| OCR Text |
Show The MPV-1 system was also used to find an optim u m distribution of the O F A flows leading to to a reduced level of C O while keeping N O x constant, as illustrated in Fig. 10. Such optimum combinations of O F A compartments may vary depending on boiler load, mills in operation and the actual O F A flows through individual compartments. That is why it was important to develop a certain strategy of using the MPV-1 system to improve boiler performance. The basic rules of boiler optimization strategy, focused on LOI reduction, include balancing the M P V bars and reducing their level by adjusting the O F A flows, S A flows to burner rows and individual burner registers, while observing N O x . Other rules could be boiler specific. Figure 10 Effect of OFA adjustments on balancing and reduction of C O Correlation with LOI In order to study correlations of the MPV bars with direct LOI measurements, special tests were conducted using 8 flyash sampling probes of different length, as illustrated in Fig. 11, where probes # 2 and # 7 were the regular boiler probes. Several sessions of special "profiling" tests were conducted using all 8 probes. At steady-state boiler conditions, 2 or 3 consecutive 15-minute ash samples were collected in each test on both South and North sides from each of the 8 probes. The collected ash samples were sent for laboratory analysis. The results were compared with average 15 minute values of the corresponding M P V bars and presented in Fig. 12 where each M P V value corresponds to the actual LOI measurement, and each LOI value corresponds to the period of time when the average M P V value was calculated. A s w e can see, both profiles well correspond to each other and it leads us to a conclusion that distribution of the M P V bars across the furnace reflects the real distribution of unburned carbon in flyash. Then we compared the average values of measured LOI for all 8 probes and for 2 regular probes. The results, due to the gradual LOI profile, were practically the same (13.7 % vs 13.6%). At the same time, when the LOI and M P V profile had sharp changes, the same values were essentially different (12.5% vs 15.0%). It means that the LOI values received using regular ash sampling probes could be essentially different than the actual average LOI values and, therefore, the plant's regular LOI measurements might be essentially distorted. Our conclusion that the MPV bars correlate with direct LOI measurements was also supported by other boiler measurements and observations, including the following: - correlation with C O during major combustion changes; - the same bias in the M P V readings and in the measured LOI values ( South» North); - the same bias in the M P V readings and in the measured C O; - when the M P V bars are balanced, C O measurements (South and North) are also balanced. 1.0CJ . ... • •^^s^Z Mk . 13.07 00 .J-.G- - ; 2 • »~4«SV 3 - -J.. 094V i.ooo : ivsjjfc RED -!?'" ~ 1000 ppm North South 8 |
