hello i am trying to run a simulation over half a front wing with a rotating wheel behind it. Even after 2000 iterations, my simulation doesn’t seem to be converging (the residual plots aren’t anywhere near 10^-6. After around 500 iterations, the residuals just remain flat. Can you have a look at it and tell me where am I going wrong? (my project is in 'Copy of 20" (V2))
Hi, although residuals are definitely important, I would highly recommend that you monitor physical parameters of interest, as well as inspect the results in the post-processor for convergence evaluation.
Focusing solely on residuals for convergence assessment poses some problems, including:
- Type of normalization that is used (which can differ from software to software)
- Initial level of residual (raw residual) that you had at the beginning of the run may impact how much residuals can drop in the first place
- Quantities of interest can stabilize at greater or smaller residuals depending on the run
Residuals in the 10e-6 range for all equations is something that you’re typically going to see for simple cases physics-wise. As you start to add more elements, residuals for some equations will stay at higher levels.