• Corpus ID: 235313934

High-order flux reconstruction method for the hyperbolic formulation of the incompressible Navier-Stokes equations on unstructured grids

@article{Kamra2021HighorderFR,
  title={High-order flux reconstruction method for the hyperbolic formulation of the incompressible Navier-Stokes equations on unstructured grids},
  author={Mohamed M. Kamra and Jabir Al-Salami and Changhong Hu},
  journal={arXiv: Computational Physics},
  year={2021}
}

• Mohamed M. Kamra, Jabir Al-Salami, C. Hu
• Published 3 June 2021
• Mathematics
• arXiv: Computational Physics

A high-order Flux reconstruction implementation of the hyperbolic formulation for the incompressible Navier-Stokes equation is presented. The governing equations employ Chorin's classical artificial compressibility (AC) formulation cast in hyperbolic form. Instead of splitting the second-order conservation law into two equations, one for the solution and another for the gradient, the Navier-Stokes equation is cast into a first-order hyperbolic system of equations. Including the gradients in the AC iterative process results in a significant improvement in accuracy for the pressure, velocity, and its gradients. Furthermore, this treatment allows for taking larger time-steps since the hyperbolic formulation eliminates the restriction due to diffusion. Tests using the method of manufactured solutions show that solving the conventional form of the Navier-Stokes equation lowers the order of accuracy for gradients, while the hyperbolic method is shown to provide equal orders of accuracy for both the velocity and its gradients which may be beneficial in several applications. Two- and three-dimensional benchmark tests demonstrate the superior accuracy and computational efficiency of the developed solver in comparison to the conventional method and other published works. This study shows that the developed high-order hyperbolic solver for incompressible flows is attractive due to its accuracy, stability and efficiency in solving diffusion dominated problems.