Hi Dan, For some occasions, $\partial_y \partial_z u$ is not equal to $\partial_z \partial_y u$. For example, a random field of u or when u is not continuous. But it seems to me that finite difference method only solve for continuous field. So you are right, I have little luck in this case.
Zhiwen On Tue, Dec 16, 2008 at 1:55 PM, Daniel Wheeler <[email protected]>wrote: > > On Tue, Dec 16, 2008 at 11:27 AM, Zhiwen Liang <[email protected]> > wrote: > > > So every element in the mesh should have diffusion coefficient like this: > > D=[0 0 1 > > 0 0 0 > > -1 0 0] > > Does this also lead to 0*x=b? > > I think so. The equation doesn't make sense, > > \partial_y \partial_z u - \partial_z \partial_y u > > is trivially zero. Right or am I losing my mind? So you are trying to > solve "0 = \alpha". > > >> > Hope all is going well. > >> > Recently, I am trying to solve for an equation like this: > >> > \frac{\partial^2 u}{\partial y \partial z}-\frac{\partial^2 > u}{\partial > >> > z > >> > \partial y}=\alpha > > The above is essentially "0 = \alpha". Do you agree? > > -- > Daniel Wheeler > >
