Positivity-Preserving Entropy-Based Adaptive Filtering for Discontinuous Spectral Element Methods

Tarik Dzanic,Freddie D. Witherden

DOI: https://doi.org/10.1016/j.jcp.2022.111501

2022-07-28

Abstract:In this work, we present a positivity-preserving entropy-based adaptive filtering method for shock capturing in discontinuous spectral element methods. By adapting the filter strength to enforce positivity and a local discrete minimum entropy principle, the resulting approach can robustly resolve strong discontinuities with sub-element resolution, does not require problem-dependent parameter tuning, and can be easily implemented on general unstructured meshes with relatively low computational cost. The efficacy of the approach is shown in numerical experiments on hyperbolic and mixed hyperbolic-parabolic conservation laws such as the Euler and Navier-Stokes equations for problems including extreme shocks, shock-vortex interactions, and complex compressible turbulent flows.

Numerical Analysis,Fluid Dynamics

What problem does this paper attempt to address?

This paper attempts to address the robustness and reliability issues of the discontinuous spectral element method (DSEM) when dealing with problems involving discontinuities. Specifically, the author proposes an entropy - based adaptive filtering method that preserves positivity to deal with the non - physical oscillations and numerical instabilities that occur when high - order discrete methods simulate strong discontinuities (such as shock waves).

### Main problems

1. **Non - physical oscillations**: Due to the Gibbs phenomenon, traditional methods will produce non - physical oscillations when dealing with discontinuities, leading to solution distortion or calculation failure.
2. **Parameter dependence**: Many existing methods require parameter adjustment according to specific problems and grids, which is not feasible in large - scale parallel computing.
3. **Guarantee of physical constraints**: Many methods cannot ensure that physical constraints (such as the positivity of density and pressure) are satisfied, which may lead to the failure of numerical schemes.

### Solutions

To solve the above problems, the author proposes an entropy - based adaptive filtering method with the following main features:

- **Preserving positivity**: By adjusting the filtering intensity, ensure that physical quantities (such as density and pressure) in the numerical solution remain positive.
- **Local discrete minimum entropy principle**: By introducing the local discrete minimum entropy principle, reduce non - physical oscillations near discontinuities.
- **No parameter adjustment required**: This method does not require parameter adjustment according to specific problems or grids, and is universal and efficient.
- **Applicable to unstructured grids**: It can be easily implemented on general unstructured grids and has a relatively low computational cost.

### Numerical experiments

The author verifies the effectiveness of this method through a series of numerical experiments, including:

- **Euler equations**: The classic Sod shock - tube problem used to evaluate shock - capturing ability.
- **Navier - Stokes equations**: Used to evaluate complex compressible turbulent flow problems.

These experiments demonstrate the superior performance of this method in dealing with extreme shock waves, shock - vortex interactions, and complex compressible turbulent flow problems.

### Formula representation

To ensure the correctness and readability of the formulas, the following are some key formulas involved in the paper represented in Markdown format:

1. **Hyperbolic conservation law**
\[
\frac{\partial u}{\partial t}+\nabla\cdot F(u) = 0,\quad u(x,0)=u_0(x),\quad x\in\Omega
\]

2. **Entropy inequality**
\[
\frac{\partial\eta(u)}{\partial t}+\nabla\cdot\psi(u)\geq0
\]
where \(\eta(u)\) is the entropy function and \(\psi(u)\) is the entropy flux.

3. **Local minimum entropy principle**
\[
\eta(u(x_0,t + \Delta t))\geq\min_{x\in D_0}\eta(u(x,t))
\]

4. **Filtering operation**
\[
H_i(\hat{u}_i)=\hat{u}_i e^{-\epsilon p_i^2}
\]
where \(p_i\) is the maximum order of the modal basis function.

5. **Filtering intensity optimization**
\[
\epsilon=\arg\min_{\epsilon\geq0}\left\{\Gamma(\tilde{u}(x_i))\geq0,\quad\eta(\tilde{u}(x_i))\geq\eta_{\text{min}}\quad\forall i\in S\right\}
\]

Through these improvements, this method can effectively capture and analyze discontinuity features while ensuring physical constraints, thereby improving the reliability and robustness of the discontinuous spectral element method in practical applications.