TY - JOUR
T1 - Discontinuous Petrov-Galerkin method based on the optimal test space norm for steady transport problems in one space dimension
AU - Niemi, Antti
AU - Collier, Nathan
AU - Calo, Victor M.
N1 - KAUST Repository Item: Exported on 2020-10-01
PY - 2013/5
Y1 - 2013/5
N2 - We revisit the finite element analysis of convection-dominated flow problems within the recently developed Discontinuous Petrov-Galerkin (DPG) variational framework. We demonstrate how test function spaces that guarantee numerical stability can be computed automatically with respect to the optimal test space norm. This makes the DPG method not only stable but also robust, that is, uniformly stable with respect to the Péclet number in the current application. We employ discontinuous piecewise Bernstein polynomials as trial functions and construct a subgrid discretization that accounts for the singular perturbation character of the problem to resolve the corresponding optimal test functions. We also show that a smooth B-spline basis has certain computational advantages in the subgrid discretization. The overall effectiveness of the algorithm is demonstrated on two problems for the linear advection-diffusion equation. © 2011 Elsevier B.V.
AB - We revisit the finite element analysis of convection-dominated flow problems within the recently developed Discontinuous Petrov-Galerkin (DPG) variational framework. We demonstrate how test function spaces that guarantee numerical stability can be computed automatically with respect to the optimal test space norm. This makes the DPG method not only stable but also robust, that is, uniformly stable with respect to the Péclet number in the current application. We employ discontinuous piecewise Bernstein polynomials as trial functions and construct a subgrid discretization that accounts for the singular perturbation character of the problem to resolve the corresponding optimal test functions. We also show that a smooth B-spline basis has certain computational advantages in the subgrid discretization. The overall effectiveness of the algorithm is demonstrated on two problems for the linear advection-diffusion equation. © 2011 Elsevier B.V.
UR - http://hdl.handle.net/10754/562741
UR - https://linkinghub.elsevier.com/retrieve/pii/S1877750311000652
UR - http://www.scopus.com/inward/record.url?scp=84875868851&partnerID=8YFLogxK
U2 - 10.1016/j.jocs.2011.07.003
DO - 10.1016/j.jocs.2011.07.003
M3 - Article
SN - 1877-7503
VL - 4
SP - 157
EP - 163
JO - Journal of Computational Science
JF - Journal of Computational Science
IS - 3
ER -