[Nek5000-users] Solving mesh Laplace equation (Re: Mesh morphing)

nek5000-users at lists.mcs.anl.gov nek5000-users at lists.mcs.anl.gov
Fri Feb 20 11:39:17 CST 2015


Dear Neks,

What I actually would like to accomplish is to solve the Laplace  
equation for the mesh (with the boundary deformation as a BC). I saw  
that this has been done already at least by Paul (a thread 2009) and  
Matt (for smoothing of a wing mesh, with zero deformation).
Could any of you please give me a hint on how to solve the Laplace  
equation for the mesh like this? Matt? Paul?
I would be very grateful.

Best regards,
Outi

Quoting Outi Tammisola <outi at mech.kth.se>:

> Dear Neks,
>
> I thought this question might be of interest for someone else as  
> well, so posting it here:
>
> I have been looking at the oscillating cylinder example. I would  
> like to do something similar, but instead of a prescribed boundary  
> velocity, would like to prescribe a given displacement of one  
> boundary, and morph my mesh smoothly everywhere to match this  
> displacement. For example, if one were to describe a deformation of  
> the cylinder rather than a velocity (but without altering the other  
> boundaries and without scaling the mesh). This just needs to be done  
> once, not at every time step.
>
> Is there an easy way to do this? Can I arrive there somehow by minor  
> modifications of the elasticity solver? What equation exactly is the  
> elasticity solver solving at every time step?
> (When I outputted the files after every time step for the  
> oscillating cylinder, the mesh deformation seemed to match the  
> prescribed one only after 2 time steps. Is there a reason for this,  
> or did I do something wrong?)
>
> Best regards,
> Outi




More information about the Nek5000-users mailing list