Abstract:We consider the Stefan problem, firstly with regular data and secondly with irregular data. In both cases is given a proof for the convergence of an approximation obtained by regularising the problem. These proofs are based on weak formulations and on compactness results in some Sobolev spaces with negative exponents.
What problem does this paper attempt to address?
This paper attempts to solve the numerical approximation of the Stefan problem and prove its convergence. Specifically, the article mainly focuses on two aspects of the problem:
1. **The Stefan problem with regular data**:
- The author considered the Stefan problem with regular data (i.e., smooth data) and provided a new method to prove the convergence of the solutions to the regularized problem.
- The new proof method is based on the weak form and the compactness results in Sobolev spaces with negative exponents, especially the compactness in the \( L^2(]0,T[, H^{-1}(\Omega)) \) space.
2. **The Stefan problem with non - regular data**:
- For non - regular data (such as \( f \in L^1(]0,T[, L^1(\Omega)) \) and \( u_0 \in L^1(\Omega) \)), the author also provided the corresponding convergence proof.
- In this case, time compactness cannot be dealt with by classical methods, so the author introduced the compactness results in Sobolev spaces with negative exponents to replace the time regularization method.
### Specific problem description
The Stefan problem is a free - boundary problem that describes the evolution of the two - phase boundary during the phase - change process of a material. Its model problem can be expressed as:
\[
\begin{cases}
\partial_t u-\Delta (\phi(u)) = f & \text{in } \Omega\times ]0, T[, \\
u(x, 0) = u_0(x) & \text{in } \Omega, \\
u(x, t) = 0 & \text{on } \partial \Omega\times ]0, T[.
\end{cases}
\]
where:
- \( \Omega \) is a bounded open set in \( \mathbb{R}^d \);
- \( \phi \) is a Lipschitz continuous and non - decreasing function;
- \( f \) is the source term;
- \( u_0 \) is the initial condition.
### Main contributions
1. **New convergence proof for regular data**:
- A new proof method based on the compactness of Sobolev spaces with negative exponents was proposed, avoiding some complex steps in traditional methods.
2. **Convergence analysis for non - regular data**:
- For non - regular data, a convergence proof method without additional time regularization was proposed, using the compactness results in Sobolev spaces with negative exponents.
### Key techniques
- **Weak form**: Transforming the problem into a weak form makes it easier to use functional analysis tools.
- **Sobolev spaces with negative exponents**: Using the compactness results in the \( L^2(]0,T[, H^{-1}(\Omega)) \) space to prove convergence.
- **Minty trick**: Used to handle the convergence of nonlinear terms.
Through these methods, the author successfully proved the convergence of the solutions to the regularized problem and extended it to the case of non - regular data.