Since its introduction, Automatic Mesh Sizing has greatly simplified the modeling process. Automatic Mesh Sizing defines a mesh that is optimized for the model and accurately represents every detail of the geometry.
A good representation of the geometry is only one requirement for a high fidelity solution. For example, the flow on a simple model with a uniform mesh can contain significant gradients. The results on the coarse mesh may not be highly accurate, but they do indicate flow trends. These trends can be used to identify where the elements should be concentrated to improve solution accuracy.
Mesh Adaptation uses solution results to progressively improve the mesh definition. The simulation is run several times. Each time the results in the previous cycle are used to improve the mesh in the next cycle. The result is a mesh that is optimized for the particular simulation. The mesh is finer for high gradient regions, and coarser elsewhere.
When Mesh Adaptation is enabled, the following occurs:
The result is an intelligently refined mesh that is tuned for the flow and temperature results fields.
Mesh Adaptation does not support the following simulation types or settings:
When using Mesh Adaptation, it is important to know when the solution has achieved mesh independence. Mesh independence is the coarsest mesh that produces results that do not change after the mesh is refined. During each adaptation step, Autodesk® CFD evaluates the pressure, velocity, and temperature fields to determine how close your solution is to mesh independence. At the conclusion of each step, Autodesk® CFD reports the mesh independence status in the Output bar in a message similar this:
Mesh independence: Pressure: 85% Velocity 98% Temperature 97%
These values indicate how converged your simulation is for each quantity. Higher values indicate less sensitivity to the mesh, which means your solution is approaching mesh independence.
For best results, you should ensure that each adaptation cycle runs to convergence (or as close as possible). If your adaptation cycles only run for a small number of iterations, this failing to reach convergence, the Mesh Independence indicators will not be accurate.
You should define an initial mesh that follows standard mesh best practices. This helps to arrive at the mesh independent solution more efficiently.
Because Mesh Adaptation runs several cycles to arrive at the final mesh, the complete analysis can take longer to run than a scenario run without Adaptation. Additionally, the overall element count can grow considerably larger than the original mesh. Be careful with complex models that are initially large, as they will take much longer to run. The benefit, however, is that the resultant solution is mesh independent.
When you click Stop on the Solve dialog, the current cycle stops, and no further adaptation cycles are run. To run additional adaptation cycles after the simulation has stopped, click Enable Adaptation on the Adaptation tab. The solver runs the prescribed number of Cycles to Run.
Note that Adaptive Meshing works for both Automatic Mesh Sizing and Manual Mesh Sizing.
The folder that contains the scenario must exist in the Design Study folder structure. Results must exist within the scenario folder to apply mesh adaptation to a scenario that has been run.
When a simulation is continued from iteration 0, all saved results are deleted and Adaptation is disabled. To adapt the mesh for the current design, first run the current scenario with Adaptation disabled. Upon completion, enable Adaptation, and continue the simulation. The completed scenario becomes the baseline for subsequent mesh adaptation cycles.
Every cycle runs either the number of specified iterations or until convergence is detected by Automatic Convergence Assessment. Because finer meshes typically require more iterations to converge, it is good practice to specify at least 300 iterations on the Solve dialog. This ensures that every mesh in the process can attain convergence.