Interpolating to a higher function space

OasisMove provides the ability to interpolate a solution to a higher function space. For instance, interpolating a \(\mathbb{P_1}\) solution to \(\mathbb{P_2}\) function space. This is particularly useful when you want to increase the resolution of your solution, or when you need a higher degree of accuracy. Note that this feature is currently implemented for the velocity field only.

Interpolating the velocity to a higher function space

The interpolation technique in OasisMove is closely related to restarting a simulation. To demonstrate interpolation of the velocity from \(\mathbb{P_1}\) to \(\mathbb{P_2}\) function space, we will consider the DrivenCavity problem. By default, the problem runs for \(T=10\) using \(\mathbb{P_1}\) elements, determined by the argument velocity_degree=1. Assume we have run the problem using the default parameters:

$ oasismove NSfracStep solver=IPCS_ABCN problem=DrivenCavity 

To restart and continue the simulation until \(T=20\) using \(\mathbb{P_2}\) elements, we may simply pass the velocity_degree=2 argument, in addition to the required parameters for restarting a simulation:

$ oasismove NSfracStep solver=IPCS_ABCN problem=DrivenCavity velocity_degree=2 restart_folder=results_driven_cavity/data/1/Checkpoint T=20 

Note that the ability to interpolate to a higher function space can be computation-intensive, especially for large simulations and higher function spaces.