Regular Article
Optimized Compact-Difference-Based Finite-Volume Schemes for Linear Wave Phenomena

https://doi.org/10.1006/jcph.1997.5836Get rights and content

Abstract

High-order compact-difference-based finite-volume schemes are developed, analyzed, and implemented for linear wave propagation phenomena with particular emphasis on computational electromagnetics in the time-domain. The formulation combines the primitive function approach with five-point spatially sixth- and fourth-order methods. Optimization in the semi-discrete case is achieved by minimizing dispersion and isotropy error. The fully discrete scheme is examined by adopting the classical fourth-order Runge–Kutta technique. Stability bounds are established and the coefficients for the spatial discretization are readjusted for optimum performance. The scheme is then formally extended to multiple dimensions. Consistent boundary conditions are presented for the reconstruction operator as well as flux specification. Several calculations in one and three dimensions confirm the properties of the method.

References (30)

  • M.H. Carpenter et al.

    Time-stable boundary conditions for finite-difference schemes solving hyperbolic systems: Methodology and application to high-order compact systems

    J. Comput. Phys.

    (1994)
  • K.S. Yee

    Numerical solution of initial boundary value problems involving Maxwell's equations in isotropic media

    IEEE Trans. Antennas Propag.

    (1966)
  • V. Shankar et al.

    A CFD-based finite-volume procedure for computational electromagnetics—Interdisciplinary applications for CFD methods

    AIAA Paper

    (1989)
  • J.S. Shang et al.

    Characteristic-based, time-dependent Maxwell equation solvers on a general curvilinear frame

    AIAA J.

    (1995)
  • Cited by (115)

    • New hybrid compact schemes for structured irregular meshes using Birkhoff polynomial basis

      2020, Journal of Computational Physics
      Citation Excerpt :

      Compact schemes have been extensively accustomed to the simulations of initial boundary value problems on various collocated and staggered meshes [1–6].

    • Hybrid sixth order spatial discretization scheme for non-uniform Cartesian grids

      2017, Computers and Fluids
      Citation Excerpt :

      Despite performance gains by the use of non-uniform compact schemes, the present method is limited to geometries with grids which vary along mutually orthogonal directions only. Many Padé compact schemes were designed with implicit closures for the boundary and near-boundary points [1,10,17,32], and it was noted [26] that the resultant schemes displayed anti-diffusion near the inflow of the domain. In [17], performance of numerical first derivative at interior nodes was characterized by an equivalent wavenumber (keq), without addressing the issues of boundary closure schemes.

    • A scalable fully implicit method with adaptive time stepping for unsteady compressible inviscid flows

      2016, Computers and Structures
      Citation Excerpt :

      Compared to the general second order methods such as the LLF scheme, the main advantage of the compact schemes is their capability to achieve arbitrarily high order formal accuracy in smooth regions while maintaining stable, sharp discontinuity transitions, and non-oscillatory. Ref. [11] combines the compact schemes with finite volume formulation, which presents a class of arbitrary high order finite volume scheme. In [10,30,39], the ideas of high order accurate weighted essentially nonoscillatory (WENO) and upwind schemes are introduced into the compact schemes, which can obtain more accurate results and lead to substantial reduction in the total computing time.

    • A new alternating bi-diagonal compact scheme for non-uniform grids

      2016, Journal of Computational Physics
      Citation Excerpt :

      Compact schemes for solving PDE originated from Padé schemes for discretization of ODEs and the former can be found in [1,8,11,12,25] among many other references. For example, optimized schemes have been used for time-domain Maxwell equation in [5]. A comprehensive account of compact schemes and GSA employed to test the numerical properties of space–time dependent problems can be found in [13].

    View all citing articles on Scopus
    1

    Corresponding author. OAI (ICOMP).

    f1

    [email protected]

    View full text