I am using preCICE 2.5 to couple my waves2foam and structural solver on Ubuntu22.04.I have verified that the modification of wave model is successful.
It has worked for a long time, but under the condition of 0.01 time step, it took 2403s ClockTime to calculate 0.12s data and it reports an error advance() cannot be called with a timestep size of 0
I use the adapter included in the example,precice-config.xml has not been modified
Has anyone encountered this problem?
Thanks in advance
How is the Courant number and the adaptive time step size calculation going? I can imagine that OpenFOAM is trying to do an unreasonably small timestep, which preCICE then interprets as zero.
Thank you for your reply, This is my controlDict file.
I have tried to set writeControl to timeStep It does resolve this error. But I don’t know whether the time step is getting smaller and smaller because of the dynamicMesh.
controlDict.txt (1.5 KB)
I didn’t delve into the causes of the above problems. I continued to simulate after modifying the controDict file, and it was successfully completed. However, the result of simulated distortion appears. I check the log file and find that the residual is not very large, perhaps not computational divergence. At present, I haven’t found the problem. I hope you can give me some suggestions
fluid_log.txt (43.9 KB)
Because the original log file is too large, I have intercepted the contents of the front and rear parts
I can imagine that what you experience here is related to accumulating errors over time, since probably OpenFOAM is doing extremely small time steps.
In controlDict, these are the relevant parts:
deltaT 0.01;
adjustTimeStep yes;
maxCo 0.25;
maxAlphaCo 0.5;
maxDeltaT 1;
I would try to test an uncoupled simulation with the same settings (e.g., comment out the adapter function object in the controlDict) and see if it shows the same behavior.
VoF simulations are generally very sensitive. The moving flap (changing cell size) could make this even more challenging.