Emmanuil H. Georgoulis,Omar Lakkis
Abstract:We derive a posteriori error bounds for a quasilinear parabolic problem, which is approximated by the $hp$-version interior penalty discontinuous Galerkin method (IPDG). The error is measured in the energy norm. The theory is developed for the semidiscrete case for simplicity, allowing to focus on the challenges of a posteriori error control of IPDG space-discretizations of strictly monotone quasilinear parabolic problems. The a posteriori bounds are derived using the elliptic reconstruction framework, utilizing available a posteriori error bounds for the corresponding steady-state elliptic problem.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to provide **a posteriori error estimates** for the discontinuous Galerkin (DG) method for a class of quasilinear parabolic problems. Specifically, the author focuses on using the interior penalty discontinuous Galerkin method (IPDG) to spatially discretize strictly monotonic quasilinear parabolic problems and derives a posteriori error bounds in the energy norm.
### Main problems and methods
1. **Problem description**:
- The research object is a quasilinear parabolic equation with strictly monotonic nonlinearity and Lipschitz growth conditions.
- The form of the equation is:
\[
u_t(t,x) - \nabla \cdot (a(t,x,|\nabla u(t,x)|) \nabla u(t,x)) = f(t,x) \quad \text{in } (0,T] \times \Omega,
\]
where \( f \in L^\infty(0,T;L^2(\Omega)) \), and the initial condition and homogeneous Dirichlet boundary conditions are given.
2. **Numerical method**:
- The hp - version of the interior penalty discontinuous Galerkin method (IPDG) is used, which is a high - order finite element method suitable for handling discontinuous solutions.
- By introducing the elliptic reconstruction technique, the error is decomposed into a parabolic part and an elliptic part, so as to utilize the existing a posteriori error estimate results for steady - state elliptic problems.
3. **Main contributions**:
- A posteriori error bounds in the energy norm are derived, and these bounds explicitly depend on the polynomial degree \( p \) of local elements.
- Explicit error estimate formulas are provided, which enable the evaluation of the quality of the numerical solution and guide the adaptive mesh refinement strategy.
### Elliptic reconstruction technique
The elliptic reconstruction technique is a key tool in this paper. It simplifies the error analysis by correlating the semi - discrete solution \( U \) of the original problem with the solution \( w \) of an elliptic problem. The specific steps are as follows:
- Define the elliptic reconstruction \( w \equiv w(t) \in H^1_0(\Omega) \) as the solution that satisfies the following elliptic problem:
\[
\langle \alpha(w), \nabla v \rangle = \langle g, v \rangle \quad \forall v \in H^1_0(\Omega),
\]
where \( g \equiv g(t)= -AU + f - \Pi f \), \( \Pi \) is the orthogonal \( L^2 \)-projection operator onto \( S_p \), and \( A \) is a discrete operator.
- Decompose the error \( e = u - U \) into two parts: \( e=\rho - \epsilon \), where \( \rho = w - u \) and \( \epsilon = w - U \).
In this way, the author can derive an explicit error estimate for \( \epsilon \), and combined with the a posteriori error estimate of the elliptic problem, finally obtain the estimate of the entire error \( e \).
### Summary
The main objective of this paper is to provide reliable a posteriori error estimates for the IPDG method for quasilinear parabolic problems, which is of great significance for verifying the accuracy of numerical solutions and guiding the design of adaptive algorithms.