|Preprint-No.:||< 331 >||Published in:||October 2011||PDF-File:||IGPM331_k.pdf|
|Title:||Numerical Simulation of Cooling Gas Injection Using Adaptive Multiscale Techniques|
|Authors:||Wolfgang Dahmen, Thomas Gotzen , Silvia Sorana Melian–Flamand and Siegfried Müller|
The interaction of a jet of cooling gas injected through single boreholes or slots with a supersonic flow field plays an essential role in the design of innovative cooling systems. In order to appropriately resolve the physically relevant effects an efficient and reliable solver is needed. For this purpose, a fully adaptive and parallel multiresolution finite volume scheme is used. Here grid adaptation is based on a multiscale decomposition of the discrete flow data in terms of cell averages at hand. The resulting multiscale data are compressed by means of hard thresholding. From the remaining data a locally refined grid is constructed. For validation purposes, first a simplified two-dimensional configuration mim- icking the gas injection through a slot of infinite length in a plate is investigated. These numerical results are thoroughly validated by van Driest’s self-similar solution for laminar boundary layers and computations performed with non- adaptive codes. In the following, three-dimensional simulations of slots of finite length are carried out and compared with experimental data. The influence of varying rate of cooling gas, injection angle etc. on the cooling efficiency is investigated.
|Keywords:||finite volume method, film cooling, cooling gas injection, multiscale techniques, grid adaptation|
|Publication:||Computers & fluids |
71, 65-82 (2013)
Proceedings / CFD 2010, 5th European Conference on Computational Fluid Dynamics : Lisbon, Portugal, June 14th - 17th, 2010
ECCOMAS, the European Community on Computational Methods in Applied Sciences. Ed.: J. C. F. Pereira ...
|Corresponding author:||firstname.lastname@example.org (Thomas Gotzen)|