simtk.org

Hello all,

After running the Coronary Normal demo project, I noticed that the flow rates in all_results-flows.txt and all_results-averages.txt don't add up to zero. There's a net flow rate of a few ml/s, which is only about 1% of the flow rate in the aorta, but it's considerable compared to the flow rates in the coronary arteries. Is there a solver setting that I should set to a more strict value to avoid this?

Thanks!
Sita

In the meantime I've been experimenting with this Residual Criteria parameter, which does seem to help improving mass conservation. I was wondering, how is the residual norm defined here? And apart from reducing the value of Residual Criteria, would it also help to increase those Maximum Number of Iterations parameters (under linear solver options)? And/or are there other useful settings to tune? I've gone through the Simulation Guide and the SimVascular paper by Updegrove et al. (2017), but I'm not very familiar with the details of the solver.

Thanks!
Sita

Generally, there are two things you can try.
1. You can consider reducing your mesh size for the coronary artery to improve mass conservation.
2. If I remember correctly, the solver adds the squares of continuity and momentum residuals for each node and then calculates the root of mean squared residual. This value is printed in the third column of histor.dat. If your residual is low, you should expect to see good mass conservation. You can increase the number of nonlinear iterations in order to reduce the residual. In solver.inp, "Step Construction : 0 1 0 1 0 1" corresponds to 3 iterations per timesteps. Just adding additional pairs of 0 1 to increase the number of iterations at each time step.

Thanks a lot, that's really helpful!
Sita