Session: 32-03 CFD Studies

Paper Number: 152665

Effects of Wall Temperature on Transonic Gas Turbine Aerothermodynamics: A Wall-Modeled LES Study 

Gas turbines operating in transonic regimes encounter significant challenges due to the complex interactions between shock waves, boundary layer separation, and high turbulence levels. Transonic flows are characterized by complex phenomena such as shock-wave/boundary-layer interactions, flow separation, and the interplay between wake turbulence and compressibility effects, making accurate prediction and control of aerodynamic performance difficult. These challenges are further compounded by the presence of coherent turbulent structures in the wake and the highly unsteady nature of the flow, especially in regions downstream of the turbine blades.

In recent years, several advancements have been made in the field of Computational Fluid Dynamics (CFD), particularly with scale-resolved methods such as Direct Numerical Simulations (DNS) and Large-Eddy Simulations (LES). DNS is widely regarded as the most accurate and reliable approach for modeling unsteady flows. However, its computational cost is still prohibitive, limiting its application to simplified or academic configurations. LES, on the other hand, offers a more feasible alternative by resolving the large, energy-carrying turbulent structures while modeling the smaller subgrid-scale turbulence. While LES is promising for industrial applications, it still faces significant computational challenges, especially in high-Reynolds-number flow scenarios. In this context, Wall-Modeled Large Eddy Simulation (WMLES) has emerged as a practical compromise between accuracy and computational efficiency. WMLES, in fact, retains the scale-resolving capability of LES in the outer flow regions, where large turbulent structures dominate, but models the near-wall regions to significantly reduce the grid resolution requirements. This approach allows for the simulation of complex flows at realistic Reynolds and Mach numbers, making the approach a viable solution for addressing industrial challenges. 

Turbulence modeling is one of many challenges when addressing industrial flow applications; geometric complexities also pose significant obstacles, particularly in pursuing fast and computationally efficient flow solvers. Body-fitted approaches are traditionally regarded as the most accurate methods for resolving wall-dominated flows, but they present challenges in data management and are difficult to integrate with modern GPU-accelerated architectures. In contrast, finite difference methods over Cartesian meshes, combined with Immersed Boundary Method (IBM) approaches, have emerged as a highly effective alternative. Thus, by combining the strengths of LES in capturing large-scale turbulence with the computational efficiency of wall modeling, the WMLES+IBM framework has already demonstrated remarkable accuracy in reproducing the flow conditions of real-world gas turbine cascades operating in transonic regimes [1]. 

However, the authors' previous studies primarily focused on the aerodynamic characteristics of turbine cascades, with limited attention given to the role of wall temperature, leaving its impact on flow dynamics largely unexplored. This research addresses that gap by focusing on the influence of wall temperature on the aerodynamic behavior of turbine cascades operating under transonic conditions. Using the same IBM+WMLES framework, this paper investigates how variations in wall temperature affect boundary layer dynamics, shock-boundary layer interactions, and overall aerodynamic losses in a fully scale-resolved approach typical of LES methods. The findings reveal that lower wall temperatures accelerate boundary layer transition and amplify shock-induced separations, resulting in increased aerodynamic losses. Conversely, higher wall temperatures smooth boundary layer behavior, delay separation and reduce losses, although at the expense of reduced cooling efficiency. This behavior is analyzed through a detailed examination of the underlying physics across multiple scales of motion, as highlighted by the scale-resolved framework. Thus, the results of this study offer crucial insights for optimizing turbine cooling strategies and enhancing aerodynamic performance in transonic turbines, advancing the current understanding of the complex interplay between thermal and aerodynamic factors in gas turbines. 

De Vanna, F, & Benini, E. "Towards New Insights in Gas Turbine Aerothermodynamics With Wall-Modeled LES and Immersed Boundary Method." Proceedings of the ASME Turbo Expo 2024: Turbomachinery Technical Conference and Exposition. Volume 12B: Turbomachinery — Axial Flow Turbine Aerodynamics; Deposition, Erosion, Fouling, and Icing. London, United Kingdom. June 24–28, 2024. V12BT30A001. ASME. https://doi.org/10.1115/GT2024-121022

Presenting Author: Francesco De Vanna Università degli Studi di Padova

Presenting Author Biography: Francesco De Vanna graduated with honors in Mechanical Engineering in 2016 and obtained his Ph.D. in Computational Fluid Dynamics from the University of Padova 2020. He currently serves as an assistant professor in Machine Fluid Dynamics at the Department of Industrial Engineering at the University of Padova. His research interests focus on advanced flow modeling and numerical methods for compressible flows, with particular attention to LES and Wall-Modeled LES approaches in machine components. Francesco is also the lead author of URANOS, a massively parallel, GPU-accelerated Navier-Stokes solver, which is distributed as open-source software.

Authors:

Francesco De Vanna Università degli Studi di Padova
Ernesto Benini Università degli studi di Padova