How do the units in the simulation relate to physical SI units?

[w.r.co...@student.rug.nl]

We're having difficulty determining how the units in the simulation relate to physical SI units. In particular, when adapting the movingObstacle example, adjusting the res variable results in different velocity profiles. This suggests that a different number of cells corresponds to different physical dimensions. Similarly, altering the timestep also appears to change the simulated physics. We're curious to know what the relation of the dimensions and timestep of the simulation is to SI units, so we can compare our simulations with real-world data. Is there a way to determine or estimate this?

Kind regards,

Wester Coenraads

[Nils Thuerey]

Hi, good question - in general, it's not advisable to directly run numerical calculations in real world units, and mantaflow uses a normalized cell size of 1 for that reason. The timestep can be set arbitrarily, but usually values around 1 are also recommended. That means you have to rescale the values to compute real world quantities. The scenes/lidDrivenCavity.py example includes viscosity and computes a Reynolds number for the flow. I'd recommend that one as a starting point.

Cheers, Nils

[w.r.co...@student.rug.nl]

Hi Nils,

Thank you for the speedy response! We already had a suspicion we might be able to figure something out based on the Reynolds number. We'll make an attempt with the scene you suggested as a starting point.

Cheers,

Wester