Computational Fluid Dynamics Simulations of High Reynolds Number and Complex Aerodynamic Flow Fields

Date

2025-2-27

Author

Orbay Akcengiz, Ezgi

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

71
views

0
downloads

Cite This

This thesis presents a comprehensive investigation of 2-D and 3-D complex external flow fields in aerodynamics and wind engineering, focusing on wind turbine airfoils at high Reynolds numbers and the blunt-body aerodynamics of tall buildings using CFD simulations performed by SU2. The performance of Krylov subspace iterative solvers used in these simulations is also investigated. To improve SU2’s efficiency and robustness in solving large-scale linear systems, a nested Krylov-based linear solver is proposed and validated through benchmark problems, showing accurate results with higher convergence rates and lower computational cost compared to the standard solver in SU2. This thesis study firstly focuses on high Reynolds number flows around the DU 00-W-212 wind turbine airfoil. The detailed CFD simulations are performed from 2-D steady-state analyses to 3-D time-accurate simulations with hybrid IDDES turbulence modeling. The challenges in the RANS and DDES-based CFD analyses of such problems are stated. The procedure for a detailed grid independence study for airfoil CFD simulations at high Reynolds numbers is presented and discussed. The aerodynamic polars are obtained and discussed for three different high Reynolds numbers, and also the results for the flow field characteristics such as transition, stall and separation are presented in detail. The proposed nested Krylov-based linear solver is tested with the 2-D and 3-D CFD simulations. The 3-D turbulent and transonic flow around the ONERA M6 wing, a common validation test case used for CFD codes, is used as a validation test case. The steady-state RANS CFD simulations of 3-D complex flow fields around tall buildings are performed. The results are compared with wind tunnel measurements performed at METU RÜZGEM. The computational performance of the standard and nested Krylov subspace linear solvers are compared regarding their accuracy in predicting the aerodynamic properties and their effects on computational cost, speed-up, and convergence rate. The efficiency and robustness of the nested Krylov-based linear solver are observed. This thesis not only contributes to the understanding of complex aerodynamic flow problems but also improves the computational efficiency by implementation of the nested Krylov linear solver in SU2 code, thus presenting a robust and efficient linear solver for the 2-D/3-D steady/unsteady CFD simulations of complex aerodynamic problems at high Reynolds numbers and also with separated flow fields.

Subject Keywords

Iterative Solvers, Krylov Subspace Linear Solvers, Computational Fluid Dynamics (CFD), Laminar to Turbulent Boundary Layer Transition, Delayed Detached Eddy Simulation (DDES), Wind Turbine Airfoil, Twisted Tall Building, Aerodynamics

URI

https://hdl.handle.net/11511/114110

Collections

Graduate School of Natural and Applied Sciences, Thesis