The use of discrete orthogonal projections in boundary element methods (Q1604027)

An overwiev of the theory of discrete orthogonal projections is presented. Discrete orthogonal projections provide a connecting framework between collocation methods on the one hand and Galerkin boundary element discretization on the other hand. The authors are mainly concentrated on qualocation methods which are situated in between. The definition of the discrete orthogonal projections and their basic properties are included in this article. Two applications of the commutator property for splines are presented. This property is a tool of central importance in stability and convergence proofs for qualocation methods for boundary integral equations with variable coefficients. Another application is the transfer of superconvergence properties from constant-coefficients boundary integral equations to the variable coefficients case. A new set of numerical experiments that confirm the theory is also included. The main conclusion is that the presence of variable coefficients of a certain smoothness does not influence the superconvergence, and the use of superconvergence-based a posteriori error estimators in this case is theoretically justified.