Learn the basic structure of a proposal.
Proposals share a general document architecture, which is usually modified to suit specificcircumstances. The overall structure of a proposal can be broken down into four parts:
End matter, or management requirements
The front matter of a proposal includes the following components:
Explain clearly how you will manage the development of your proposal project. Proposal reviewerspay strict attention to this section because here you show that you have the required know-how tobring a project to completion. In this section, you will present the following information:
See Work Plan for a discussion of work plans.
|PRINCIPAL INVESTIGATOR:||Ian A. Waitz, |
Assistant Professor, Aeronautics and Astronautics
|PERIOD OF INVESTIGATION:||November 1, 1994 - October 31, 1997|
|FIRST YEAR AMOUNT:||$XXX,000|
|2.||Background and Motivation||1|
|3.||Applications for Tailored Strain Field Mixing for Low-Emissions Gas Turbines||2|
|4.1 Numerical simulations of simplified model flows||5|
|4.2 Diffusion flame experiments||6|
|4.3 Review and analysis of mechanisms for driving controlled-strain mixing environments||8|
|Appendix A.||Resume of Professor Ian A. Waitz||14|
|Appendix B.||Descriptions of the MIT Aero-Environmental Research Laboratory (AERL) and the MIT Gas Turbine Laboratory (GTL)||16|
|Appendix C.||Budget Estimates for Proposed Project||20|
|Figure 1.||Strained diffusion interface||3|
|Figure 2.||Simplified computational flow model 1||5|
|Figure 3.||Simplified computational flow model 2||6|
|Figure 4.||Experimental model flow||7|
|Figure 5.||Work schedule||9|
Mixing enhancement is critical for implementation of low-NOx combustortechnologies. Current mixing augmentation techniques being developed for application in lowemissions gas turbine engines rest largely upon increased intensity of conventional turbulent mixingwith relatively weak spatial and temporal coherence in length scales responsible for the majority ofthe critical fine-scale mixing. The objective of this project is to investigate mixing concomitant withspatially well-defined strain fields such that the emission of pollutant species is reduced. Strategieswill be studied that use high strain rates to selectively enhance or attenuate pertinent reactionprocesses in the course of mixing.
A joint Massachusetts Institute of Technology/Naval Research Laboratory program of analysis,numerical simulation, and experimental research is proposed to determine the feasibility and impactof tailored strain field mixing for control of pollutant emissions in marine and aircraft gas turbineapplications.
The strong sensitivity of pollutant formation processes to local temperature and mixture fractiondictates atomization and mixing with extreme rapidity and spatial uniformity in low-NOx combustion systems. In lean-premixed-prevaporized (LPP), lean direct injection(LDI), and rich burn-quench-lean burn (RQL) combustors, molecular mixing of fuel and air thatminimizes stoichiometric burning (and the associated local heat release that leads to fixation ofatmospheric nitrogen) is desired. This must be accomplished in a manner that allows completeoxidation of CO, soot, and unburned hydrocarbons, while achieving combustor performanceobjectives and satisfying a variety of design constraints.
The critical mixing requirements are set by the kinetic rates of pertinent oxidation and formationreaction mechanisms. These reaction rates are dependent on the local temperature and pressure,and relevant species concentrations. The progression of these reactions is also strongly dependenton the rate of strain in the fluid since this directly affects the fuel/air interfacial area, and diffusionof heat and chemical species.
For the most part, molecular mixing and transfers of momentum and energy in current low-NOx combustion systems are the result of turbulent processes (e.g., jet injection fromdilution holes). That is, there is largely random organization of strain rates in length-scalesresponsible for the majority of the mixing. Mixing using well-defined, tailored strain fields may beused to exert greater control over chemical reactions during the course of molecular mixing andexchange of momentum and energy. As reviewed below, depending on magnitude, strain can eitherincrease or decrease the progression of a given reaction.
Consider two semi-infinite planar regions separated by the x-axis, one containing fuel and the otheroxidizer, as shown in Figure 1a. Infinitely-fast chemical kinetics are assumed so that locally thereaction
Because of the broad range of kinetic rates and heat release associated with primary and pollutantspecies chemistry, the reactions will be influenced differently by straining. Then, tailoring of thestrain field may be used to satisfy the necessary fuel/air mixing and combustion requirements of gasturbine applications in a manner which minimizes pollutant emission.
Figure 1: Strained diffusion interface.
One candidate for leveraging the effects of strain is to allow a reaction interface or strip of reactantsto be convected in the field of an axial vortex such as that shed downstream of a lobed or forcedmixer nozzle. Strain rates on the order of 1000/s are realizable for such geometries at flowconditions gerinane to gas turbine combustor applications. This may allow quenching of the primaryhydrocarbon reactions in regions where stoichiometric burning might otherwise occur. For gaseoushydrocarbon flames (e.g. propane) extinction at atmospheric pressures occurs for strain rates between300/s and 1200/s. In addition, the characteristic chemical time for such a reaction is 10-4 s. This is on the same order or faster than that for some pertinent reactions in gas turbinecombustors, so similar impacts of strain on reaction processes are expected.
Four specific research questions are posed:
The research plan described below is designed to address these four questions.
A joint Massachusetts Institute of Technology/Naval Research Laboratory program of analysis,numerical simulation, and experimentation is proposed. The objective of the research is todetermine the effects of time-varying strain rates on ignition and extinction of the primaryhydrocarbon reaction, and on pollutant species formation/oxidation in gas turbine engines. Theapproach will be to investigate these effects using simplified model flow fields which capture theessential physics present in a gas turbine combustor setting. This will provide the greatest clarity indistinguishing the relevant mechanisms and opportunities for emissions reduction. The functionalrequirements for these simplified model flows are:
Numerical simulations will be performed by an MIT graduate student in residence at the NavalResearch Laboratory. The student will work under the guidance of Dr. Elaine S. Oran. Simulationswill be conducted using the CMRFAST-2D code, which is a fully-parallelized, twodimensional,reacting, Navier-Stokes solver currently running on the Thinking Machines CM-5 ConnectionMachine. The code was originally developed at NRL [Oran, Boris and DeVore, 1992] and employsa flux-corrected-transport algorithm.
Two model flows which satisfy the functional requirements described in the beginning of Section 4will be investigated. The first, shown in Figure 2, is a fuel/product strip subject to a time-varyingstrain rate. A similar configuration was used as a basis for the simulations of Thévenin andCandel . The primary independent variables of interest are the fuel/product speciescomposition and temperature, and the variation of strain rate with time.
The second model flow which will be studied is a fuel strip in the strain rate field of a vortex asshown in the schematic of Figure 3. This flow is more representative of a likely combustorimplementation, containing both spatially and temporally varying strain rates. Large spatialvariations in the strain rate field may limit the emissions reducing potential of the combustionstrategies being investigated, and study of this model flow is intended in part to address this concern.
Experiments will be carried out in parallel with the numerical simulations both to validate thenumerical results, and to study the impact of large-scale straining in reacting flows with embeddedfine-scale turbulence. The work will be performed in a model combustor to be constructed usingexisting combustion lab facilities in the MIT Aero-Environmental Research Laboratory. Thefacilities provide air flow up to 0.7 lb/s and automated control of gaseous fuel flow rate and dilution,in addition to a hot exhaust capability.
A schematic of the experimental configuration is shown in Figure 4. Two splitter plates with asingle convolution in the trailing edge will be stacked one upon the other. A gap between theplates will be used for introduction of gaseous fuels. The convolution in the trailing edge willgenerate a single counter-rotating vortex pair. The vortex pair strength, and thus the strain ratefield, may be directly correlated with the slitter plate geometry (see for example, Waitz etal., 1994].
When the air and fuel velocities are matched, the experiment is similar to the simplifiednumerical model flow shown in Figure 3. Vortex strength and size with respect to the fuel stripwidth can be varied by changing the geometry of the stacked splitter assembly.
At the half-way point of the proposed program, after results of the numerical and experimentalinvestigations begin to be obtained, analysis will commence of the (l) feasibility (2) impact, and(3) methods for application, of tailored strain field mixing in gas turbine engines. This willinclude analysis of existing streamwise-vorticity-enhanced mixing technology, as well asinvestigation of non-traditional approaches for introducing tailored strain fields.
A work schedule for the three year program is shown in Figure 5. The majority of the first yearwill be devoted to modifying the computational fluid dynamics code to enable its use for the flowfields of interest for this study. Simulations of the model flow fields will be carried out duringthe second year, followed by design and construction of the experimental test apparatus. 'Meexperiments will be conducted during the third year of the effort.
The principal investigator for this project will be Professor Ian A. Waitz, Assistant Professor ofAeronautics and Astronautics, Director of the MIT Aero-Environmental Research Laboratory(AERL), and member of the MIT Gas Turbine Laboratory (GTL). A resume is attached inAppendix A; descriptions of AERL and GTL are included in Appendix B.
Co-investigators include Dr. Elaine S. Oran of the Naval Research Laboratory and ProfessorEmeritus Frank E. Marble of California Institute of Technology. Bi-annual meetings will be heldamong the program participants. Dr. Oran will be responsible for advising an MIT student on theimplementation of the CMRFAST-2D code, and will participate as a consultant throughout thecourse of the program. She is an expert in the field of numerical modeling of reacting flows and hasserved a similar role in advising students from the University of Maryland. Professor Marble willserve as a consultant at a level of effort of approximately 15 days/year. He is a recognized leaderin the field of combustion science and has extensive experience in modeling the effects of strain ondiffusion flames. He is the author of a seminal paper on the interaction of a diffusion flame witha vortex [Marble, 1985]. Professor Marble has a long standing relationship, with the MIT GasTurbine Laboratory and makes several visits each year. Interaction with Professor Marble will bethrough these visits and possibly through one visit each year of the MIT student to CaliforniaInstitute of Technology.
Funding is requested to support one MIT doctoral student for the three-year period. The programwill take advantage of existing NRL facilities and personnel; funding for NRL is not requested forthese activities. Budget estimates are contained in Appendix C.
Chomiak, J. "Application of Chemiluminescence Measurement to the Study of Turbulent FlameStructure", Combustion and Flame, Vol. 18, pp. 429-433, 1972.
Driscoll, J. F., Sutkus, D. J., Roberts, Wm., L., Post, M. E., and Goss, L. P., "The Strain Exerted bya Vortex on a Flame-Determined from Velocity Field Images", AIAA 93-0362,1993.
Macaraeg, M. G., Jackson, T. L., and Hussaini, M. Y., "Ignition and Structure of a Laminar DiffusionFlame in the Field of a Vortex," Combustion Science and Technology, Vol. 87, pp. 364-387, 1992.
Marble, F. E., "Growth of a Diffusion Flame in the Field of a Vortex," Recent Advances in theAerospace Sciences, Plenum Publishing, 1985.
The McGraw-Hill Companies, Inc..