Abstract

This paper is concerned with improving the prediction reliability of CFD modeling of gas turbine combustors. CFD modeling of gas turbine combustors has recently become an important tool in the combustor design process, which till now routinely used the old “cut and try” design practice. Improving the prediction capabilities and reliability of CFD methods will reduce the cycle time between idea and a working product. The paper presents a 3D numerical simulation of the BSE Ltd. YT-175 engine combustor, a small, annular, reversal flow type combustor. The entire flow field is modeled, from the compressor diffuser to turbine inlet. The model includes the fuel nozzle, the vaporizer solid walls, and liner solid walls with the dilution holes and cooling louvers. A periodic 36 deg sector of the combustor is modeled using a hybrid structured/unstructured multiblock grid. The time averaged Navier-Stokes (N-S) equations are solved, using the k-ε turbulence model and the combined time scale (COMTIME)/PPDF models for modeling the turbulent kinetic energy reaction rate. The vaporizer and liner walls’ temperature is predicted by the “conjugate heat transfer” methodology, based on simultaneous solution of the heat transfer equations for the vaporizer and liner walls, coupled with the N-S equations for the fluids. The calculated results for the mass flux passing through the vaporizer and various holes and slots of the liner walls, as well as the jet angle emerging from the liner dilution holes, are in very good agreement with experimental measurements. The predicted location of the liner wall hot spots agrees well with the position of deformations and cracks that occurred in the liner walls during test runs of the combustor. The CFD was used to modify the YT-175 combustion chamber to eliminate structural problems, caused by the liner walls overheating, that were observed during its development.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call