Discussion Forum

September 13, 2010 8:49pm UTC

2D Navier-Stokes w/ moving interior boundary

Hello all,

I'm trying to simulate a 2D actuator pump that is a rectangle with a moving interior boundary which moves like a propagating wave constrained at one end. Water fills the two subdomains partitioned by the interior boundary. My goal is to solve the navier-stokes equations and investigate the pressure at the constrained end of the moving interior boundary. My method henceforth has been to try to solve for the mesh velocity w/ appropriate boundary conditions and then solve for the ns variables (u,v) using the stored mesh velocities as input for the ns boundary conditions on the interior boundary.
I have been able to successfully calculate the moving mesh velocities, but when I go to solve for the ns variables, I get an error message relating that the solver "failed to find consistent initial boundary conditions" and that the last time step did not converge. These obstacles, along with many inverted mesh warnings, are frustrating me. Does anyone know of a better way to solve this model/ have suggestions for how to correct these errors?
Thanks,
Robert Polackwich

Reply  |  Reply with Quote  |  Send private message  |  Report Abuse