Example of particle tracking

[markm0705]

Dear Elmer Forum

I've only recently discovered Elmer and looking for guidance as to whether this is a tool I can use for particle investigations. I'm a geologist and geostatistician with an interest in sampling theory. I have been using the 2D FEM tool 'Algodoo' to investigate various aspects of mechanical sampling of particles (see https://www.linkedin.com/pulse/brazil-n ... plash=true). However, I have quickly found that 2D is not that realistic for some of the sampling scenarios I wish to investigate - really a 3D method is required. I found Elmer through the YouTube video (https://www.youtube.com/watch?v=GNt1Z-vk76g), which seems to depict the types of scenarios I would like to model - basically particles interacting with various surfaces and sampling tools. So the question is - Is Elmer the right tool for types of investigations I would like to make? If so I would particularly like to get a copy of the code use to create the YouTube video mentioned previously as first investigation to see how how Elmer works for the type of scenarios I wish to investigate. I would also be greatful as to any other suggestions on the a way forward to learn what I need would be helpful - I have a basic programming experience with Fortran but by am by no means an expert so tend to rely on compiled code of others.

Regards
Mark Murphy

[raback]

Hi Mark

The particle tracking utility of Elmer was something that was designed to be an additional feature on top of finite element analysis. Compared to simple particle tracking methods the particles are followed in the finite elements and not in uniform grid. This results to an overhead which means that it is quite suboptimal to use Elmer for particle-only simulations. Rather the particles should be affected by the continuous fields.

Your nice application included several particle types. This should be added somehow. Basically any property can be attached to the particle but still there would be some additional work. So far the applications have included only one type of particles. The most common applications have been electrons and passive tracers.

To conlude I see some potential but still some non-trivial steps might be needed to cover your application. I don't now have any example at hand...

-Peter

[markm0705]

Peter

Thanks for your quick reply. You have confirmed my initial suspicions regarding that Elmer may not be the best for individual particle tracking. I've found a number of commercial packages available that may be up to the task but I'm afraid the expense is outside the budget of my curiosity!. No matter - I will put this on the 'investigate in future shelf' and perhaps have a crack at writing some code for the task when I have more free time to explore.

Thanks and regards

Mark M

[raback]

Hi Mark

I think LAMMPS could be a good OS option. There seems to be work on that also for "granular flow".

-Peter

[markm0705]

Peter

Thanks again - I'll shall have a look at you suggestion

Regards

Mark M

[mzenker]

Hi,

you might also want to check out radiance (http://radsite.lbl.gov/radiance/), which is a raytracing soft ware I have used ages ago for my PhD thesis. I have no idea how it has evolved since then and if it suits your needs. It may be worth a look, though.

HTH,

Matthias