"Igor" wrote in message <firstname.lastname@example.org>... > Thank you Nasser for such quick reply. > > > how many elements are you using? > 16 elements which concentration I would like to find with oscillating BC > > > > As I know pdepe uses FEM method. Am I right? > > yes. > Where can I read about method in details which used for pdepe? >
Skeel, R. D. and M. Berzins, "A Method for the Spatial Discretization of Parabolic Equations in One Space Variable," SIAM Journal on Scientific and Statistical Computing, Vol. 11, 1990, pp.1?32.
> >> Can we trust them? > > Do not trust any one. > I agree. And in this article they used only steady state conditions. > > 1. Maybe pdepe not so good for oscillating? > > 2. What do you think about using FD method for this task?
The discretization method is not important in 1d.
> > 3. How can I calculate the accuracy of pdepe?
Double the number of discretization points and look how much the results change.
> > 4. What tools can I use for optimization of calculation speed? Is there any sense to use Matlab runtime compiler and import it to c++ code? Speed will be better?
Check it out.
> > 5. Can I use somehow parallel computations here?