LES MODELING OF HIGH-LIFT HIGH-WORK LPT BLADES: PART II—VALIDATION AND APPLICATION

Abstract

Abstract Over the years, computational fluids dynamics (CFD) has matured to such a state so as to be indispensable in turbine design. In the past two decades, significant advances in turbine design have been made through the use of CFD—in particular, through the use of Reynolds-Averaged Navier-Stokes (RANS) simulations. Currently, turbine design is RANS-driven; however, significant gains in performance and efficiency are becoming more difficult to achieve using RANS. For this reason, the turbomachinery CFD community is moving toward Large-Eddy Simulations (LES). In the design of low-pressure turbine (LPT) blades, LES is particularly beneficial owing to its ability to capture accurately both transition and separation. In this paper, LES is used to characterize a new family of high-lift high-work LPT blades—designated the LXFHW-LS family—designed at the U.S. Air Force Research Laboratory (AFRL). LES simulations are conducted in accordance with the methodology outlined in Part I of this paper. The purpose of this paper is twofold: 1) to use LES to predict the performance of the LXFHW-LS family and compare to measurements in a low-speed linear cascade and, in doing so, 2) to illustrate how LES may be used in LPT design as it evolves from RANS-driven to LES-driven. For each blade in the family, the loading distribution and loss coefficient are computed for sixteen separate Reynolds numbers. Computational results are validated using detailed experimental measurements from a low-speed linear cascade wind tunnel.