On 11/16/2016 03:30 PM, thomas stephens wrote:
but I'm not sure it will redistribute mesh points in tangential
directions. I think there's a second step to this that moves vertices
around.
The important realization is that there are infinitely many ways of
parametrizing the same surface. That's easy to see if you just think
about a curve C that you want to describe as a function (x(s),y(s))
mapped from a reference domain s \in [0,1]. There are many such
functions x(s), y(s) that lead to the same curve. Some move along C
slower in the beginning and faster at the end, some do it the other way
around. One particular one moves at constant speed -- that's the one
that uses the arc length (times a constant) for the parameter s.
The same is true with Eulerian mappings: If you move some of the nodes
tangentially along the surface, it's still the same surface. (At least
up to the discretization accuracy.)
Best
W.
--
------------------------------------------------------------------------
Wolfgang Bangerth email: [email protected]
www: http://www.math.colostate.edu/~bangerth/
--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see
https://groups.google.com/d/forum/dealii?hl=en
---
You received this message because you are subscribed to the Google Groups "deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
