Discussion Forum

Hello,
I'm trying to model a very flexible membrane wing that has been significantly pretensioned and is at a fixed angle of attack in a wind tunnel test section. I've tried a handful of approaches and none of them have worked out, so I'm hoping someone here can suggest what I'm doing wrong. I'm using the FSI module.

First, I tried straining the membrane manually, within the FSI module, with a combination of prescribed displacements, rollers, and a fixed point. If I only model the membrane (and not the fluid) with this approach, it strains exactly as I would expect. However, if I then add the fluid modeling, I get an error that a bunch of nodes giving "NaN/Inf in the vector for the variable x". But I can't see why that would happen, if it worked when only the membrane was being modeled...

Next, I tried using the "Initial stress/strain" condition under the Linear Elastic Material definition in the FSI module. To get the appropriate stress values, I split the study into two stationary steps. In the first step I used Solid Mechanics module to solve for the membrane stress tensor. Then, in the second stationary step, I used the Domain ODEs and DAEs module to get the stress tensor values from the first step, and the FSI module with the initial stress defined from those values. It solves, but the results don't look like I would expect. There's no cambering of the membrane (i.e., curvature due to the aerodynamic pressure), and the membrane dimensions are somewhere in between the initial and prestressed values.

I've attached a model, illustrating both these approaches in the two studies. Any suggestions or pointers would be appreciated!

Regards,
Jillian