Non-homogeneous Dirichlet Boundary condition for Stokes problem

[Hemanta822]

Hi All,

I am new to MFEM and trying to impose non-homogeneous Dirichlet boundary conditions for the Stokes problem. I am modifying ex5 (Darcy code). I managed to get optimal rates while using Taylor-Hood elements by the following way:

a) I have boundary condition u=uexact, where u is the Stokes velocity. I defined additional variable : w= u - uexact.
b) Changed the Stokes equation in terms of w. Then my boundary condition becomes w=0.
c) Solve for w.
d) Get u back using u=w+uexact.

I also asked for help in the following issue.
#80 (comment)

I don't want to do this all the time. Is there any simple way to impose non-homogeneous Dirichlet boundary, without defining additional variable. Any help or Direction would be greatly appreciated. Thank you.