Varying Bond model parameters to mimic soil particle cohesion/stiction

[cnmatu]

Hi,

I am running an internal erosion model where particles get eroded and exit out of an aluminium cylindrical mould (image attached). However, the soil particles are cohesive in nature and therefore I have to use the bond model module which I have used but I am wondering the best way to calibrate these bond model parameters (Tangential stiffness per area, Normal stiffness per area, shear stress limit and tensile stress limit)so that I achieve the exact cohesion of my soil so that when I run a one-way or two-way coupled simulation, it simulates the exact experimental result. Any advice on this? Any reading material that could help?