<p>Various complex interface shapes are often encountered when solving interface problems numerically, which may causes certain difficulties. This paper is dedicated to the development of a Special Immersed Finite Volume (SIFV) method to deal with elliptic interface problems, even in cases where the interface has a non-trivial geometry. This SIFV method works on unfitted meshes without requiring stabilization terms or penalty parameters. By employing the Aubin-Nitsche technique, an optimal <Emphasis Type="BoldItalic">L</Emphasis><sup><b>2</b></sup> convergence is established for the elliptic interface problem. Numerical tests demonstrate that the SIFV method possesses the following numerical performance: (a) it is capable of achieving second order accuracy in the <Emphasis Type="BoldItalic">L</Emphasis><sup><b>2</b></sup> norm; (b) it is applicable to elliptic interface problems with complex interface geometries; (c) it maintains its convergence order with the mesh refinement.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

\(L^2\) error estimate for solving elliptic interface problems by using a special immersed finite volume method

  • Kai Liu,
  • Pengfei Zhu

摘要

Various complex interface shapes are often encountered when solving interface problems numerically, which may causes certain difficulties. This paper is dedicated to the development of a Special Immersed Finite Volume (SIFV) method to deal with elliptic interface problems, even in cases where the interface has a non-trivial geometry. This SIFV method works on unfitted meshes without requiring stabilization terms or penalty parameters. By employing the Aubin-Nitsche technique, an optimal L2 convergence is established for the elliptic interface problem. Numerical tests demonstrate that the SIFV method possesses the following numerical performance: (a) it is capable of achieving second order accuracy in the L2 norm; (b) it is applicable to elliptic interface problems with complex interface geometries; (c) it maintains its convergence order with the mesh refinement.