Brittany D. Froese,Adam M. Oberman
Abstract:The elliptic Monge-Ampère equation is a fully nonlinear Partial Differential Equation that originated in geometric surface theory and has been applied in dynamic meteorology, elasticity, geometric optics, image processing and image registration. Solutions can be singular, in which case standard numerical approaches fail. Novel solution methods are required for stability and convergence to the weak (viscosity) solution.
In this article we build a wide stencil finite difference discretization for the \MA equation. The scheme is monotone, so the Barles-Souganidis theory allows us to prove that the solution of the scheme converges to the unique viscosity solution of the equation.
Solutions of the scheme are found using a damped Newton's method. We prove convergence of Newton's method and provide a systematic method to determine a starting point for the Newton iteration.
Computational results are presented in two and three dimensions, which demonstrates the speed and accuracy of the method on a number of exact solutions, which range in regularity from smooth to non-differentiable.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is to develop an effective numerical method for solving the viscosity solution of the elliptic Monge - Ampère equation. Specifically, the article aims to:
1. **Handle singular solutions**: When the conditions are not met, classical numerical methods may fail or be unstable. Therefore, new discretization and solution methods are required to approximate singular solutions.
2. **Ensure numerical stability**: Due to the nonlinear and elliptic nature of the Monge - Ampère equation, the solution method must maintain numerical stability and be able to correctly select convex solutions.
3. **Prove convergence**: By constructing a monotone finite - difference discretization scheme, ensure that the numerical solution converges to the unique viscosity solution.
### Specific problem description
The Monge - Ampère equation is a fully nonlinear partial differential equation (PDE), in the form of:
\[
\text{(MA)} \quad \det(D^2 u(x)) = f(x), \quad x \in \Omega
\]
where \( D^2 u \) is the Hessian matrix of the function \( u \), and \( f(x) \) is a given function. To ensure the ellipticity of the equation, a convexity constraint also needs to be satisfied:
\[
\text{(C)} \quad u \text{ is convex}
\]
In addition, the Dirichlet boundary condition is also considered:
\[
\text{(D)} \quad u(x) = g(x), \quad x \in \partial \Omega
\]
### Numerical challenges
1. **Singular solutions**: For singular solutions, standard numerical methods may become unstable or select the wrong non - convex solution.
2. **Weak solutions**: For singular solutions, an appropriate concept of weak solutions (such as viscosity solutions or Aleksandrov solutions) needs to be used.
3. **Convexity constraint**: The convexity constraint is crucial for uniqueness and stability, and any approximation method requires some selection principle to ensure convex solutions.
### Solution
The paper proposes a wide stencil finite - difference discretization method and proves the monotonicity and convergence of this method. The specific steps include:
1. **Construct a discretization scheme**: Discretize the second - order directional derivatives of the Hessian matrix through wide stencil to ensure the consistency and monotonicity of the discretization scheme.
2. **Solution method**: Use the damped Newton’s method to solve the discretized equation and provide a systematic method for determining the initial point of Newton iteration.
3. **Theoretical proof**: Use the Barles - Souganidis theory to prove that the solution of the discretization scheme converges to the unique viscosity solution.
### Experimental results
The paper shows the calculation results in two - dimensional and three - dimensional cases, verifying the speed and accuracy of this method, which is applicable to various exact solutions from smooth to non - differentiable.
In summary, the main objective of this paper is to develop a stable and convergent numerical method for solving the viscosity solution of the Monge - Ampère equation and to handle the possible problems of singular solutions.