Discrete Miranda-Talenti estimates and applications to linear and nonlinear PDEs (Q2423689)
From MaRDI portal
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Discrete Miranda-Talenti estimates and applications to linear and nonlinear PDEs |
scientific article |
Statements
Discrete Miranda-Talenti estimates and applications to linear and nonlinear PDEs (English)
0 references
20 June 2019
0 references
The paper is concerned with the numerical solution of second-order elliptic equations in non-divergent form with non-differentiable coefficients. In this case, the variational formulation cannot be derived by the standard approach using Green's theorem. The authors focus on the case where a domain \(\Omega\) is convex and the coefficient matrix is possibly discontinuous, but essentially bounded and satisfies the Cordes condition. It is known that then there exists a strong solution \(u \in H^2(\Omega)\). The authors derive the discrete Miranda-Talenti inequality, which relates the \(L^2\)-norm and the \(H^2\)-seminorm for a function \(v_h\) from the Lagrange finite element space. Based on this inequality, they propose the finite element method for linear elliptic problems in non-divergence form and derive optimal error estimates. The results are then extended to the nonlinear Hamilton-Jacobi-Bellman equation. Numerical experiments are provided to confirm the theoretical results and to illustrate the efficiency and applicability of the method.
0 references
non-divergence form
0 references
Miranda-Talenti inequality
0 references
Hamilton-Jacobi-Bellman equation
0 references
finite element methods
0 references
convergence
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references