Restart file solution gives weird outputs

raviramesh10 · March 17, 2021, 04:12

Hello all,

I am currently trying to obtain the plots of $C_{l}$ and $C_{d}$ vs. convective time units (non-dimensional) for an airfoil in deep stall (having an angle of attack of 60) using the DDES approach, using the SA turbulence model and a 2nd order numerical scheme (dual-time stepping scheme).

I have an issue with plotting the results and comparing them with the ones obtained from experimental data from Swalwell as attached with the comment below. For some reason, when I restart my solution from somewhere in between, there is a mismatch in the residuals that I obtain. This does not seem to make sense from the numerical perspective, as the solution should simply start from where the solver left off.

Furthermore, the fluctuations in $C_{l}$ and $C_{d}$ seem to have diminished for the given time traces, when there is actually supposed to a significant deviation from a certain mean value for both coefficients. Is this on account of high numerical dissipation in the current scheme?

Hoping to hear from you soon.

Ravi

raviramesh10 · March 24, 2021, 03:04

Quote:

Originally Posted by raviramesh10

Hello all,

I am currently trying to obtain the plots of $C_{l}$ and $C_{d}$ vs. convective time units (non-dimensional) for an airfoil in deep stall (having an angle of attack of 60) using the DDES approach, using the SA turbulence model and a 2nd order numerical scheme (dual-time stepping scheme).

I have an issue with plotting the results and comparing them with the ones obtained from experimental data from Swalwell as attached with the comment below. For some reason, when I restart my solution from somewhere in between, there is a mismatch in the residuals that I obtain. This does not seem to make sense from the numerical perspective, as the solution should simply start from where the solver left off.

Furthermore, the fluctuations in $C_{l}$ and $C_{d}$ seem to have diminished for the given time traces, when there is actually supposed to a significant deviation from a certain mean value for both coefficients. Is this on account of high numerical dissipation in the current scheme?

Hoping to hear from you soon.

Ravi

Hello all, so as a follow-up to the previous post that I had put up last week, I had found out that this error was due to the fact that I had been plotting the time averaged values of $C_{l}$ and $C_{d}$ all this while for this case.

I had also run a URANS k- $\omega$ model simulation to compare my results with them. It seems like the DDES model does not have an option of plotting the $C_{l}$ and $C_{d}$ values for each time step in the history file. I tried looking under the options of AERO_COEFF, but the history file only seems to be writing the time averaged values for DDES simulations and not for URANS ones, and hence, the fluctuations in $C_{l}$ and $C_{d}$ are captured in the latter but not the former. Can someone please help me out with this?

I have attached relevant snapshots of the history as well as the configuration files for the two cases with the post for reference.

Hoping to hear from the others soon.

Ravi

pcg · March 25, 2021, 06:08

From your first plot it looks like you are plotting all the inner iterations.
But without more information it is impossible to help.
Attach your config, the screen output..., the history file..., otherwise the only people in the world that might understand the problem you are describing are the ones who already had it.

raviramesh10 · March 25, 2021, 06:45

Quote:

Originally Posted by pcg

From your first plot it looks like you are plotting all the inner iterations.
But without more information it is impossible to help.
Attach your config, the screen output..., the history file..., otherwise the only people in the world that might understand the problem you are describing are the ones who already had it.

Thank you for your reply, Pedro. Indeed, I did not put up the original files, and I will share it again. Hope it helps better with understanding the problem at hand.

With warm regards,

Ravi

raviramesh10 · July 1, 2021, 08:00

Quote:

Originally Posted by raviramesh10

Thank you for your reply, Pedro. Indeed, I did not put up the original files, and I will share it again. Hope it helps better with understanding the problem at hand.

With warm regards,

Ravi

Dear all,

As a late update to this, this issue was caused because of the fact that the time-averaged windowing part was turned on, because of which the instantaneous solution remained more or less constant with the number of time iterations.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
how to calculate mass flow rate on patches and summation of that during the run?	immortality	OpenFOAM Post-Processing	104	February 16, 2021 08:46
[OpenFOAM] Annoying issue of automatic "Rescale to Data Range " with paraFoam/paraview 3.12	keepfit	ParaView	60	September 18, 2013 03:23
[swak4Foam] Error bulding swak4Foam	sfigato	OpenFOAM Community Contributions	18	August 22, 2013 12:41
[swak4Foam] build problem swak4Foam OF 2.2.0	mcathela	OpenFOAM Community Contributions	14	April 23, 2013 13:59
friction forces icoFoam	ofslcm	OpenFOAM	3	April 7, 2012 10:57

March 25, 2021, 06:08		#3
pcg Senior Member Pedro Gomes Join Date: Dec 2017 Posts: 465 Rep Power: 13	From your first plot it looks like you are plotting all the inner iterations. But without more information it is impossible to help. Attach your config, the screen output..., the history file..., otherwise the only people in the world that might understand the problem you are describing are the ones who already had it.