Error analysis for discontinuous Galerkin time-stepping methods for nonlinear parabolic equations via maximal regularity

Georgios Akrivis,Stig Larsson

2024-12-11

Abstract:We consider the discretization of a class of nonlinear parabolic equations by discontinuous Galerkin time-stepping methods and establish a priori as well as conditional a posteriori error estimates. Our approach is motivated by the error analysis in [9] for Runge-Kutta methods for nonlinear parabolic equations; in analogy to [9], the proofs are based on maximal regularity properties of discontinuous Galerkin methods for non-autonomous linear parabolic equations.

Numerical Analysis

What problem does this paper attempt to address?

This paper attempts to solve the error analysis problem of nonlinear parabolic equations when they are discretized using the discontinuous Galerkin (dG) time - stepping method. Specifically, the author focuses on the a priori and a posteriori error estimates of a class of nonlinear parabolic equations after discretization by the dG time - stepping method.

### Problem Description
The paper considers the initial - boundary value problem of the nonlinear parabolic equation in the following form:
\[
\begin{cases}
u_t=\nabla\cdot f(\nabla u, u)&\text{in }\Omega\times(0, T],\\
u = 0&\text{on }\partial\Omega\times(0, T],\\
u(\cdot, 0)=u_0&\text{in }\Omega,
\end{cases}
\]
where $\Omega\subset\mathbb{R}^d$ is a bounded region with a smooth boundary, the initial value is $u_0$, and it is assumed that the flux function $f:\mathbb{R}^d\times\mathbb{R}\to\mathbb{R}^d$ is smooth and satisfies the local elliptic condition.

### Research Objectives
1. **A priori error estimate**: Establish the a priori error estimate of the discontinuous Galerkin time - stepping method.
2. **A posteriori error estimate**: Establish the conditional a posteriori error estimate.

### Main Results
- **Theorem 1.1 (A priori error estimate)**: Under appropriate regularity conditions, it is proved that the solution $U$ of the dG method converges to the exact solution $u$, and the order of convergence is $k^q$.
- **Theorem 1.2 (Conditional a posteriori error estimate)**: Under certain conditions, it is proved that the error of the reconstructed solution $\hat{U}$ can be estimated by the residual $R(t)$, and the error estimate is of the optimal order.

### Methods and Techniques
- **Maximum regularity**: Utilize the maximum regularity property of non - autonomous linear parabolic equations, which is the key to error analysis.
- **Radau IIA method**: Interpret the dG method as a modified Radau IIA method, thereby using its maximum regularity property.
- **Sobolev embedding theorem**: Used to handle the embedding relationships in space - time Sobolev spaces.
- **Schaefer fixed - point theorem**: Used to prove the existence of solutions.

Through these methods and techniques, the author successfully established the error estimation theoretical framework of the dG time - stepping method in solving nonlinear parabolic equations.