A \(p\)-version finite element method for nonlinear elliptic variational inequalities in 2D (Q861656)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | A \(p\)-version finite element method for nonlinear elliptic variational inequalities in 2D |
scientific article |
Statements
A \(p\)-version finite element method for nonlinear elliptic variational inequalities in 2D (English)
0 references
30 January 2007
0 references
This paper deals with the design, analysis, and implementation of a \(p\)-finite element method (FEM) method for nonlinear variational inequalities arising from the obstacle problem for second-order quasi-linear elliptic operators. The solution is approximated by controlling the obstacle condition in images of the Gauss-Lobatto points. In the general case the FEM subspaces can be \(h\), \(p\) or \(hp\). The \(h\)-version FEM obtains the convergence by mesh refinement, whereas the \(p\)-version FEM obtains the convergence by increasing the polynomial degree of the subspace. The authors deduce the existence and the uniqueness of the discrete solution for the \(p\)-version for the obstacle problem. They also prove the convergence with respect to the energy norm, and assuming some additional regularity properties for the original solution, obtain an a priori error estimate which corresponds to the one of \textit{R. S. Falk} for the \(h\)-version [Math. Comput. 28, 963--971 (1974; Zbl 0297.65061)]. Numerical experiments are reported and the authors asserts that the \(p\)-version turns out to be superior to the \(h\)-version concerning the convergence rate and the number of unknowns necessary to get a prescribed exactness of the approximation.
0 references
finite element method
0 references
nonlinear elliptic variational inequalities
0 references
second-order quasi-linear elliptic operators
0 references
convergence
0 references
mesh refinement
0 references
\(p\)-version
0 references
obstacle problem
0 references
a priori error estimate
0 references
numerical experiments
0 references
0 references
0 references