
|
|
By Graham Goldin and Jens Madsen, Fluent Inc.; and Bill Rogers and Douglas Straub, NETL, Morgantown, WV A Trapped Vortex Combustor (TVC) is an advanced concept for gas turbine engines. In a TVC, the flame is stabilized by a vortex in a cavity adjacent to the main air stream. While the TVC is still under development at research facilities such as the U.S. Department of Energys National Energy Technology Laboratory (NETL), it promises more stable, compact, fuel-flexible flames with lower emissions than conventional combustors. The trapped vortex that is used to stabilize the flame is also the site where the fuel and secondary air are injected. Since there is not enough oxygen to completely oxidize the fuel in this region, a significant amount of the fuel is reformed into H2 and CO. The main air-stream, oriented along the axis of the combustor, oxidizes the reformed fuel under lean conditions. In fact, the total amount of air injected into the combustor can be up to 250% of the theoretical amount required to completely oxidize the fuel. Lean flame conditions such as this produce low emissions. When simulating a TVC, the burning of the reformed fuels, especially the CO, is difficult to capture with conventional combustion models. Because the CO burns slowly, it is never in a state of chemical equilibrium, so the non-premixed PDF/ mixture fraction model, which is based on an equilibrium assumption, is not adequate. The eddy dissipation model also fails because it does not incorporate real chemistry the rate of combustion is determined instead by the rate at which turbulence can mix fuel and oxidizer into the combustion zone, where the chemistry is considered to occur rapidly. Kinetically controlled species such as CO and NOx are best modeled using a finite rate formulation, but there is a major obstacle in using a finiterate chemistry model that incorporates dozens of species and hundreds of reactions in multi-dimensional CFD simulations. The obstacle is that the chemical mechanisms are invariably stiff, with reaction time scales that can span several orders of magnitude. To solve chemical systems of this type, enormous amounts of CPU time are required.
|
FluentNEWS |
||