Page 98 -
P. 98

82     2. Boundary Integral Equations

                              As for the exterior boundary value problems, in order to ensure the
                                                               c
                           uniqueness of the solution of (2.4.1) in Ω , we need to augment (2.4.1) with
                           an appropriate radiation condition (see (2.3.18)). We require that
                                              A 1 · x

                                u(x)= A 0 r +        r log r + O(r)  as  r = |x|→∞     (2.4.13)
                                               |x|
                           for given constant A 0 and constant vector A 1 . Under the condition (2.4.9),
                                                                                          c
                           we then have the representation formula for the solution of (2.4.1) in Ω ,
                                                                     ∂u

                                          u(x)= −V (Mu, Nu)+ W u,        + p(x),       (2.4.14)
                                                                     ∂n
                           where p ∈R is a polynomial of degree less than or equal to one.
                              Before we formulate the boundary integral equations we first summarize
                           some classical basic results.
                           Theorem 2.4.1. (Gakhov [90], Mikhlin [208, 209, 211] and Muskhelishvili
                           [223]). Let Γ ∈ C 2,α  , 0 <α< 1.
                           i) Let
                                                          ∂u        3+α       2+α


                                     ϕ =(ϕ 1 ,ϕ 2 ) = u| Γ ,  | Γ  ∈ C  (Γ) × Γ  (Γ)
                                                          ∂n
                                                                                    4
                           be given. Then there exists a unique solution u ∈ C 3+α (Ω) ∩ C (Ω) of the
                           interior Dirichlet problem satisfying the Dirichlet conditions
                                                               ∂u
                                                u| Γ = ϕ 1 and   | Γ = ϕ 2 .           (2.4.15)
                                                               ∂n
                                                             2
                              For given A 0 ∈ IR and A 1 ∈ IR there also exists a unique solution
                                               4
                                                   c
                                      c
                           u ∈ C 3+α (Ω ∪ Γ) ∩ C (Ω ) of the exterior Dirichlet problem which behaves
                           at infinity as in (2.4.13).
                           ii) For given
                                                                        α

                                             ψ =(ψ 1 ,ψ 2 ) ∈ C 1+α (Γ) × C (Γ)
                           satisfying the compatibility conditions (2.4.12), i.e.,
                                                 ∂v


                                              ψ 1   + ψ 2 v ds =0 for all v ∈R ,       (2.4.16)
                                                 ∂n
                                           Γ
                           the interior Neumann problem consisting of (2.4.1) and the Neumann condi-
                           tions
                                                                                       (2.4.17)
                                                Mu| Γ = ψ 1 and Nu| Γ = ψ 2
                                                       4
                           has a solution u ∈ C 3+α (Ω) ∩ C (Ω) which is unique up to a linear function
                           p ∈R.
                                                                    c
                              If, for the exterior Neumann problem in Ω , in addition to ψ the linear
                                                                                       c
                           function p ∈R is given, then it has a unique solution u ∈ C 3+α (Ω ∪ Γ) ∩
                             4
                           C (Ω) with the behaviour (2.4.13), (2.4.14) where
   93   94   95   96   97   98   99   100   101   102   103