Topology Optimization of Turbulent Flows Using a Parametric Level-Set Method with Implicit Wall Functions
Please login to view abstract download link
Turbulent flow is widely present in various industrial applications, and its flow properties are crucial to system efficiency and reliability. However, due to the strong nonlinearity of turbulent flow and the limitations of traditional turbulence topology optimization methods in handling precise wall boundary conditions, a significant performance gap exists between the optimized structure and the reconstructed model. Topology optimization for turbulent flow still faces many challenges. This paper proposes a parametric level-set TO framework for turbulent flows, in which implicit wall function is introduced. In this research, fixed coarse mesh is applied during the topology optimization process. Various two-dimensional design problems are investigated to demonstrate the effectiveness of the proposed approach, including a classical pipe-bend benchmark and lift-to-drag trade-off optimization problems. Preliminary results indicate that the proposed framework consistently maintains a relatively well-defined fluid-solid interface, which enables the stable application of implicit wall functions and leads to enhanced robustness during iterations. Besides, the proposed framework exhibits strong robustness and stability, even at high Reynolds numbers up to Re=106, and remains effective on relatively coarse meshes. The proposed topology optimization framework is implemented through a coupled COMSOL Multiphysics and MATLAB environment.
