Implicit high-order gas-kinetic schemes for compressible flows on three-dimensional unstructured meshes I: steady flows

Yaqing Yang,Liang Pan,Kun Xu

2024-03-04

Abstract:In the previous studies, the high-order gas-kinetic schemes (HGKS) have achieved successes for unsteady flows on three-dimensional unstructured meshes. In this paper, to accelerate the rate of convergence for steady flows, the implicit non-compact and compact HGKSs are developed. For non-compact scheme, the simple weighted essentially non-oscillatory (WENO) reconstruction is used to achieve the spatial accuracy, where the stencils for reconstruction contain two levels of neighboring cells. Incorporate with the nonlinear generalized minimal residual (GMRES) method, the implicit non-compact HGKS is developed. In order to improve the resolution and parallelism of non-compact HGKS, the implicit compact HGKS is developed with Hermite WENO (HWENO) reconstruction, in which the reconstruction stencils only contain one level of neighboring cells. The cell averaged conservative variable is also updated with GMRES method. Simultaneously, a simple strategy is used to update the cell averaged gradient by the time evolution of spatial-temporal coupled gas distribution function. To accelerate the computation, the implicit non-compact and compact HGKSs are implemented with the graphics processing unit (GPU) using compute unified device architecture (CUDA). A variety of numerical examples, from the subsonic to supersonic flows, are presented to validate the accuracy, robustness and efficiency of both inviscid and viscous flows.

Numerical Analysis

What problem does this paper attempt to address?

### What problem does this paper attempt to solve?

This paper aims to solve the problem of slow convergence rate when simulating steady - state compressible flows on three - dimensional unstructured grids. Specifically, the authors developed the Implicit High - Order Gas - Kinetic Scheme (HGKS) to accelerate the convergence rate of steady - state flows.

#### Background and problem description

1. **Compressible fluid simulation of complex geometries**
- In computational fluid dynamics, simulating compressible flows with complex geometries is an important problem.
- Unstructured grids are widely used because of their flexibility, but the convergence speed of traditional explicit time - discretization methods drops significantly when dealing with steady - state flows.

2. **Limitations of existing methods**
- Explicit methods have a slow convergence rate in steady - state flow problems and require a large number of time steps to reach the steady state.
- Although the LUSGS method is more efficient than explicit methods, it still requires thousands of time steps to reach the steady state and has the problem of large memory consumption.

3. **Advantages of implicit methods**
- Implicit methods can accelerate convergence by solving large sparse linear equation systems, especially by using Krylov subspace methods such as GMRES.
- However, implicit methods usually require a large amount of memory to store the Jacobian matrix, which may not be feasible for large - scale problems.

#### Solutions in the paper

1. **Develop implicit non - compact and compact high - order gas - kinetic schemes (HGKS)**
- For non - compact schemes, the simple WENO reconstruction method is used to achieve spatial accuracy and is combined with the nonlinear Generalized Minimum Residual (GMRES) method for implicit discretization.
- For compact schemes, the Hermite WENO (HWENO) reconstruction method is used, where the reconstruction stencil contains only one layer of neighboring cells, thereby improving resolution and parallelism.

2. **GPU - accelerated computation**
- To further accelerate the computation, the authors implemented the implicit non - compact and compact HGKS methods on the Graphics Processing Unit (GPU) through CUDA.
- The experimental results show that the GPU code achieves an 8 - fold speedup compared to the CPU code.

3. **Numerical verification**
- Through a series of three - dimensional numerical experiments from subsonic to supersonic flows, the accuracy, robustness and efficiency of the new method are verified.

In summary, this paper solves the problem of slow convergence in steady - state compressible flow simulations by developing implicit high - order gas - kinetic schemes and GPU - acceleration techniques, providing an efficient computational tool for flow simulations of complex geometries.