New: Defense of M.Sc. thesis:
The thesis defense by Mr. Erfan Dabbaghchian a M.Sc. student in aerospace, entitled "Simulation of flow leakage in impulse turbines' blade tip", will be held on 2024-September-03 at 14:00 (Tehran time) at the School of Mechanical Engineering of Iranian University of Science and Technology (IUST). The members of the Space Propulsion Research Laboratory (SPRL) invite all interested parties in this field and the other relevant areas to participate in this defense.
 
Abstract
A supersonic turbine is one of the main components of a turbopump, which serves as the power source for a liquid propulsion engine. When the flow rate is low, a high specific work is required for the turbine's operation. Due to the low flow rate, the blade height and consequently the aspect ratio will also be small. To reduce the losses associated with a low aspect ratio, partial admission turbines are used in turbopumps. In this approach, the gas flow only impacts a portion of the rotor blades. At the blade tips, a clearance is considered to prevent contact between the blade and the casing, which results in tip leakage flow. 
The focus of this study is to investigate the behavior of tip leakage flow in the space above the blades of a supersonic partial admission impulse turbine. This leakage flow is controlled by cavities specifically designed at the blade tips for this purpose. The study's approach involves numerical simulation and the use of artificial intelligence algorithms to predict the flow behavior within the turbine. To validate the numerical results, the performance of an existing partial admission turbine is simulated, and the obtained results are validated against its experimental data. Subsequently, the same turbine is simulated with modifications to the upper blade profile using several defined geometric parameters, including angle, setback, and cavity depth. 
Further steps in this research include analyzing the results of these modifications, creating a database to utilize artificial intelligence algorithms to replace simulations, and predict the turbine's performance. It is worth mentioning that in the end, the optimal blade tip cavity profile is provided by the artificial intelligence algorithm with a negligible error margin (less than 1%).
 
Keywords: Supersonic impulse turbine, blade tip leakage, neural network, simulation