Spectral methods for the Navier-Stokes equations with one infinite and two periodic directions
Journal of Computational Physics
A spectral method for polar coordinates
Journal of Computational Physics
Zonal embedded grids for numerical simulations of wall-bounded turbulent flows
Journal of Computational Physics
Matrix computations (3rd ed.)
A spectral method for unbounded domains
Journal of Computational Physics
B-spline method and zonal grids for simulations of complex turbulent flows
Journal of Computational Physics
A critical evaluation of the resolution properties of B-Spline and compact finite difference methods
Journal of Computational Physics
A high order multivariate approximation scheme for scattered data sets
Journal of Computational Physics
Hi-index | 31.45 |
The numerical method presented in this paper aims at solving the incompressible Navier-Stokes equations in unbounded domains. The problem is formulated in cylindrical coordinates and the method is based on a Galerkin approximation scheme that makes use of vector expansions that exactly satisfy the continuity constraint. More specifically, the divergence-free basis vector functions are constructed with Fourier expansions in the θ and z directions while mapped B-splines are used in the semi-infinite radial direction. Special care has been taken to account for the particular analytical behaviors at both end points r = 0 and r → ∞. A modal reduction algorithm has also been implemented in the azimuthal direction, allowing for a relaxation of the CFL constraint on the timestep size and a possibly significant reduction of the number of DOF. The time marching is carried out using a mixed quasi-third order scheme. Besides the advantages of a divergence-free formulation and a quasi-spectral convergence, the local character of the B-splines allows for a great flexibility in node positioning while keeping narrow bandwidth matrices. Numerical tests show that the present method compares advantageously with other similar methodologies using purely global expansions.