Resolved particle inlet

[jcs]

Good morning,

I am aiming to model a filtration process using the HLBM resolved particle method in OpenLB. My intention for an initial model is to introduce particles through a constant velocity inlet boundary towards a circular obstruction, with a pressure outlet opposite the inlet and periodic boundaries for the other two BC’s. This will be a 2D domain.

First of all I’f be grateful if you can see anything wrong with using HLBM for this task, or any other glaring issues. If it seems like a reasonable approach then my question is how can I go about introducing particles from “outside” the domain?

I could not find an example in the “examples”, or a BC in src/boundary or src/particles/boundary that would achieve this, though I’ve seen the periodic boundary implementation in the below link that seems to be adjacent to what I want.
https://www.openlb.net/particle-swarm-sedimentation/
https://doi.org/10.1016/j.cpc.2024.109321

Many thanks,
James

[mathias]

The setup from the video uses periodic boundary conditions. For partcles emerging, you can update the proiosity field as done in the inital step but date it up every timestep, so basically move the particle manually until it is entirely in the domain, then add that particle the the particle dynamics class!

[jcs]

Thanks Mathias for your fast response. I’ve been attempting your suggestion in the meantime, but encountered the issue of a pressure wave emanating from the particle once the particle dynamics were activated. I’ve managed to avoid this issue now by using the setField function to set the particle velocity immediately after executing the particle manager for the first time.

auto particleA = particleSystem.get( 0 );
Vector<T,2> velocitynu = {0.1,0};
particleA.template setField<MOBILITY,VELOCITY>(velocitynu);

[mathias]

Nice that it works, yes of course, that makes sence, you need to udate the velocity and porousity!

[Author]

Posts