ACERC
Abstracts - 1991 |
ACERC
Abstracts - 1991 |
|
|
Thrust Area 6: Model Evaluation Data and Process Strategies |
Hancock, R.D.; Hedman, P.O.
and Kramer, S.K.
Combustion and Flame, 87:77-88, 1991. Funded by ACERC.
Coherent anti-Stokes Raman spectroscopy (CARS) is a laser diagnostic technique that can be used to determine temperature and major species concentrations in harsh combustion environments. CARS has been successfully applied to clean gas flames, but much less attention has been given to particle-laden flames like those encountered in industrial coal burners. Typically, experimental CARS spectra are obtained from a flame and then compared with theoretical CARS spectra to determine temperature and species concentration information. This information is more difficult to acquire in coal flames due to background and nonresonant interferences. These interferences alter the shape and intensity of the CARS signal, thus making analysis with unmodified version of standard CARS fitting codes impractical. Nitrogen temperature measurements were obtained in heavily coal-seeded natural gas/air flames. Two different coals and several coal feed rates and stoichiometries were investigated in order to determine possible limits associated with making CARS measurements in coal flames. Carbon monoxide signals were observed in some of the fuel-rich coal-seed flames but the signals were weak and of poor quality, therefore, quantitative results are not reported. Temperature measurements were obtained with nonresonant background levels caused by particle-induced breakdown as high as 100% of the peak N2 resonant signal. CARS N2 temperatures generally agreed with equilibrium code calculations.
Sturgess, G.J.; Sloan, D.G.;
Roquemore, W.M.; Reddy, V.K.; Schouse, D.; Lesmerises, A.L.; Ballal, D.R.; Heneghan,
S.P.; Vangsness, M.D. and Hedman, P.O.
10th International Society of Air Breathing Engines, Nottingham, England,
September 1991. Funded by US Air Force.
A progress report is presented on a comprehensive research program aimed at improving the design and analysis capabilities for flame stability and lean blowout in the combustors of aircraft gas turbine engines. The motivation and aims of the program are reviewed, and the unusual approach adopted to address the research issues is outlined. The supporting experimental program and the test vehicles involved are described, together with some major results obtained to date. The modeling techniques that are being explored are summarized. Their potential and limitations are highlighted. Although much work remains yet to be done, the progress made thus far gives rise to reasonable optimism for achieving the program objectives.
Cannon, J.N.; Webb, B.W.
and Queiroz, M.
Fossil Fuel Combustion, 33:49-56, (R. Ruiz, ed.), The American Society
of Mechanical Engineers, 1991. Funded by Empire State Electric Energy Research
Corp. and ACERC.
Brigham Young University (BYU)/ACERC testing of an 80 MWe corner fired coal power plant in June of 1989 sought data to validate 3-D comprehensive combustion computer codes. The major controlled variables during the two weeks of testing were coal grind fineness, O2 content in the flue-gas, plant load, and burner tilt. A full complement of board data was taken along with coal fineness, coal split between coal feed lines and on-line trial instruments on carbon-in-ash and air flows. Selective spatially resolved data was collected throughout the radiant section of the boiler from the near field in the burners to the platen area with water cooled probes for gas temperature, gas composition, particle size distribution and composition, velocity (magnitude, direction and turbulence) measurements, and radiant flux. The water cooled probe results yielded spatially resolved data in the near and far combustion field that will be compared to computer modeling runs in the future at BYU. The effects of O2, coal fineness, load variation and burner tilt are clearly discernable on the spacially resolved gas temperature, gas composition, velocity, radiant flux and particle size and composition measurements and are reviewed in this paper. Centrifugal effects on particle size are also discernable as are particle size effects on composition.
Roquemore, W.M.; Reddy,
V.K.; Hedman, P.O.; Post, M.E.; Chen, T.H.; Goss, L.P.; Trump, D.; Vilimpoc,
V. and Sturgess, G.J.
AIAA 29th Aerospace Sciences Meeting, Reno, NV, January 1991. Funded
by US Air Force.
This paper reports the results of an investigation to determine the flow and flame characteristics of a burner which has been carefully designed to "specifically reproduce recirculation patterns and lean-blow-out (LBO) processes that occur in a real gas turbine combustor." When operated in a fuel rich mode, the flame is very stable and is anchored in the jet shear layer by a pilot flame attached to the step, near the outer edge of the air supply tube. As the equilivalence ratio is reduced, the flame becomes less stable, and eventually reaches a point where the pilot flame becomes detached (lifts) from the base region, and the entire flame structure becomes stabilized downstream. Thus, there are two distinct operating modes for the combustor: a fully attached flame and a lifted flame. As the fuel equivalence ratio is further reduced, the flame becomes progressively less stable in its lifted condition, and eventually blows out. Photographs of the flame clearly illustrate the attached and lifted flame operational regimes of the combustor. However, visual observation and conventional photographic techniques are unable to quantify the precise details of the flame transition from an attached to a lifted condition. A Computational Fluid Dynamics (CFD) model with one step chemistry was used to investigate the time-averaged features of the reacting and non-reacting flow fields. The difficulties of predicting the characteristics of the attached flame with a time averaged CFD type model are discussed. Measurements, using OH emissions and gray body radiation from 14 µm diameter filaments located near the base of the flame, clearly indicate the dynamic or intermittent nature of both the attached and lifted flames. It is theorized that unburned hydrocarbon combustion products are transported into the recirculation zone by the intermittent process; and it is these products that provide the fuel needed for the flame to attach to the outer edge of the step.
Cannon, J.N.; Webb, B.W.
and Queiroz, M.
Advanced Combustion Engineering Research Center, Empire State Electrical
Energy Research Corp. Report, 1991. Funded by Empire State Electric Energy Research
Corp. and ACERC.
The Goudey Station testing has direct purpose for validation as well as an indirect purpose in future boiler design and operation. All of the spatially resolved data are gathered by research type instruments under controlled and common conditions. Data are not taken at every internal node (i.e. every 6 inch in spacing), but at sample locations where boiler internal access is available.
For the second series of tests in 1991, spatially resolved data were gathered from seven, four-inch ports and thirty, two inch ports, during twenty-five different tests. The data gathered consisted of gas temperature, gas and particle composition, particle velocity, concentration, size distribution, and radiation flux, all of which are combustion code predicted variables. The major test variables were load, excess air and coal grinded fineness. All three operating variables had significant influence on the test variables. In addition, burner tilt had a strong influence on the spatially resolved variables. Soot blowing was constrained during these tests but could not eliminate the effects of ash depositions during the tests. Predictions are expected to show the same trends.
The radiant flux showed significant differences from port to port and boiler level differences. Direct probing of the burner plume was of sufficient accuracy to identify coordinated O2 concentration and gas temperature variation through the burner plume. This compared favorably with the two-dimensional PCGC-2 predictions for the first 1.5 meters of the burner plume before three-dimensional effects altered the two-dimensional flow characteristics. The particle size distribution, concentration and velocity data were accurate enough to see the centrifugal effects of a corner-fired furnace on cyclone type particle separation. In addition, a pulsating particle concentration could be identified that was not related to any causal source but required temporal averaging. Gas temperature and radiation flux data averaged by level showed a mid-furnace flattening that was unexpected. Three-dimensional, gas only, predictions did not provide any explanation of the flattening nor were the predictions expected necessarily to resolve the observations. Carbon burnout was correlated with furnace level enough to indicate nearly complete combustion by the seventh level in the furnace.
|