Toggle menu
71
399
28
3.6K
Medusa: Coordinate Free Meshless Method implementation
Toggle preferences menu
Toggle personal menu
Not logged in
Your IP address will be publicly visible if you make any edits.

Hyperviscosity: Difference between revisions

From Medusa: Coordinate Free Meshless Method implementation
Created page with "One of the most commonly mentioned drawbacks of the RBF-FD method is its stability. Furthermore, the RBF-FD method, due to scattered nodes, sometimes produces system (differen..."
 
No edit summary
 
Line 1: Line 1:
One of the most commonly mentioned drawbacks of the RBF-FD method is its stability. Furthermore, the RBF-FD method, due to scattered nodes, sometimes produces system (differential) matrices with spurious eigenvalues. Those are far more common than with for example FVM or FEM method. The instabilities arise fastly even with linear PDEs.
One of the most common drawbacks of the RBF-FD method is its stability. Furthermore, the RBF-FD method, due to scattered nodes, sometimes produces system (differential) matrices with spurious eigenvalues. Those are far more common than with for example FVM or FEM method.  


With meshless methods (especially RBF-FD) one usually stabilises such schemes by adding a higher-order Laplacian (hyperviscosity) operator on the right-hand side.
For such unstable systems, one usually considers numerical diffusion in the form of hyperviscosity. The hyperviscosity scheme introduces higher-order Laplacian operator (i.e. hyperviscosity operator) to the right hand side of the PDE


$$
$$
  +\gamma \Delta^\alpha u
  +\gamma \Delta^\alpha u,
$$
$$


The final scheme is referred to as hyperviscosity scheme
where $\gamma$ is a hyperviscosity constant and $\alpha$ is the order of hyperviscosity.

Latest revision as of 14:24, 23 March 2024

One of the most common drawbacks of the RBF-FD method is its stability. Furthermore, the RBF-FD method, due to scattered nodes, sometimes produces system (differential) matrices with spurious eigenvalues. Those are far more common than with for example FVM or FEM method.

For such unstable systems, one usually considers numerical diffusion in the form of hyperviscosity. The hyperviscosity scheme introduces higher-order Laplacian operator (i.e. hyperviscosity operator) to the right hand side of the PDE

$$

+\gamma \Delta^\alpha u,

$$

where $\gamma$ is a hyperviscosity constant and $\alpha$ is the order of hyperviscosity.