ACERC
Abstracts - 1999 |
ACERC
Abstracts - 1999 |
|
|
Research Area 4: Simulation/Validation |
Xu, H.; Smoot, L.D. and
Hill, S.C.
Energy & Fuels, 13:411-420, 1999.
Advanced reburning is a NOx reduction process wherein injection of a hydrocarbon fuel such as natural gas downstream of the combustion zone is followed by injection of a nitrogen-containing species such as ammonia. The authors recently reported a seven-step, 11-species reduced mechanism for NO reduction by advanced reburning processes. However, inclusion of even a seven-step reduced mechanism into a CFD code for turbulent combustion leads to substantial computational demands. In this work, the authors have further simplified the kinetic mechanism. A simpler four-step, eight-species reduced mechanism for NO reduction by advanced reburning has been developed from a 312-step, 50-species full mechanism through the use of a systematic reduction method. The four-step reduced mechanism is in good agreement with the full mechanism for most laminar flow cases. It also agrees qualitatively with three sets of experimental data, which show the influences of temperature, CO concentration, O2 concentration, and the ratio (NH3/NO)in. It can be applied for coal-, gas-, and oil-fired combustion. The four-step reaction sequence has been integrated into a comprehensive CFD combustion code for turbulent combustion, PCGC-3. The method of integration is described. Several computations are reported with the combined code to demonstrate the predictive behavior of the advance reburning mechanism in turbulent, pulverized coal combustion. The model calculations show the effects of temperature and concentrations of CO, O2, and NH3 on NO reduction.
Hayes, R.R.; Wang, J.; McQuay,
M.Q.; Webb, B.W. and Huber, A.M.
Glastechnisched Berichte - Glass Science and Technology, August 1999.
This study reports optically measured glass surface temperatures along the furnace centerline in the combustion space of a side-port, 550-ton/day industrial, gas-fired flat glass furnace. The measurements were made using a water-cooled two-color pyrometer inserted through holes in the crown at six locations along the length of the furnace. Both average and time-resolved glass surface temperature measurements were performed during the approximately 20-second reversal period of the furnace. The measured glass surface temperature data are supplemented by observations of the batch location using a specially designed, water-cooled video probe. The average temperatures were found to rise from a low near 1700 K near the batch blanket to a peak of approximately 1900 K, and then drop to a level of 1800 K. Evidence of batch islands or "logs" is observed in the surface temperature data collected at the measurement location nearest the batch blanket; large temperature excursions are seen here, indicative of measurement alternately of both the batch surface and the molten glass. Also reported in this study are results of a numerical model for the three-dimensional melt flow and heat transfer in the tank, coupled with a batch melting model. The radiant heat flux distribution incident on the melt and batch blanket surfaces is assumed. The melt tank model includes bubbling. The numerical predictions agree well with the time-averaged glass surface temperature data collected experimentally. The measurements and model predictions illustrate the complex transport phenomena in the melting section of the furnace.
McQuay, M.Q.; Webb, B.W.
and Huber, A.M.
Combustion Science and Technology, Revised, March 1999.
Post-rebuild profiles of velocity, species concentration (O2, CO, and CO2), and gas temperature are reported in the portnecks of a regenerative, side-port, 550-ton/day, gas-fired, flat-glass furnace. These measurements are also compared to similar ones made before the same furnace was rebuilt. Measurements were also made below one of the regenerators in the tunnel leading to the furnace stack after the rebuild. Fewer variations were observed in the exhaust profiles of most measured variables after the rebuild. Flat inlet velocity profiles were measured with a magnitude of approximately 11 m/s before and after the rebuild. The temperature of the inlet preheat air was generally speaking higher and the furnace exhaust temperature lower before the rebuild. Locations of low O2 concentration in the effluent are consistent with high CO concentrations before and after the furnace rebuild. CO2 concentrations are nearly uniform across the portneck height, more so after the rebuild. The measurements in the tunnel after the rebuild indicate a stratification effect in the species concentration measurements. These measurements also indicate that the combustion reactions continue inside the regenerators resulting in overall complete combustion as indicated by the very low CO levels in the tunnel. A mass balance analysis for the overall combustion reaction based on the measurements of O2 and CO2 and fuel flow rate in each port showed that (1) before and after the furnace rebuild the predicted CO2 formed in the glass is within 15% of the value estimated by Ford personnel; and (2) the overall stoichiometry was not much different before and after the rebuild (22.5% excess air before compared to 19.2% after). The total airflow rate calculated by this analysis after the rebuild is within 9% of the plant-measured value.
Eaton, A.M.; Smoot, L.D.;
Hill, S.C. and Eatough, C.N.
Progress in Energy and Combustion Science, 25:387-436 (1999).
Development and application of comprehensive, multidimensional, computational combustion models are increasing at a significant pace across the world. While once confined to specialized research computer codes, these combustion models are becoming more readily accessible as features in commercially available computational fluid dynamics (CFD) computer codes. Simulations made with such computer codes offer great potential for use in analyzing, designing, retrofitting, and optimizing the performance of fossil-fuel combustion and conversion systems.
The purpose of this paper is to provide an overview of comprehensive combustion modeling technology as applied to fossil-fuel combustion processes. This overview is divided into three main parts. First, a brief review of the state-of-the-art of the various components or submodels that are required in a comprehensive combustion model is presented. These submodels embody mathematical and numerical representations of the fundamental principles that characterize the physico-chemical phenomena of interest. The submodel review is limited to those required for characterizing non-premixed, gaseous and pulverized coal gasification and combustion processes. A summary of the submodels that are available in representative computer codes is also presented.
Second, the kinds of data required to evaluate and validate the predictions of comprehensive combustion codes are considered. To be viewed with confidence, code simulations must have been rigorously evaluated and validated by comparison with appropriate experimental data, preferably from a variety of combustor geometries at various geometric scales. Three sets of validation data are discussed in detail. Two sets are from the highly instrumented, pilot-scale combustor called the controlled profile reactor (CPR) (one natural gas-fired and one coal-fired), and the other set is for a full-scale, corner-fired 85 MWe utility boiler.
Third, representative applications of comprehensive combustion models are summarized, and three sets of model simulations are compared with experimental data. The model simulations for the three test cases were made using two commonly used, CFD-based computer codes with comprehensive combustion model features, PCGC-3 and FLUENT 4.4. In addition to the standard version of FLUENT, predictions were also made with a version of FLUENT incorporating advanced submodels for coal reactions and NO pollutant formation.
|