On the application of Goal Based Mesh Adaptation (GBMA) using the Adjoint solver of Ansys Fluent

In a Computational Fluid Dynamics (CFD) investigation, accurately predicting the flow requires a thorough evaluation of the discretization error. Conducting a mesh sensitivity study is considered a best practice to achieve a grid-independent solution. This involves generating a series of consistently refined computational grids, but the effectiveness of this approach heavily relies on human interaction with pre-processing tools, often requiring advanced mesh generation skills. Additionally, this process is frequently time-consuming, especially when dealing with complex geometries or regions exhibiting intricate flow phenomena that demand high accuracy predictions. Without a careful mesh strategy, it is common to encounter issues such as an excessively refined computational grid, particularly resulting from the refinement of non-key areas. To address these challenges, various mesh adaptation techniques have been developed over the years. Ansys Fluent introduces a new technology called Goal Based Mesh Adaptation (GBMA), which employs the adjoint method and gradient sensitivities to enable automatic mesh adaptation. By calculating adjoint sensitivities, GBMA drives a user-defined quantity of interest toward a grid-independent solution, facilitating optimal grid refinement. This study aims to assess the robustness and effectiveness of the GBMA approach, demonstrated using 2D and 3D external aerodynamic compressible test cases in the high-subsonic and transonic regimes. Overall, the technology proves to be robust, achieving effective mesh adaptation that leads to improved aerodynamic force predictions and up to ×7 reduction in cell count compared to traditional mesh adaptation methods. This technology represents a significant advancement in achieving accurate and efficient simulations within Ansys Fluent.