Okay, I think I get what you mean. I originally thought I do have to use FE_Nothing, which of course seems to be weird. That was my misunderstanding from Step-46. I will try to re-make it in a scalar way and see.
Thanks, Weixiong 在 2017年6月16日星期五 UTC-7上午10:35:47，Wolfgang Bangerth写道： > > > > Physically, the problem is scalar per direction (radiation transport). > > So you are suggesting I don't need to use FE_Nothing in the FESystem? > > I don't think it's become completely clear what you are doing, but from > what I infer is that your problem is *not* in fact scalar -- but a > system of equations each of which is scalar. Is this correct? > > If so, then of course you need an FESystem -- you have a vector-valued > problem (namely a system of equations). But if that is the case then the > result of FEValues::get_function_values needs to be a vector of vectors, > namely the vector-valued solution at each of the quadrature points. The > fact that *some* of the components of the vector-valued solution are > zero (because you are using an FE_Nothing for this component) doesn't > mean that you don't have to provide a std::vector<Vector<double>> as > argument. > > Does that make sense? > > Best > W. > > -- > ------------------------------------------------------------------------ > Wolfgang Bangerth email: bang...@colostate.edu > <javascript:> > www: http://www.math.colostate.edu/~bangerth/ > -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
