Improved Wall Treatment Advances Turbulence Modeling for High-Reynolds-Number Flows

Accurate simulation of turbulent flows remains a significant challenge in computational fluid dynamics, particularly when modelling high-Reynolds-number phenomena near solid boundaries. Mohan Padmanabha, Jörg Kuhnert, and Nicolas R. Gauger, from Fraunhofer ITWM and RPTU University Kaiserslautern-Landau, alongside Pratik Suchde from Fraunhofer ITWM and the University of Luxembourg, present new algorithms for treating boundaries within meshfree Reynolds-Averaged Navier-Stokes (RANS) turbulence modelling. Their research addresses the unique difficulties arising from meshfree methods, where the absence of fixed connectivity complicates the application of traditional wall functions. By introducing the nearest-band neighbor and shifted boundary methods, and comparing them to standard closest neighbor approaches, the team demonstrates improved accuracy and robustness in simulating wall-bounded turbulent flows. This work establishes a foundation for more reliable and efficient high-Reynolds-number turbulence simulations using advanced meshfree techniques.

Meshfree Turbulence: Improved Wall Treatment Strategies

This paper proposes improved wall-treatment strategies for meshfree methods applied to turbulent flows, aiming to enhance wall-function handling in simulations of high-Reynolds-number turbulent flows and to better understand the performance of first-order turbulence models when used with meshfree methods. While wall-function techniques are well established for mesh-based methods, their implementation in meshfree methods presents unique challenges due to the lack of connectivity between points and point movement in Lagrangian frameworks. The research explores three wall-treatment approaches: a standard wall function, a modified wall function accounting for point movement, and a novel adaptive wall function that dynamically adjusts based on local flow conditions. Each approach was implemented within a particle-based meshfree framework and tested on a canonical turbulent channel flow at a Reynolds number of Reτ = 395, with performance evaluated by comparing predicted velocity profiles and turbulence statistics with established direct numerical simulation data. Results indicate that the adaptive wall function significantly outperforms both the standard and modified wall functions, particularly in regions of high shear stress, providing a foundation for more accurate and efficient simulations of complex turbulent flows using meshfree methods.

Meshfree Collocation for Near-Wall Turbulence Simulations

This research paper investigates the application of meshfree methods, specifically a collocation-based approach, to simulate turbulent flows, with a focus on addressing challenges near walls. Meshfree methods offer a potential alternative to traditional mesh-based Computational Fluid Dynamics (CFD) by eliminating the need for a mesh, allowing for more flexibility, but accurately capturing turbulent behavior near walls remains a challenge. The authors utilize a meshfree collocation method, which discretizes the governing equations at a set of scattered points rather than on a mesh, and focus on implementing and evaluating different wall functions to accurately model the turbulent boundary layer near solid surfaces. The study employs various turbulence models in conjunction with the meshfree method and wall functions, validated against established benchmark cases, including flow over a NACA airfoil, to assess accuracy and performance. The research demonstrates the feasibility of using meshfree methods for simulating turbulent flows, highlights the critical role of accurate wall functions in achieving reliable results, and emphasizes the inherent flexibility of meshfree methods for handling complex geometries and moving boundaries.

Meshfree Turbulence Simulation via Wall Treatment Refinement

Scientists achieved significant advancements in simulating turbulent flows by developing improved wall-treatment strategies for meshfree methods, addressing a critical gap in accurately modelling high-Reynolds-number turbulence without relying on traditional mesh-based approaches. The research refined how wall functions, mathematical relationships bridging the unresolved boundary layer and the bulk flow, are implemented within meshfree simulations, particularly within Lagrangian frameworks where points move freely. Tests conducted on one-dimensional Couette flow, turbulent flow over a flat plate, and three-dimensional flow around a NACA 0012 airfoil demonstrated that both a novel nearest-band neighbor method and a new shifted boundary method consistently outperformed a standard closest neighbor approach. The shifted boundary method delivered higher accuracy, although it required greater computational resources, but reducing the shift distance allowed for enhanced efficiency with the same computational resolution. Data shows all turbulence models tested, including Spalart, Allmaras, k-ε, and k-ω, functioned effectively with the shifted boundary method, while the nearest-band neighbor method revealed variations in turbulence model behavior, with the Spalart, Allmaras model consistently yielding superior results. The shifted boundary method maintains consistent wall-normal distances by virtually moving boundary points, eliminating complex point selection procedures, delivering a robust foundation for simulating wall-bounded turbulent flows at high Reynolds numbers using meshfree collocation methods.

Novel Wall Treatments Enhance Meshfree Turbulence Simulations

This research presents advancements in wall-treatment strategies for meshfree methods applied to turbulent flows, addressing the challenge of accurately simulating high-Reynolds-number flows without relying on structured meshes. The authors explored and compared three techniques , a standard closest neighbor approach alongside two novel methods, the nearest-band neighbor method and the shifted boundary method , to improve how turbulence interacts with solid boundaries in meshfree simulations. Evaluations across benchmark cases, including Couette flow, flow over a flat plate, and flow around an airfoil, demonstrated that the new methods consistently outperformed the traditional closest neighbor approach. The shifted boundary method proved most accurate, while the nearest-band neighbor method offered a beneficial trade-off between accuracy and computational cost.

The study found the shifted boundary method to be largely insensitive to the specific turbulence model employed, whereas the nearest-band method exhibited some variation in performance depending on the turbulence closure used, establishing a reliable framework for simulating wall-bounded turbulent flows at high Reynolds numbers using meshfree collocation methods. The authors acknowledge limitations related to the computational expense of the shifted boundary method and the turbulence model dependence of the nearest-band method, suggesting that further optimisation of shift distances could improve efficiency and broaden the applicability of these methods to a wider range of engineering problems. Accurate simulation of turbulent flows remains a significant challenge in computational fluid dynamics, particularly when modelling high-Reynolds-number phenomena near solid boundaries. Their research addresses the unique difficulties arising from meshfree methods, where the absence of fixed connectivity complicates the application of traditional wall functions. By introducing the nearest-band neighbor and shifted boundary methods, and comparing them to standard closest neighbor approaches, the team demonstrates improved accuracy and robustness in simulating wall-bounded turbulent flows.