Abstract:Building on and extending tools from variational analysis, we prove Kuratowski convergence of sets of simplicial area minimizers to minimizers of the smooth Douglas-Plateau problem under simplicial refinement. This convergence is with respect to a topology that is stronger than uniform convergence of both positions and surface normals.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is whether the discrete minimal surfaces converge to the solutions of the smooth Douglas - Plateau problem during the simplification process (i.e., by increasing the mesh resolution). Specifically, the author wants to prove that, under the condition of invariant topological structure, as the mesh is refined, the set of discrete minimal surfaces will converge to the set of smooth minimal surfaces, and this convergence is carried out in a topology stronger than the uniform convergence of position and normal.
### Core Problems of the Paper
1. **Convergence of Discrete Minimal Surfaces**: Prove whether the discrete minimal surfaces will converge to the smooth minimal surfaces during the mesh refinement process.
2. **Topological Structure of Convergence**: Ensure that this convergence is carried out in a topology stronger than the uniform convergence of position and normal.
3. **Handling Non - convex Optimization Problems**: The minimal area problem is usually non - convex, so it is necessary to find appropriate methods to handle this problem.
### Background and Motivation
The minimal surface problem is a classic mathematical problem, which involves finding the surface with the minimum area under given boundary conditions. Although solutions already exist for some specific cases (such as disc - like surfaces), there are still many challenges for more general cases, especially for the minimal volume problems in high - dimensional manifolds. In particular, when calculating these minimal surfaces, how to use finite - dimensional approximation methods (such as triangular meshes) to approximate the smooth solutions is an important research direction.
### Method Overview
The author adopts variational analysis tools, especially the concept of Kuratowski convergence, to prove the convergence of discrete minimal surfaces. Specific steps include:
- **Define Sampling and Reconstruction Operators**: Map the smooth manifolds to discrete manifolds and vice versa.
- **Consistency, Stability and Proximity**: By introducing consistency error, stability conditions and proximity error, ensure that the discrete solutions can approximate the smooth solutions.
- **Select Appropriate Metric Spaces**: Select appropriate metric spaces to compare the discrete and smooth solutions to ensure that the convergence is meaningful.
### Main Results
The main result of the paper is to prove that the set of discrete minimal surfaces converges to the set of smooth minimal surfaces in the sense of Kuratowski. This result not only verifies the effectiveness of the discrete method but also provides theoretical support for numerical calculations.
### Formula Summary
- **Kuratowski Convergence**: Let \( (A_n)_{n \in \mathbb{N}} \) be a sequence of sets in the topological space \( X \), and \( A \) is also a set in \( X \). If all the accumulation points of \( (A_n)_{n \in \infty} \) belong to \( A \), and each \( a \in A \) is a limit point of a certain sequence \( (a_n)_{n \in \mathbb{N}} \) (where \( a_n \in A_n \)), then \( A_n \) is said to converge to \( A \) in the sense of Kuratowski.
- **Total Consistency Error**: Defined as:
\[
\delta_n=\delta_S^n + \delta_R^n
\]
where,
\[
\delta_S^n=\sup_{a \in A}(F_n\circ S_n(a)-F(a))^+
\]
\[
\delta_R^n=\sup_{a \in A_n}(F\circ R_n(a)-F_n(a))^+
\]
- **Proximity Error**: Defined as:
\[
\varepsilon_n = \max(\varepsilon_S^n,\varepsilon_R^n)
\]
where,
\[
\varepsilon_S^n=\sup_{a \in A}d_X(\Psi_n\circ S_n(a),\Psi(a))
\]
\[
\varepsilon_R^n=\sup_{a \in A_n}d_X(\Psi\circ R_n(a),\Psi_n(a))
\]