Abstract:The mixed finite element method for the Poisson problem with the Raviart-Thomas elements of low-level can be interpreted as a finite volume method with a non-local gradient. In this contribution, we propose a variant of Petrov-Galerkin type for this problem to ensure a local computation of the gradient at the interfaces of the elements. The shape functions are the Raviart-Thomas finite elements. Our goal is to define test functions that are in duality with these shape functions: Precisely, the shape and test functions will be asked to satisfy a L2-orthogonality property. The general theory of Babuška brings necessary and sufficient stability conditions for a Petrov-Galerkin mixed problem to be convergent. We propose specific constraints for the dual test functions in order to ensure stability. With this choice, we prove that the mixed Petrov-Galerkin scheme is identical to the four point finite volumes scheme of Herbin, and to the mass lumping approach developed by Baranger, Maitre and Oudin. Finally, we construct a family of dual test functions that satisfy the stability conditions. Convergence is proven with the usual techniques of mixed finite elements.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: in the framework of mixed finite elements, how to ensure the local calculation of the discrete gradient and make this method consistent with the results of the finite - volume method. Specifically, the author hopes to achieve a discrete - gradient calculation similar to the finite - volume method in the Petrov - Galerkin - type Raviart - Thomas finite - element space. To achieve this goal, the author proposes a new method for selecting test functions, so that these test functions satisfy a certain duality (orthogonality) with the Raviart - Thomas shape functions, thereby ensuring the locality and stability of the discrete gradient.
### Problem Background
The finite - volume method is widely used because it can explicitly and locally calculate the discrete gradient. This calculation only depends on the values of a given scalar field. However, for elliptic partial differential equations (such as the Poisson equation), traditional mixed finite - element methods (such as the Raviart - Thomas finite - element method) usually cannot directly provide a local discrete gradient. Therefore, the author hopes to find a method in the framework of mixed finite elements so that the discrete gradient can be locally calculated at element interfaces.
### Solution
The author proposes the following solutions:
1. **Define Dual Test Functions**:
- Select the test - function space so that it forms a dual relationship with the Raviart - Thomas shape - function space, that is, satisfies certain orthogonality conditions.
- These test functions should be orthogonal to the shape functions in the sense of the \(L^2\) inner product, that is, \((\phi_a^*, \phi_b)_0 = 0\) when \(a\neq b\).
2. **Ensure the Locality of the Discrete Gradient**:
- By selecting appropriate dual test functions, make the calculation of the discrete gradient depend only on the values of the scalar field near the element interfaces, not on the source term.
3. **Stability Analysis**:
- Use Babuška's general theory to give the necessary and sufficient conditions for the convergence of the mixed Petrov - Galerkin problem.
- Propose specific constraint conditions to ensure that the selection of dual test functions can meet these stability conditions.
4. **Relationship with the Finite - Volume Method**:
- Prove that the proposed mixed Petrov - Galerkin scheme is equivalent to Herbin's four - point finite - volume scheme and the mass - condensation method proposed by Baranger et al.
### Results
Through the above methods, the author has proved that:
- The proposed mixed Petrov - Galerkin scheme can ensure the local calculation of the discrete gradient.
- This scheme is equivalent to existing finite - volume methods (such as the four - point finite - volume scheme) and has good stability and convergence.
### Mathematical Formulas
The key mathematical formulas include:
- Definition of the discrete gradient:
\[
(\nabla_T u, \phi_a)_0=\begin{cases}
\frac{u_L - u_K}{(\phi_a, \phi_a^*)_0} & \text{if } a\in T_1^i, \partial c_a=(K, L) \\
\frac{-u_K}{(\phi_a, \phi_a^*)_0} & \text{if } a\in T_1^b, \partial c_a=(K)
\end{cases}
\]
- Orthogonality conditions for dual basis functions:
\[
(\phi_a^*, \phi_b)_0 = 0\quad \text{for } a\neq b
\]
- Flux normalization condition:
\[
\int_b \phi_a^*\cdot n_b \, ds=\delta_{ab}
\]
Through these formulas and conditions, the author has successfully constructed a discrete - gradient calculation method that can maintain locality and ensure stability.