TY - JOUR
T1 - Enhanced solution of 2D incompressible Navier–Stokes equations based on an immersed-boundary generalized harmonic polynomial cell method
AU - Yu, Xueying
AU - Fuhrman, David R.
AU - Shao, Yanlin
AU - Liao, Kangping
AU - Duan, Wenyang
AU - Zhang, Yunxing
PY - 2021
Y1 - 2021
N2 - Poisson equation lies in the heart of the numerical solutions of incompressible Navier–Stokes equations (NSEs) based on the popular projection methods, which decouple the pressure and velocities fields. This paper presents enhanced numerical solutions of the 2D incompressible NSEs inspired by a newly-developed Generalized Harmonic Polynomial Cell (GHPC) method for the Poisson equation, which has a fourth-order spatial accuracy. To achieve that, the original GHPC method is adapted to accommodate immersed boundaries, necessary when a staggered grid for velocities and pressure is applied. The finite difference method (FDM) is used in the spatial discretization of the diffusion and advection terms. Numerical analyses of lid-driven cavity flow, a Taylor–Green vortex, and flow around a smooth circular cylinder all show encouraging results, confirming the accuracy and efficiency of the new solver. Through comparison with an NSEs solver which applies second-order FDM for the Poisson equation, we demonstrate that the accuracy of the pressure solution can be greatly improved by applying the more accurate GHPC method for the Poisson equation, even though the accuracy of the velocities solutions are limited by the numerical schemes for advection-diffusion equations. The accuracy in the pressure solution can also be translated into CPU time saving to achieve a predefined accuracy. The present immersed-boundary GHPC Poisson equation solver can easily be ‘plugged’ into other existing NSEs solvers utilizing staggered grids.
AB - Poisson equation lies in the heart of the numerical solutions of incompressible Navier–Stokes equations (NSEs) based on the popular projection methods, which decouple the pressure and velocities fields. This paper presents enhanced numerical solutions of the 2D incompressible NSEs inspired by a newly-developed Generalized Harmonic Polynomial Cell (GHPC) method for the Poisson equation, which has a fourth-order spatial accuracy. To achieve that, the original GHPC method is adapted to accommodate immersed boundaries, necessary when a staggered grid for velocities and pressure is applied. The finite difference method (FDM) is used in the spatial discretization of the diffusion and advection terms. Numerical analyses of lid-driven cavity flow, a Taylor–Green vortex, and flow around a smooth circular cylinder all show encouraging results, confirming the accuracy and efficiency of the new solver. Through comparison with an NSEs solver which applies second-order FDM for the Poisson equation, we demonstrate that the accuracy of the pressure solution can be greatly improved by applying the more accurate GHPC method for the Poisson equation, even though the accuracy of the velocities solutions are limited by the numerical schemes for advection-diffusion equations. The accuracy in the pressure solution can also be translated into CPU time saving to achieve a predefined accuracy. The present immersed-boundary GHPC Poisson equation solver can easily be ‘plugged’ into other existing NSEs solvers utilizing staggered grids.
KW - Pressure Poisson equation
KW - Projection method
KW - Immersed boundary method
KW - Generalized harmonic polynomial cell method
KW - Incompressible Navier–Stokes equations
U2 - 10.1016/j.euromechflu.2021.04.009
DO - 10.1016/j.euromechflu.2021.04.009
M3 - Journal article
SN - 0997-7546
VL - 89
SP - 29
EP - 44
JO - European Journal of Mechanics B - Fluids
JF - European Journal of Mechanics B - Fluids
ER -