OCR Text |
Show 8 Coal properties influence NOx emissions in many more important ways than the fuel nitrogen content. (The latter does not vary widely within a given coal rank to the extent that fuel sulfur content varies.) One strategy for reducing NOx emissions on pre-NSPS units is to implement burners out-of-service (BOOS) or simulated overfire air operation. This typically involves terminating the coal flow to the top row of burners by taking a mill out-of-service leaving the air registers partially open. However, this requires spare pulverizer capacity, which depends on the coal grindability and the maintenance history of the pulverizers. Switching coals from high heating value or low ash deposition characteristics to one of lower heating value or higher ash depOSition can significantly change the heat balance or distribution in the boiler. This in tum affects the tradeoff between reheat steam temperature, burner tilts, and O2 level on tangentially-fired units. Another coal property that directly affects NOx formation and ash carbon content is the fuel ratio (fixed carbonlvolatile matter ratio). Although most utilities are reluctant to purchase a coal strictly because of the effect of the coal properties on NOx emissions, it is increasingly likely that coal properties besides fuel sulfur content will be taken into consideration with regard to meeting emission regulations for NOx as well as S02. Since most emission and boiler performance tests are conducted at a fixed load under steady state test conditions, many field test engineers do not realize that NOx emissions can vary Significantly under transient or load following operating conditions. During one recent test program on a large opposed wall-fired cell burner unit, a three megawatt per minute load ramp was initiated from 350 to 625 MW while NO emissions were monitored at the economizer exit. However, this load ramp was temporarily interrupted for about 15 minutes due to a system constraint caused by a tie line connection problem. The NO emissions and operating data from this test are shown in Figure 6. The control operator was forced to reduce the firing rate (coal flow) slightly shortly after one p.m., because the steam temperatures on the supercritical unit overshot the normal operating set point. Initially, the NOx emissions were relatively flat as the coal firing rate was increased (up to 12:40 p.m.) because the operating O2 level was gradually decreased by the automatic combustion controls. The NOx emissions began to increase rapidly when another pulverizer was added and the firing rate increased. The rate of the NOx increase was somewhat higher than expected, but the subsequent decrease in NOx, back to the initial starting level with the slight coal flow decrease, was even more surprising. Once the generating ramp resumed and the coal flow increased, the NOx emissions once again increased to an even higher level. It should be noted that this unit has a high volumetric heat release rate and a significant fraction of the NOx emissions are thought to be thermal NOx in origin. The main point of this discussion is to point out the wide variation in NOx emissions that can occur over the load range during transient operations compared to steady state test conditions. On some units, a combustion air leak during a load ramp can lead to much higher NOx emissions than normal steady state operation. |