Efficient high-order discontinuous Galerkin schemes with first-order hyperbolic advection-diffusion system approach

被引:8
作者
Mazaheri, Alireza [1 ]
Nishikawa, Hiroaki [2 ]
机构
[1] NASA, Langley Res Ctr, Hampton, VA 23681 USA
[2] Natl Inst Aerosp, Hampton, VA 23666 USA
关键词
High-order; Discontinuous Galerkin (DG); DG with hyperbolic first-order system (DG-H); Advection-diffusion; WENO; Interior Penalty (IP); FINITE-VOLUME SCHEMES; RESIDUAL-DISTRIBUTION SCHEMES; NAVIER-STOKES EQUATIONS; ELLIPTIC PROBLEMS; CONSERVATION-LAWS; INTERIOR PENALTY; UNSTRUCTURED MESHES; ELEMENT-METHOD; TRIANGULAR MESHES; WENO LIMITERS;
D O I
10.1016/j.jcp.2016.06.006
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
We propose arbitrary high-order discontinuous Galerkin (DG) schemes that are designed based on a first-order hyperbolic advection-diffusion formulation of the target governing equations. We present, in details, the efficient construction of the proposed high-order schemes (called DG-H), and show that these schemes have the same number of global degrees-of-freedom as comparable conventional high-order DG schemes, produce the same or higher order of accuracy solutions and solution gradients, are exact for exact polynomial functions, and do not need a second-derivative diffusion operator. We demonstrate that the constructed high-order schemes give excellent quality solution and solution gradients on irregular triangular elements. We also construct a Weighted Essentially Non-Oscillatory (WENO) limiter for the proposed DG-H schemes and apply it to discontinuous problems. We also make some accuracy comparisons with conventional DG and interior penalty schemes. A relative qualitative cost analysis is also reported, which indicates that the high-order schemes produce orders of magnitude more accurate results than the low-order schemes for a given CPU time. Furthermore, we show that the proposed DG-H schemes are nearly as efficient as the DG and Interior-Penalty (IP) schemes as these schemes produce results that are relatively at the same error level for approximately a similar CPU time. Published by Elsevier Inc.
引用
收藏
页码:729 / 754
页数:26
相关论文
共 75 条
[1]   Hybrid scheduling for the parallel solution of linear systems [J].
Amestoy, PR ;
Guermouche, A ;
L'Excellent, JY ;
Pralet, S .
PARALLEL COMPUTING, 2006, 32 (02) :136-156
[2]   A fully asynchronous multifrontal solver using distributed dynamic scheduling [J].
Amestoy, PR ;
Duff, IS ;
L'Excellent, JY ;
Koster, J .
SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 2001, 23 (01) :15-41
[3]  
[Anonymous], P 20 AIAA COMP FLUID
[4]   Unified analysis of discontinuous Galerkin methods for elliptic problems [J].
Arnold, DN ;
Brezzi, F ;
Cockburn, B ;
Marini, LD .
SIAM JOURNAL ON NUMERICAL ANALYSIS, 2002, 39 (05) :1749-1779
[5]   AN INTERIOR PENALTY FINITE-ELEMENT METHOD WITH DISCONTINUOUS ELEMENTS [J].
ARNOLD, DN .
SIAM JOURNAL ON NUMERICAL ANALYSIS, 1982, 19 (04) :742-760
[6]  
Arnold DN, 1979, THESIS
[7]  
BAKER GA, 1977, MATH COMPUT, V31, P45, DOI 10.1090/S0025-5718-1977-0431742-5
[8]   PIECEWISE SOLENOIDAL VECTOR-FIELDS AND THE STOKES PROBLEM [J].
BAKER, GA ;
JUREIDINI, WN ;
KARAKASHIAN, OA .
SIAM JOURNAL ON NUMERICAL ANALYSIS, 1990, 27 (06) :1466-1485
[9]   A high-order accurate discontinuous finite element method for the numerical solution of the compressible Navier-Stokes equations [J].
Bassi, F ;
Rebay, S .
JOURNAL OF COMPUTATIONAL PHYSICS, 1997, 131 (02) :267-279
[10]  
Bassi F., 1997, 2 EUROPEAN C TURBOMA, P99