OCR Text |
Show tl 96391* I- ~ 0.101 . . ft5W9ff I 0.101 [P j P Hours Hours Figure 3 MPV-1 system display for Homer City Station, Unit 3 The MPV optical sensors are usually mounted in existing boiler observation ports. After the sensor is installed, the customer may continue to use the port for observation. W e recommend sensors to be vertically lined up with burner columns. O n H C Units 1 and 2, four sensors in the upper furnace positioned above each of the 4 burner columns. O n HC-3, eight sensors positioned above each of the 8 burner columns, as shown in Fig. 4 and 5. In the course of long-term testing, we made a number of observations and conclusions: 1. The new MPV-1 system provides consistent, repeatable and sensitive response to changes in combustion conditions. W h e n the same boiler or burner procedure (e.g. unloading and loading of one mill) is repeated several times, the system response is consistent and repeatable. However, the response is often different when the same operation repeated for different mills. In this case, essential differences across the width of the furnace reflect differences in distribution of fuel and airflow in different groups of burners. 2. The MPV distribution profile across the furnace corresponds to other boiler measurements (CO and O ). This profile can be changed and corrected by variations in O F A flows, S A flows and individual burner register adjustments, as illustrated in Fig. 6a-c. 3. The system correlates with LOI analysis of flyash samples, and has a potential to monitor other combustion parameters, such as N O x , using other M P V algorithms or other optical sensors. 4. The new system responds to boiler and mill disturbances and identifies unstable situations. On many occasions, it helped diagnose serious boiler problems and provided early warnings (for example, a developing fire inside a burner, or a sharp reduction of mill S A flow due to a broken linkage on a mill S A drive). 4 |