Goal-Oriented Adaptive Space-Time Finite Element Methods for Regularized Parabolic p-Laplace Problems

B.Endtmayer,U. Langer,A. Schafelner

2023-06-12

Abstract:We consider goal-oriented adaptive space-time finite-element discretizations of the regularized parabolic p-Laplace problem on completely unstructured simplicial space-time meshes. The adaptivity is driven by the dual-weighted residual (DWR) method since we are interested in an accurate computation of some possibly nonlinear functionals at the solution. Such functionals represent goals in which engineers are often more interested than the solution itself. The DWR method requires the numerical solution of a linear adjoint problem that provides the sensitivities for the mesh refinement. This can be done by means of the same full space-time finite element discretization as used for the primal non-linear problems. The numerical experiments presented demonstrate that this goal-oriented, full space-time finite element solver efficiently provides accurate numerical results for different functionals.

Numerical Analysis

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: How to perform goal - oriented adaptive spatio - temporal finite - element discretization for the regularized parabolic \(p\)-Laplace problem on a completely unstructured spatio - temporal grid. Specifically, the paper focuses on efficiently solving the regularized variational \(p\)-Laplace problem and accurately calculating certain possibly nonlinear target functionals without using additional stabilization techniques through the goal - oriented adaptive method (GOAST).

### Background and Problem Description
- **Research Object**: The regularized parabolic \(p\)-Laplace equation, which has the form:
\[
\partial_t u-\text{div}_x\left((|\nabla_x u|^2 + \varepsilon^2)^{\frac{p - 2}{2}}\nabla_x u\right)=f\quad\text{in }Q,
\]
where \(Q=\Omega\times(0, T)\), \(\Omega\) is the spatial domain, \(T\) is the final time, \(\varepsilon> 0\) is the regularization parameter, and \(p\in(1,\infty)\) represents the degree of nonlinearity.

- **Goal**: To accurately calculate certain specific target functionals, which usually represent quantities that engineers are more interested in rather than the solution itself.

### Main Methods
- **Spatio - Temporal Finite - Element Method**: Use the spatio - temporal finite - element method to discretize the above equation on a completely unstructured spatio - temporal grid.
- **Goal - Oriented Adaptive Method**: Use the dual - weighted residual method (DWR) to drive the adaptive process in order to improve the calculation accuracy of the target functionals.
- **Adjoint Problem**: In order to achieve goal - oriented adaptation, it is necessary to solve a linear adjoint problem, which provides sensitivity information for grid refinement.

### Key Contributions
- **Efficient Adaptive Algorithm**: Propose a goal - oriented adaptive spatio - temporal finite - element method that can efficiently provide accurate numerical results under different target functionals.
- **Theoretical Analysis**: Prove the effectiveness of this method through theoretical analysis of error estimates and adaptive strategies.
- **Numerical Experiments**: Verify the effectiveness of the method through numerical experiments and demonstrate its performance under different target functionals.

### Conclusion
The paper successfully solves the regularized parabolic \(p\)-Laplace problem through the goal - oriented adaptive spatio - temporal finite - element method and demonstrates efficient computational performance under different target functionals. This method not only improves the calculation accuracy of the target functionals but also provides a powerful tool for engineering applications.