|M.Sc Student||Dolnik Alexander|
|Subject||Numerical Investigation of Burning Processes in Gas|
|Department||Department of Aerospace Engineering||Supervisor||Professor Emeritus Yeshayahou Levy|
|Full Thesis text - in Hebrew|
Development of a new small gas turbine combustor was accompanied with numerous tests to achieve the desired temperature radial profile at combustor exit and it was delayed for many months because of the need to overcome different design problems such as thermal damage to vaporizers. Available and reliable CFD model would allow to spare large part of the tests and shorten significantly the time needed to bring the combustor to the final working configuration. The present research focuses on the modeling and simulation of gas turbine combustor. This study provides insight into physical and chemical processes in combustion, and evaluates variations of combustion performance and exit temperature profiles for different configurations of combustion chamber. The primary objectives of this study are: 1) to establish an efficient end accurate numerical framework for the support of development stage of a new combustor; and 2) to investigate the parameters and mechanisms influencing and responsible for the temperature profile at the combustion chamber exit. Simulation results showed good agreement with experimental data and were able to obtain and point out the exact place of thermal damage caused to the vaporizer. Numerical study was expanded to include last changes made to the vaporizer and showed its impact on the combustor exit temperature profile. Simulation of the modified combustor at turbine engine design point showed significant difference with optimal temperature radial profile at combustor exit. This finding is raising serious worry about turbine life time at given conditions and requires additional experiments to be conducted to check it out. The last part of this study presents an optimization process in which the objective was to receive the optimal temperature profile at combustion chamber exit. This objective is received in a relatively small number of iterations and this is due to parametric approach and advanced visualization tools of software postprocessor.