Discussion Forum

Hi all,

I’m trying to simulate blood flow through a straight artery with linearly elastic walls using FSI in COMSOL 5.1.

I can get results from a stationary study with no issues, but when I use the identical boundary conditions to run a time-dependent study, I invariably get the error message “Failed to find consistent initial values. Last time step is not converged” after approximately 20 iterations.

I’ve tried various things to troubleshoot, including:

• Using the stationary study results as initial values for the time-dependent study
• Ramping velocity using a step function
• Decreasing initial time step to 1e-10
• Decreasing relative tolerance to 1e-5
• Adding a pressure point constraint
• Using a pressure inlet instead of velocity inlet

Additionally, when running the time-dependent study (but not the stationary study), I get a warning about an inverted mesh element: “Inverted mesh element near coordinates [coordinates very near (0,0,0)]”,

for which I’ve tried:

• Making the mesh 'fine' instead of 'normal'
• Making the geometry shape order linear
• Using Automatic Remeshing based on mesh quality and on distortion
• Using Adaptive Mesh Refinement

but all to no avail. I continue to get the inverted mesh element warning, and a failure to find consistent initial values.

I’m not sure whether there might be an issue with my boundary conditions, but if so, I can’t figure out what it might be!

The .mph file is attached. Any help would be greatly appreciated!

Thanks in advance,

Emma.