Abstract:We present a new fixed mesh algorithm for solving a class of interface inverse problems for the typical elliptic interface problems. These interface inverse problems are formulated as shape optimization prob- lems whose objective functionals depend on the shape of the interface. Regardless of the location of the interface, both the governing partial differential equations and the objective functional are discretized optimally, with respect to the involved polynomial space, by an immersed finite element (IFE) method on a fixed mesh. Furthermore, the formula for the gradient of the descritized objective function is de- rived within the IFE framework that can be computed accurately and efficiently through the discretized adjoint procedure. Features of this proposed IFE method based on a fixed mesh are demonstrated by its applications to three representative interface inverse problems: the interface inverse problem with an internal measurement on a sub-domain, a Dirichlet-Neumann type inverse problem whose data is given on the boundary, and a heat dissipation design problem.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is interface - related inverse problems, especially for typical elliptic interface problems. Specifically, such inverse problems are formulated as shape optimization problems, and their objective functions depend on the shape of the interface. Regardless of the position of the interface, the governing partial differential equations and the objective functions are optimally discretized on a fixed grid by the Immersed Finite Element (IFE) method. In addition, the authors derive formulas for the gradients of the discretized objective functions, and these gradients can be calculated efficiently and accurately within the IFE framework.
### Specific description of the problem
1. **Background**:
- The paper focuses on solving a class of interface inverse problems, which usually appear in engineering and scientific applications, such as Electrical Impedance Tomography (EIT) and groundwater or oil reservoir simulations.
- Such inverse problems are usually ill - posed, especially when the available data is limited.
2. **Problem types**:
- **Parameter estimation problem**: Determine the physical properties of materials, that is, the value of the coefficient \(\beta\).
- **Inverse geometry problem**: Detect the position and shape of inclusions / interfaces using data \(u_k\) measured on a subset of the domain or a subset of the boundary \(\partial\Omega\).
3. **Specific mathematical model**:
- Consider a domain \(\Omega\) divided by an interface \(\Gamma\) into two sub - domains \(\Omega^-\) and \(\Omega^+\), each occupied by different materials, represented as a piecewise - constant function \(\beta\), which is discontinuous on \(\Gamma\).
- For a typical second - order elliptic equation:
\[
-\nabla\cdot(\beta\nabla u_k)=f_k, \quad \text{in }\Omega^-\cup\Omega^+
\]
where \(u_k\) satisfies the following boundary conditions and jump conditions:
\[
u_k = g_k^D, \quad \text{on }\partial\Omega_k^D\subseteq\partial\Omega
\]
\[
\frac{\partial u_k}{\partial n}=g_k^N, \quad \text{on }\partial\Omega_k^N\subseteq\partial\Omega
\]
\[
[u_k]_{\Gamma}:=u_k^+ - u_k^- = 0
\]
\[
[\beta\nabla u_k\cdot n]_{\Gamma}:=(\beta^+\nabla u_k^+-\beta^-\nabla u_k^-)\cdot n = 0
\]
4. **Objective**:
- Use the given measurement data \(u\) to recover the position and geometric shape of the material interface \(\Gamma\).
### Solution
The paper proposes an Immersed Finite Element (IFE) method based on a fixed grid to solve the above - mentioned inverse geometry problems. The advantages of this method are as follows:
- **Fixed grid**: Avoids the problem in traditional finite element methods of constantly updating the grid to adapt to interface changes.
- **Efficient computation**: The gradients of the discretized objective functions can be calculated efficiently and accurately through the discrete adjoint process.
- **Wide applicability**: This method is applicable to various types of interface inverse problems, such as internal measurement, Dirichlet - Neumann type inverse problems, and heat dissipation design problems.
In short, the paper aims to develop a numerical method that is efficient, stable, and applicable to multiple application scenarios.