Dear community,
I'm currently performing Navier-Stokes simulations of granular collapses on different slopes, based on Pierre-Yves Lagrée's example entitled '
Continuum collapse of a column of grains real 2D, no slip at the base'. The general idea is to vary the slope angle up to 36° in order to compare the results with large-scale experiments already carried out, using Coulomb and mu(I)-rheology.
I have several issues, all of which are likely related to pressure calculation, including the slope effect.
- Regardless of rheology, the calculated pressure appears unstable as the slope increases without affecting the velocity (see attached videos). I included the slope only in gravity and pressure initiation (see attached script).
- I have no deposits along the plane above theta ~ 25°, while the angle of repose is set to 32°.
- Inherent to the numerical method, bubbles are trapped at the front, which significantly affect the flow dynamics at steep slopes. There is no solution to this problem, and therefore this numerical method is ineffective at steep slopes?
Thank you for your help,