Page 133 -
P. 133

116    3. Finite Element Methods for Linear Elliptic Problems


          As in the example of (2.27) (cf. Lemma 2.10), V h is defined by specifying
        P K and the degrees of freedom on K for K ∈T h . These have to be chosen
        such that, on the one hand, they enforce the continuity of v ∈ V h and,
        on the other hand, the satisfaction of the homogeneous Dirichlet bound-
        ary conditions at the nodes. For this purpose, compatibility between the
        Dirichlet boundary condition and the triangulation is necessary, since it
        will be required in (T6).
          As can be seen from the proof of Lemma 2.10, it is essential
                that the interpolation problem, locally defined on K ∈
        (F1)                                                        (3.41)
                T h by the degrees of freedom, is uniquely solvable in P K ,
                that this also holds on the (d−1)-dimensional boundary
                surfaces F of K ∈T h for the degrees of freedom from F
                and the functions v| F where v ∈ P K ; this then ensures
        (F2)                                                        (3.42)
                the continuity of v ∈ V h ,if P K and P K   matchinthe
                sense of P K | F = P K  | F for K, K ∈T h intersecting in F

                (see Figure 3.2).
                  .                              .
                             P  = P                     P     = P
                              K    1                    K  F  1
                                  .
                          .                              .

        Figure 3.2. Compatibility of the ansatz space on the boundary surface and the
        degrees of freedom there.

          The following finite elements defined by their basic domain K(∈T h ),
        the local ansatz space P K , and the degrees of freedom Σ K satisfy these
        properties.
          For this, let P k (K) be the set of mappings p : K → R of the following
        form:
                                                              α
                                          x
            p(x)= p(x 1 ,...,x d )=  γ α 1 ...α d 1 α 1  ··· x α d  =  γ α x ,  (3.43)
                                                  d
                                |α|≤k                 |α|≤k
        hence the polynomials of order k in d variables. The set P k (K)forms
        a vector space, and since p ∈P k (K) is differentiable arbitrarily often,
        P k (K) is a subset of all function spaces introduced so far (provided that
        the boundary conditions do not belong to their definition).
                                    d
          For both, K ∈T h and K = R we have

                                                  d + k
                                            d
                        dim P k (K)= dim P k (R )=      ,           (3.44)
                                                    k
                   d
        as even P k (R )| K = P k (K) (see Exercise 3.12). Therefore, for short we will
        use the notation P 1 = P 1 (K) if the dimension of the basic space is fixed.
   128   129   130   131   132   133   134   135   136   137   138