Some questions about solving gradients

[15175829310]

May I use the finite difference method to calculate the gradient and validate the adjoint method? How should I operate？

[15175829310]

I have read your previous replies to others. Do you have any cases related to flow field inversion

[friedenhe]

There is no automatic way to do FD for field inversion because there are so many inputs. What you can do is to manually perturb one beta value in the runScript.py file and run the primal to get the FD derivative. Check where you prescribe the initial values for beta in the runScript.py file. You can prescribe some perturbed beta values there.

[15175829310]

Sorry, do you have any reference scripts? I'm not sure how to operate them

[15175829310]

I saw an example of the periodic hill in your article before and would like to know how you wrote the script
“An Open-source Adjoint-based Field Inversion Tool for Data-driven RANS Modelling”

[friedenhe]

https://github.com/DAFoam/tutorials/blob/607a4f7738d067dd0921eecd4a96750c69443811/Ramp/steady/train/runScript_FI.py#L201
Here is where you set the initial value for beta val=np.ones(nCells), you can set a different initial value by manually perturbing beta at for certain cell.