Recovery Techniques for Finite Element Methods

Hailong Guo,Zhimin Zhang

2024-12-05

Abstract:Post-processing techniques are essential tools for enhancing the accuracy of finite element approximations and achieving superconvergence. Among these, recovery techniques stand out as vital methods, playing significant roles in both post-processing and pre-processing. This paper provides an overview of recent developments in recovery techniques and their applications in adaptive computations. The discussion encompasses both gradient recovery and Hessian recovery methods. To establish the superconvergence properties of these techniques, two theoretical frameworks are introduced. Applications of these methods are demonstrated in constructing asymptotically exact {\it a posteriori} error estimators for second-order elliptic equations, fourth-order elliptic equations, and interface problems. Numerical experiments are performed to evaluate the asymptotic exactness of recovery type a posteriori error estimators.

Numerical Analysis

What problem does this paper attempt to address?

### Problems the paper attempts to solve

This paper aims to explore the recovery techniques in the finite element method, especially the gradient recovery and Hessian recovery methods, in order to improve the accuracy of numerical solutions and achieve superconvergence. Specifically, the paper mainly focuses on the following aspects:

1. **Improving the accuracy of finite - element approximation**: Enhance the accuracy of the finite - element approximate solution through post - processing techniques, especially when solving second - order elliptic equations, fourth - order elliptic equations and interface problems.

2. **Constructing reliable a posteriori error estimators**: Develop easy - to - use asymptotically exact a posteriori error estimators, which are crucial for adaptive computation. These estimators can evaluate the error of the numerical solution and guide the mesh refinement process.

3. **Establishing a theoretical framework**: Introduce two theoretical frameworks to prove the superconvergence properties of the recovery techniques. These two frameworks are:
- Based on supercloseness results on slightly structured meshes.
- Based on superconvergence by different quotients on translation - invariant meshes.

4. **Practical applications and numerical experiments**: Verify the asymptotic exactness of the a posteriori error estimators based on the recovery techniques through numerical experiments and show their performance in practical applications.

### Key concepts

- **Gradient Recovery**: Construct a more accurate gradient field by locally fitting the finite - element solution. Commonly used methods include the simple weighted average method, the Superconvergent Patch Recovery (SPR) and the Polynomial Preserving Recovery (PPR).

- **Hessian Recovery**: Construct the Hessian matrix by applying the gradient recovery operator again to the gradient recovery result.

- **Superconvergence**: The phenomenon that the convergence rate of the numerical solution is higher than the optimal convergence rate at some special points. For example, when using \(k\)-order Lagrange elements, the solution at the nodes can achieve \(2k\)-order superconvergence.

- **Asymptotic Exactness**: As the mesh is refined, the error of the error estimator gradually approaches the true error.

### Research background

The finite - element method is widely used in scientific computing, but the accuracy of the standard finite - element solution is limited. In order to improve the accuracy and achieve adaptive computation, researchers have proposed a variety of post - processing techniques. Among them, the recovery techniques have attracted much attention due to their excellent performance in improving accuracy and achieving superconvergence. The Superconvergent Patch Recovery (SPR) method proposed by Zienkiewicz and Zhu is a milestone in this field, and subsequent research has further improved and developed these methods, such as the Polynomial Preserving Recovery (PPR).

### Conclusion

This paper systematically reviews the Polynomial Preserving Recovery techniques and their applications in adaptive computation, and shows their advantages in improving the accuracy of numerical solutions and constructing reliable error estimators. Through theoretical analysis and numerical experiments, the effectiveness and superiority of these recovery techniques are proved.