Session: 13-01 Heat Transfer in Turbines

Submission Number: 176389

Robustness Analysis of the Impact of Step Misalignment on Cooling Performance Under Realistic Swirling Inflow Conditions 

With the development of miniaturization of gas turbines, the influence of geometric deviations on the heat transfer and cooling performance of turbines has become increasingly significant. The upstream endwall misalignment, caused by thermal expansion and assembly or manufacturing tolerance, commonly exists between the combustor exit and the nozzle guide vane (NGV) platform and has attracted wide attention. The previous studies show that the endwall misalignment significantly affects the scale of cavity vortex, which further affects the robustness of cooling performance, under the uniform inflow. However, most of these investigations did not address the influence of the realistic swirling inflow, which is caused by the lean premixed combustion and characterized by nonuniform velocity distribution, on the scale of cavity vortex. In this paper, uncertainty quantification (UQ) analysis is performed to quantify the endwall cooling performance at three design step misalignments (ΔH): a baseline configuration (ΔH= 0mm), two misaligned configurations with forward step (ΔH = −5mm) and backward step (ΔH= 5mm), respectively. The studies are performed at engine-representative Ma_exit = 0.85, Re_{exit, Cax} = 1.5×10⁶, and engine-representative density ratio of 1.95. A detailed comparison is conducted between the uncertainty quantification result of ideal uniform flow and that of realistic swirl flow to reveal the interaction between step misalignment and swirl inflow and its impacts on the robustness of cooling performance.

Presenting Author: Mingyang Hao Xi'an Jiaotong University

Presenting Author Biography: Mingyang Hao is a PhD student at Xi’an Jiaotong University and a visiting student at the National University of Singapore. His main research interests focus on uncertainty quantification of aero-thermal performance for the gas turbine and experimental data processing techniques based on machine learning. He has published three academic papers in the Journal of Turbomachinery as the first author. Additionally, he has presented two conference papers at the ASME Turbo Expo.

Authors:

Mingyang Hao Xi'an Jiaotong University
Zhiyu Li Xi'an Jiaotong University
Zhigang Li Xi'an Jiaotong University
Jun Li Xi'an Jiaotong University