Thanks for your answer. I know that there's no need to solve advection problem like that. Actually I'm going to add some more complicated terms to the advection term and see if it works well. That's why I'm about to verify that in the simplest way and then generalize that to 2D and 3D.
Thanks, Mohsen On Thu, Jun 23, 2011 at 7:21 PM, Wolfgang Bangerth <[email protected]>wrote: > > Mohsen, > > > I am a beginner user of deal.ii. for the work I am going to deal with I > > think I should start with Step-9. I have start with the simplest case and > > then try to improve my code. That's why I need to run the step-9 in 1-D. > > But, as I tried to run it, I saw that in the code, there's an object like > > "QGauss<dim-1> face_quadrature_formula(2);" which makes error when > solving > > for 1-D. can anybody help e how to resolve this problem? > > Yes, 1d often behaves differently: > > http://dealii.sourceforge.net/index.php/Deal.II_Questions_and_Answers#deal.II_programs_behave_differently_in_1d_than_in_2.2F3d > > The problem is that in 1d, integrating over a face makes no real sense: > it's > just a point evaluation, and consequently, the FEFaceValues class you try > to > use doesn't work in 1d. There isn't a real simple solution for this right > now > apart from coding these face terms by hand. > > Why is it that 1d problems are of particular interest to you? A 1d > advection > equation is simply an ODE for which many good solvers exist. > > Best > W. > > > ------------------------------------------------------------------------- > Wolfgang Bangerth email: [email protected] > www: http://www.math.tamu.edu/~bangerth/ >
_______________________________________________ dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
