Discretization of the Poisson equation with non-smooth data and emphasis on non-convex domains (Q2821188)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 6628234
Language Label Description Also known as
default for all languages
No label defined
    English
    Discretization of the Poisson equation with non-smooth data and emphasis on non-convex domains
    scientific article; zbMATH DE number 6628234

      Statements

      Discretization of the Poisson equation with non-smooth data and emphasis on non-convex domains (English)
      0 references
      0 references
      0 references
      0 references
      16 September 2016
      0 references
      Poisson equation
      0 references
      non-smooth data
      0 references
      polygonal domain
      0 references
      method of transposition
      0 references
      very weak formulation
      0 references
      finite element method
      0 references
      regularization
      0 references
      convergence
      0 references
      error estimate
      0 references
      numerical example
      0 references
      The paper is concerned with the Poisson equation \(- \Delta y = f\) on a bounded polygonal domain \(\Omega \subset \mathbb{R}^2\) with boundary conditions \(y=u\) on \(\partial \Omega\) if \(u \in L^2 \left( \partial \Omega \right)\) is not smooth enough. In this case it cannot be expected that there exists a solution \(y \in H^1 \left( \Omega \right)\). The authors discuss several approaches for solving the problem. First, they study the method of transposition that leads to a very weak formulation and discuss a choice of trial and test spaces. Furthermore, Berggren's approach is studied and it is shown that it is equivalent to the method of transposition for a particular choice of test functions. Finally, the data are regularized and the solution is computed as the standard solution of the regularized problem by the finite element method. The advantage of the regularization approach is its simplicity. Convergence of the scheme using piecewise linear functions is proved and error estimates are derived. Numerical examples show that the experimental orders of convergence correspond to the theoretical convergence rates for domains with various maximal interior angles.
      0 references
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references