Abstract:The focus of this paper is on the concurrent reconstruction of both the diffusion and potential coefficients present in an elliptic/parabolic equation, utilizing two internal measurements of the solutions. A decoupled algorithm is constructed to sequentially recover these two parameters. In the first step, we implement a straightforward reformulation that results in a standard problem of identifying the diffusion coefficient. This coefficient is then numerically recovered, with no requirement for knowledge of the potential, by utilizing an output least-square method coupled with finite element discretization. In the second step, the previously recovered diffusion coefficient is employed to reconstruct the potential coefficient, applying a method similar to the first step. Our approach is stimulated by a constructive conditional stability, and we provide rigorous a priori error estimates in $L^2(\Omega)$ for the recovered diffusion and potential coefficients. Our approach is stimulated by a constructive conditional stability, and we provide rigorous a priori error estimates in $L^2(\Omega)$ for the recovered diffusion and potential coefficients. To derive these estimates, we develop a weighted energy argument and suitable positivity conditions. These estimates offer a beneficial guide for choosing regularization parameters and discretization mesh sizes, in accordance with the noise level. Some numerical experiments are presented to demonstrate the accuracy of the numerical scheme and support our theoretical results.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to simultaneously reconstruct the diffusion coefficient \(D(x)\) and the potential coefficient \(\sigma(x)\) in elliptic/parabolic equations, and two internal measurements are used to achieve this goal. Specifically, the researchers focus on how to recover the exact diffusion coefficient \(D^\dagger(x)\) and the potential coefficient \(\sigma^\dagger(x)\) from two internal observations.
### Problem Background
In many practical applications, such as groundwater flow simulation, medical imaging and other fields, it is necessary to determine the diffusion coefficient and the potential coefficient in the medium. These coefficients are usually unknown, but can be estimated by inverting the internal or boundary observation data of the system. However, this inverse problem is usually ill - posed, that is, small measurement errors may lead to large changes in the solution. Therefore, developing effective numerical algorithms to stably recover these coefficients is an important research topic.
### Main Contributions of the Paper
1. **Decoupling Algorithm**: The paper proposes a decoupling algorithm that can sequentially recover the diffusion coefficient and the potential coefficient. First, through a simple rewrite, the problem is transformed into a standard problem of identifying the diffusion coefficient. Then, the output least - squares method combined with finite - element discretization is used to numerically recover the diffusion coefficient. In the second step, the recovered diffusion coefficient is used to reconstruct the potential coefficient, and the method is similar to the first step.
2. **Conditional Stability**: Based on the constructive conditional stability, the researchers provide strict a priori error estimates. These estimates not only help to select the regularization parameters and the discretization grid size, but also provide theoretical support for the accuracy of the numerical scheme.
3. **Numerical Experiments**: The paper presents several numerical experiments to verify the accuracy of the proposed numerical scheme and the validity of the theoretical results.
### Mathematical Model
Consider the following elliptic boundary - value problem:
\[
\begin{cases}
-\nabla\cdot(D\nabla u)+\sigma u = f, & \text{in }\Omega,\\
u = g, & \text{on }\partial\Omega,
\end{cases}
\]
where \(f\) is a given source term, \(g\) is the boundary data, \(\Omega\subset\mathbb{R}^d\) is a convex polyhedral domain, and \(\partial\Omega\) is its boundary. The goal of the problem is to recover the exact coefficients \(D^\dagger(x)\) and \(\sigma^\dagger(x)\) from two internal observations \(u_1(D^\dagger,\sigma^\dagger)\) and \(u_2(D^\dagger,\sigma^\dagger)\).
### Specific Steps of the Decoupling Algorithm
1. **First Step: Recover the Diffusion Coefficient**
- By rewriting the equation, the potential coefficient \(\sigma\) is eliminated, and a new equation containing only the diffusion coefficient \(D\) is obtained.
- The output least - squares method and finite - element discretization are used to numerically recover the diffusion coefficient \(D\).
2. **Second Step: Recover the Potential Coefficient**
- Using the diffusion coefficient \(D\) recovered in the first step, a new equation is constructed to recover the potential coefficient \(\sigma\).
- A similar method is used to numerically recover the potential coefficient \(\sigma\).
### Error Estimation
The paper provides strict a priori error estimates, which are derived in the \(L^2(\Omega)\) norm. These estimates consider not only the discretization error of the numerical solution, but also the influence of measurement noise. Through these estimates, the selection of regularization parameters and discretization grid size can be guided, thereby improving the accuracy of the numerical solution.
### Conclusion
This paper proposes an effective and stable numerical algorithm for simultaneously recovering the diffusion coefficient and the potential coefficient in elliptic/parabolic equations. Through strict mathematical analysis and numerical experiments, the effectiveness and accuracy of this method are proved. These results are of great significance for parameter identification problems in practical applications.