kgs"1 leakage air after super heaters, i.e. 32.3 kgs ' from furnace 2) S O L V O * simulation based on measurements To avoid these problems, an extra evaporator surface was added on and the heat transfer was designed to be as effective as possible in the lower part of the furnace. According to simulations, the peat with a 5 0 % moisture 4 |