High Resolution Optimized High-Order Schemes for Discretization of Non-Linear Straight and Mixed Second Derivative Terms

Hemanth Chandravamsi,Steven H. Frankel

2024-06-02

Abstract:In this paper, we propose a new set of midpoint-based high-order discretization schemes for computing straight and mixed nonlinear second derivative terms that appear in the compressible Navier-Stokes equations. Firstly, we detail a set of conventional fourth and sixth-order baseline schemes that utilize central midpoint derivatives for the calculation of second derivatives terms. To enhance the spectral properties of the baseline schemes, an optimization procedure is proposed that adjusts the order and truncation error of the midpoint derivative approximation while still constraining the same overall stencil width and scheme order. A new filter penalty term is introduced into the midpoint derivative calculation to help achieve high wavenumber accuracy and high-frequency damping in the mixed derivative discretization. Fourier analysis performed on the both straight and mixed second derivative terms show high spectral efficiency and minimal numerical viscosity with no odd-even decoupling effect. Numerical validation of the resulting optimized schemes is performed through various benchmark test cases assessing their theoretical order of accuracy and solution resolution. The results highlight that the present optimized schemes efficiently utilize the inherent viscosity of the governing equations to achieve improved simulation stability - a feature attributed to their superior spectral resolution in the high wavenumber range. The method is also tested and applied to non-uniform structured meshes in curvilinear coordinates, employing a supersonic impinging jet test case.

Numerical Analysis,Computational Physics,Fluid Dynamics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to develop a high - resolution numerical discretization scheme for calculating the nonlinear straight second - order derivative terms and mixed second - order derivative terms that appear in the compressible Navier - Stokes equations. Specifically, the author proposes new mid - point - based high - order discretization schemes to improve the computational accuracy and spectral characteristics of these derivative terms.

### Main problems

1. **Improving numerical accuracy**: Although traditional fourth - order and sixth - order discretization methods can provide a certain degree of accuracy, they perform poorly in dealing with complex flow problems, especially in terms of spectral characteristics in the high - wavenumber range. Therefore, higher - order and more accurate discretization methods need to be developed to improve the accuracy of numerical simulations.

2. **Optimizing spectral characteristics**: In order to ensure the efficiency of the discretization scheme at different wavenumbers and the minimum numerical viscosity, the author proposes an optimization process to adjust the order and truncation error of the mid - point derivative approximation while maintaining the same overall stencil width and scheme order.

3. **Handling mixed derivative terms**: In the compressible Navier - Stokes equations, in addition to the straight second - order derivative terms, there are also mixed second - order derivative terms (such as \(\frac{\partial}{\partial x}\left(\frac{\partial\phi}{\partial y}\right)\)). The discretization of these mixed derivative terms also requires special treatment to avoid odd - even decoupling effects and improve high - frequency attenuation performance.

4. **Applying to complex grids**: This method is not only applicable to uniform structured grids but also extended to non - uniform structured grids and curvilinear coordinate systems, and is especially verified in complex flow problems such as supersonic impinging jets.

### Overview of solutions

- **Baseline scheme**: First, the traditional fourth - order and sixth - order discretization schemes are described in detail. These schemes use the central difference method to calculate second - order derivative terms.

- **Optimization scheme**: By introducing new filtering penalty terms, the calculation of the mid - point derivative is optimized, thereby achieving high - wavenumber accuracy and high - frequency attenuation in the mixed - derivative discretization. In addition, the high - spectral efficiency and minimum numerical viscosity of the optimization scheme for straight second - order derivative terms and mixed second - order derivative terms are verified by Fourier analysis.

- **Numerical verification**: Through numerical verification of multiple benchmark test cases, the theoretical accuracy and solution resolution of the optimization scheme are evaluated. The results show that the optimized scheme can effectively utilize the inherent viscosity of the control equations to achieve better simulation stability, especially showing excellent spectral resolution in the high - wavenumber range.

### Key formulas

1. **Discretization of straight second - order derivative terms**:
\[
\left(\frac{\partial^{2}\phi}{\partial x^{2}}\right)_j\approx\frac{a^{*}}{\Delta x}\left(\phi'_{j +\frac{1}{2}}-\phi'_{j-\frac{1}{2}}\right)+\frac{b^{*}}{3\Delta x}\left(\phi'_{j+\frac{3}{2}}-\phi'_{j-\frac{3}{2}}\right)+\frac{c^{*}}{5\Delta x}\left(\phi'_{j+\frac{5}{2}}-\phi'_{j-\frac{5}{2}}\right)
\]
where the coefficients \([a^{*}, b^{*}, c^{*}]\) take values according to the required order of accuracy.

2. **Calculation of mid - point derivative**:
\[
\left(\frac{\partial\phi}{\partial x}\right)_{j+\frac{1}{2}}\approx\sum_{p = - 3}^{3}a_p^{\pm}\frac{\phi_{j + p}}{\Delta x}\pm\psi_1^{\frac{1}{2}}\frac{\partial^{5}\phi}{\partial x^{5}}\Delta x^{5}