Hi Kun-Dar,

There is actually someone using the 4th order term in the way you
highlight below and seems to be happy with it, so it may work. I would
still treat it with suspicion and make sure you have a really good
test case for it before you have convinced yourself it works. We have
absolutely no test for using it in this fashion and I haven't tried in
any on my work. Also, even if it works, I'm not sure what happens with
boundary conditions so you'll have to treat those carefully too (that
goes for all anisotropic coefficients). If you come up with a
relatively simple test case and you don't believe it is working then
I'll be happy to take a look at it and see if it would be possible to
get it working in that case.

Cheers

2009/6/12 Kun-Dar Li <[email protected]>:
> Dear All,
> Recently, I just updated FiPy to version 2.0, and try to solve the
> anisotropy 4th order PDE, k1 is tensor or vector, and k2 is scalar.
> From the User's Guide, I knew currently FiPy does not support the anisotropy
> higher order in the form of DiffusionTerm([[1, 0], [0, 1]]). But in this
> case, it will like: DiffusionTerm([[1, 0], 1]). Does this work in FiPy?
> If not, do I have other way to describe this anisotropy problem? Please give
> me some hints. Thanks in advance. Have a good weekend.
>
> Kundar



-- 
Daniel Wheeler

Reply via email to