| OCR Text |
Show 1. INTRODUCTION In today's industrial environment there is a need for a fast, reliable computational tool that predicts NO formation in conjunction with carbon burnout and furnace outlet temperature for commercial-scale furnaces. This is especially true in advanced firing systems which employ staged combustion scenarios for NOx reduction. As recently reviewed by Coimbra, et al. (1994) and Weber, et al. (1993), the modeling of pulverized coal furnaces has advanced from an empirical and global approach to detailed phenomenological models that mathematically predict combustion and its related physical and chemical processes in three dimensions. These authors, however, also point out that the practical use of today's most detailed models is limited by computational economics. SpeCifically, a designer/engineer can not afford to make detailed 3-dimensional computer runs over the range of all coal types, all furnace geometries, and all operating conditions that are or will be used. Typical engineering applications must also be completed in a short time frame and long, tedious computations are not feasible. It is possible, however, that the primary mechanisms affecting furnace outlet temperature and carbon loss (Le. time-temperature history of the particle, fuel-air mixing) can be simulated in two dimensions rather than three, thus reducing computational time considerably. However, in order to transform a three-dimensional furnace into a 2-dimensional equivalent, important furnace design parameters that may affect time-temperature history and mixing must be preserved. If correctly transformed, the 2-dimensional furnace equivalent could be analyzed using reacting CFO-based engineering tools to provide reasonable predictions of process conditions. The important aspects of fuel-air mixing are thus included providing a more realistic sim ulation. Predicted furnace outlet gas temperatures, combustion efficiencies, and NO emissions for the tangential fired furnaces presented in this paper were obtained using JASPER, a 2-dimensional computational fluid dynamic (CFD) code from Reaction Engineering International (REI). JASPER is a fully coupled reacting CFD code which is able to mathematically model coal combustion, radiation, 2-dimensional fluid dynamics and NOx formation. Mapping of 3-dimensional furnace effects into the 2-dimensional framework was accomplished through experimental testing and empirical factors derived from field data and experience. Important inputs to this model include experimental results from ABB Power Plant Laboratories' Drop Tube Furnace System (Thornock, et aI., 1993). Experimental inputs, for a range of coal types, have been calibrated for coal devolatilization/combustion and kinetics via a detailed series of comparison tests and simulations between the experimental Drop Tube Furnace System (OTFS) results and 2-D JASPER model. Experimental data from the OTFS have also been used to characterize fuel nitrogen conversion to NO over a wide range of stoichiometric conditions and fuel types. Fuel nitrogen conversion rates are used directly in JASPER's NOx prediction model to obtain the commercial-scale prediction. The attainment of the proper time/temperature history in the 2-dimensional simulation requires input regarding the wall conditions of the fumace. For example, following well established Combustion Engineering tangential fired furnace modeling practice 2 |
