Higher-order conservative discretizations on arbitrarily varying non-uniform grids

A. Arun Govind Neelan,Raimund Bürger,Manoj T. Nair,Samala Rathan
DOI: https://doi.org/10.1007/s40314-024-02987-x
2024-11-10
Computational and Applied Mathematics
Abstract:Conservative discretizations of transport equations are based on integral formulations that include the finite volume method (FVM) and conservative finite difference methods (CFDMs). The FVM is used by most fluid dynamics simulation packages and requires smoothly shifting grids for better convergence. This motivates the study of the order of accuracy and rate of convergence of the FVM on non-uniform grids. It is difficult to do such an analysis of the FVM on an unstructured grid; however, the FVM is reduced to a CFDM on a Cartesian grid. The effect of the order of accuracy and the rate of convergence of higher-order CFDMs on arbitrarily varying grids are investigated. It is shown that higher-order conservative discretizations on arbitrarily varying non-uniform grids need some smoothness in the grid transition to be first-order accurate. The condition to achieve first-order accuracy is also presented. If the grid is replaced by a gradually varying grid, it is shown that conservative discretizations yield a better rate of convergence. In this situation, a rate of convergence between one and the theoretical maximum can be achieved in dependence on the grid stretch/contraction ratio. Numerical examples, including the linear convection-diffusion equation, the lid-driven cavity problem, and the Taylor-Green vortex problem, are presented.
mathematics, applied
What problem does this paper attempt to address?