How do you deal with complex geometries and meshing issues in turbomachinery simulations?

1 Why are turbomachinery simulations challenging?

Turbomachinery simulations involve solving the fluid dynamics and heat transfer equations in three-dimensional domains that often have curved, twisted, and rotating surfaces. These geometries require high-quality meshes that can capture the flow features and boundary layers accurately and efficiently. However, generating such meshes can be time-consuming, error-prone, and computationally expensive. Moreover, the mesh quality can affect the convergence, stability, and accuracy of the numerical solution, as well as the post-processing and analysis of the results.

2 How to choose the right meshing strategy?

When simulating turbomachinery, there is no one-size-fits-all meshing strategy. You need to consider several factors, such as the geometry complexity, flow regime, physics model, solver settings, and performance requirements. Common meshing strategies include structured mesh, unstructured mesh, and hybrid mesh. Structured mesh is ideal for simple and smooth geometries due to its high resolution and low numerical diffusion. However, it is difficult to generate for complex and curved geometries and may introduce artificial singularities or discontinuities. Unstructured mesh is suitable for complex and curved geometries as it can conform to any surface and fill any volume. However, it may have lower resolution and higher numerical diffusion than structured mesh, as well as require more cells and memory. Hybrid mesh combines structured and unstructured elements to take advantage of the benefits of both types of mesh while also inheriting their disadvantages.

3 How to improve the mesh quality?

The mesh quality is a measure of how well the mesh reflects the geometry and physics of the problem. This quality depends on several parameters, such as cell size, shape, aspect ratio, skewness, orthogonality, smoothness, and alignment. Poor mesh quality can lead to numerical errors, convergence issues, instability, and inaccurate results. To improve the mesh quality, you can use appropriate meshing tools and techniques like CAD cleaning or surface meshing. It is also important to check and validate the mesh using quality metrics like minimum and maximum values or contour plots. You can also modify and optimize the mesh with manual or automatic methods like deleting, splitting, merging, or moving cells.

4 How to deal with rotating domains?

Simulations of turbomachinery, such as blades, impellers, or rotors, can be complex due to the relative motion and centrifugal effects. To address these challenges, you should consider the appropriate reference frames, such as inertial, rotating, or moving. Additionally, you should select the right boundary conditions depending on the geometry symmetry and rotation direction. Finally, you should choose a mesh interface based on the level of detail and accuracy desired, as well as the computational cost.

5 How to verify and validate the simulation results?

The verification and validation (V&V) of the simulation results are crucial steps to ensure reliability and credibility. Verification is the process of checking the correctness of the numerical solution, while validation is the process of comparing the numerical solution with experimental or analytical data. To perform V&V, one can use grid convergence studies to assess the sensitivity of the solution to the mesh size and quality, using metrics such as grid refinement ratio, grid convergence index, or Richardson extrapolation. Solution convergence studies can also be used to assess the sensitivity of the solution to solver settings and parameters, using metrics such as residuals, mass flow rate, pressure drop, or efficiency. Further, uncertainty quantification studies can be done to assess uncertainty due to input data, model assumptions, numerical errors, or measurement errors. Methods such as sensitivity analysis, error propagation, and Monte Carlo simulation can be used for this purpose. Turbomachinery simulations are difficult but rewarding tasks that can help you design and optimize your devices. With these tips and tricks in mind, you can confidently handle complex geometries and meshing issues.

6 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?