Ruhollah,
> Kelley paper is too old :o, I'm just curious that why there is not (at
> least) implementatin of other methods in deal ii library which is basically
> developed for AFEM (and its developpers are AFEM mans)
Oh, there is of course: step-14 is the dual weighted residual method that we
discuss in the book by Rannacher and myself.
And other people have implemented other methods as well. They just aren't part
of the library because nobody has contributed them.
> A very common estimator (ZZ, in particular in the field of linear
> elacticity) is based on patch-wise flux recovery (C0 or H1 recovery of
> grad) and computing difference between element-wise energy-norm of
> original solution and recovered one at superconvergence points; barlow
> points; what is your idea about this estimator; weakness or benefits; if
> you have some idea?
I'm sure it works as well or as bad as the Kelly estimator.
W.
-------------------------------------------------------------------------
Wolfgang Bangerth email: [EMAIL PROTECTED]
www: http://www.math.tamu.edu/~bangerth/
_______________________________________________
