OCR Text |
Show NELSON DEWEY REBURNING SYSTEM MIXING EVALUATION The mixing of the Nelson Dewey reburning system was evaluated, not only to optimize NO. reduction, but also to ensure proper reburning burner flame penetration into the furnace. Over-penetration or under-penetration of the reburning burner flame could cause tube wastage and flame stability problems. Both physical and numerical flow modeling methods were used. A 1/12-scale plexiglas model of the Nelson Dewey unit was constructed and baseline tests were performed with isothermal air flow through the model. Both models were compared to the velocity measurements obtained from the Nelson Dewey plant for validation purposes. After validation of these models, the reburning system mixing tests were performed. Boiler Description The boiler was recently described in detail by Yagiela, et al. (~). Only a brief description of the boiler pertinent to flow behavior is given here. The full-scale boiler, shown in Figure 9, is a B&W radiant boiler. The design rate is 100 MWe but the electric generating capacity is currently 110 MWe. The boiler is frred with three cyclone furnaces located horizontally on the front wall with clockwise swirl. Hot combustion gases exit the cyclones at temperatures above 3000·F. A target wall in the boiler directs the cyclone flow downward toward the floor of the boiler. The gas then turns upward and passes through slag screen tubes where it enters the main furnace. The lower furnace and slag screen are insulated to keep the slag in a molten state. The tube walls of the boiler above the slag screen are free of refractory. There is no running slag above the refractory region. The gases are cooled to approximately 2200·F prior to entering the convection pass. Baseline Flow Patterns Due to the critical importance of mixing, a comprehensive study was first performed for the baseline configuration of Nelson Dewey, including physical and numerical flow modeling and full-scale velocity measurements. Velocity profiles at the approximate proposed reburning burner elevation were measured for cold flow conditions (using air only) and hot flow conditions (frring oil at approximately 50% load). A comparison of numerical, physical, and field cold flow results is shown in Figure 10 and a comparison of numerical and field hot flow velocity profiles are shown in Figure 11. Predictions are in general agreement with the physical flow data, and the field velocity measurements; thus, this information provides the means for validating the physical and numerical flow models. In all cases, the velocity is highest near the rear wall, and a large recirculation zone exists in the main furnace with flow moving downward along the front wall and , target wall. Some disparities exist between data and predictions, however. The numerical model shows a bias in flow to the left side, due to the clockwise cyclone swirl, that could not be confrrmed by the data. The velocity gradient from front to rear is also somewhat steeper at Nelson Dewey than for the numerical model (Figure 11). These differences are not significant however. A sensitivity analysis was performed with the numerical model to ensure that the mixing results were not affected by these differences. Reburning System Evaluation The host site boiler was inspected for suitable burner and OF A locations. It was necessary to arrange the reburning burners on the rear wall to achieve uniform mixing across the width of the boiler, and at an elevation above the slagging zone to prevent slag build-up around the reburning burners. Due to the space limitations, a maximum of four reburning burners could be utilized. An average plug flow residence time of approximately 0.5 second in the reburn zone was used to locate the OF A ports in a horizontal elevation above the reburning burners. Physical and numerical models were then utilized to compare the mixing effectiveness of three or four reburning burners in combination with three or four OF A ports. As mentioned earlier, numerical model predictions were in general agreement with the baseline results of the Nelson Dewey boiler measurements. In order to validate the numerical model for penetration from the reburning burners and OFA ports, numerical simulation of the physical model was performed. In this simulation, the measured velocity distribution approaching the reburning burner elevation was used as the inlet boundary conditions of the numerical model. This eliminated the uncertainties associated with differences between the numerical flow predictions and the steep velocity gradients of the physical flow measurements. Numerical model predictions of jet penetration were in qualitative agreement with flow visualization in the physical model using smoke injection. The methodology used for numerical simulation of the physical model was used for the full-scale reburning system. The measured velocity profiles from the Nelson Dewey boiler were used as the inlet boundary condition of the numerical model. The predicted flow patterns and stoichiometry distribution in the furnace are shown in Figure 12. The shaded region is the reducing zone where NOll destruction takes place. The large recirculation zone that was present during baseline conditions was fortunately eliminated during reburning conditions. The reburning burner flow has adequate penetration without impinging on the target wall because the location of low stoichiometry is near the center of furnace. Adequate reburning burner penetration will maintain flame stability and prevent tube wastage. The OF A flow also penetrates adequately and will be discussed later. |