________________________________ From: Wolfgang Bangerth <[email protected]> To: [email protected] Cc: Pietro Maximoff <[email protected]> Sent: Wednesday, July 29, 2009 4:33:38 AM Subject: Re: [deal.II] 1D Problems > Is there any tutorial floating about on how to use deal.ii to solve 1D > problems. I've just started to learn deal.ii and I'd like to test it > out on some 1D problems. > > I have had a look at Step-15 but at this point it's too complicated (it > is the 15th Step) plus it's tangled with other stuff. Any help is > appreciated. >> 1d is really no different than 2d, pretty much all the code that we show in >> the tutorial would compile equally well for 1d. The only thing that is a >> bit different is the handling of boundary indicators: in 2d and 3d you can >> ask a face (i.e. a line or a quad) what part of the boundary it is on, but >> you can't ask a node that in 1d. There, by default the leftmost node has >> indicator zero, and the rightmost one indicator one. That is pretty much >> the only difference of any significance. >> W. >> ------------------------------------------------------------------------- >> Wolfgang Bangerth email: [email protected] >> www: http://www.math.tamu.edu/~bangerth/ Thanks for that, Wolfgang. I managed to play around with the boundary indicators yesterday. If you don't mind, could you tell me how to also resolve these: 1. For example, yesterday, I tried to solve a 1D problem to which I know the exact solution. The range for the mesh was, say, 1 to 2. The (exact) y-values also run from 1 to 2. However, when I ran it through deal.ii, the more I refined the mesh, the more different the solution values were (in terms of scale) even though the shapes of the plots were the same. Is there something I've forgotten to set? 2. In 1d, is there a way for me to specify different properties for different parts/ranges of the mesh. For example, if my line runs from 0 to 100, I'd like the following: 0 < x < 40 to be for steel with its own specific properties 40 < x < 60 to be for copper and 60 < x <100 to be for steel. Thus, there's a sort of discontinuity which should be reflected in the solution. Apologies if this questions are too trivial. Pietro
_______________________________________________ dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
