Discontinuous Galerkin Finite Volume Element
Methods for Second-Order Linear
Elliptic Problems
Sarvesh Kumar, Neela Nataraj, Amiya K. Pani
Industrial Mathematics Group, Department of Mathematics, Indian Institute of
Technology Bombay, Powai, Mumbai 400076, India
Received 3 October 2007; accepted 20 August 2008
Published online 30 April 2009 in Wiley InterScience (www.interscience.wiley.com).
DOI 10.1002/num.20405
In this article, a one parameter family of discontinuous Galerkin finite volume element methods for approx-
imating the solution of a class of second-order linear elliptic problems is discussed. Optimal error estimates
in L
2
and broken H
1
- norms are derived. Numerical results confirm the theoretical order of convergences.
© 2009 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq 25: 1402–1424, 2009
Keywords: discontinuous Galerkin methods; finite volume element methods; optimal error estimates
I. INTRODUCTION
In this article, we formulate and analyze a one parameter family of discontinuous Galerkin finite
volume element methods (DGFVEM) for approximating the solution of the following elliptic
problem: Given f , find u such that
−∇ · (K ∇u) = f in , (1.1)
u = 0 on ∂,
where is a bounded, convex polygonal domain in R
2
with boundary ∂ and K = (k
ij
(x))
2×2
is a real valued, symmetric and uniformly positive definite matrix, i.e., there exists a positive
constant α
0
such that
ξ
T
Kξ ≥ α
0
ξ
T
ξ ∀ξ ∈ R
2
. (1.2)
In recent years, there has been a renewed interest in Discontinuous Galerkin (DG) methods
for the numerical approximation of partial differential equations. This is due to their flexibility
in local mesh adaptivity and handling nonuniform degrees of approximation for solutions whose
smoothness exhibit variation over the computational domain. DG methods have an advantage
Correspondence to: Neela Nataraj, Industrial Mathematics Group, Department of Mathematics, Indian Institute of
Technology Bombay, Powai, Mumbai 400076, India (e-mail: nataraj.neela@gmail.com)
© 2009 Wiley Periodicals, Inc.