There are currently several options for open LBM (lattice-Boltzmann) codes. Quick search gives at least the following: OpenLB, Walberla, Palabos and Musubi.
Our application will be flow in tomographic images of fibrous porous materials (porosity 50-90%) with low Reynolds number. The networks are homogeneous enough so that load balancing is not obligatory. Also, no need for grid refinement. Solid-fluid boundaries will be accurate enough with simple bounce back combined with the MRT collision operator - more sophisticated boundaries are not needed. The simulations are performed with a single powerful windows workstation.
My simple question is: What open LBM code to choose for these simulations? We have currently looked mainly at openLB, but would in our case another choice be preferable?
I would very much endorse openLB. There is an active community which will make it easier to finish your plans.
OpenLB has strong emphasize on industrial (turbulent flows). There are few examples on fluid flow in porous media, but the requirements I have are rather simple and should be easily met by OpenLB?
I can also recommend OpenLB. There is also LBMHexMesh, which is based on OpenFoam I think, but I have never used that.
You could look at https://github.com/CFD-GO/TCLB
It was evaluated in case similar to yours:
Article Pressure drop in flow across ceramic foams—A numerical and e...
And I'm more then happy to help with your application
Thank you all for answering. Especially, thank you, Michael Dzikowski, for offering help!
Hello, I think also LBSim could fit your needs:
https://github.com/fabioskomori/lbsim/wiki
Executables existed for windows and MacOSX, they can still be found via the wayback machine of the Internet Archive:
https://web.archive.org/web/20181224110706/http://lbsim.com.br/index.php/download
Hello, I can mention this application of Palabos with porous media that might be of interest for you https://github.com/je-santos/MultiphasePorousMediaPalabos. His author is active in the Palabos forum.
I admit to be biased in favour of palabos for clear conflict of interest, but I witness that Palabos has been used many times for porous media applications.
In addition we have the plan to add viscosity independent error bounday conditions that are interesting for porous media.
Cheers!
Dear Francesco, thank you for this suggestion. There seem to be many interesting options available!
Dear Antti, I am not sure if you are still in search of codes for this but I can second Michal Dzikowski above in regards to the open source TCLB code. In the codes experimental branch there is also an option for an arbitrary grid to remove the overhead of the solid nodes if the porosity of the sample is low (the overhead is balanced with additional connectivity overhead, so is best for low porosity samples).
The website for the code is still progressing, but there is a tutorial that walks through calculating the permeability of a Digital Rock Physics Benchmark that you may find useful: https://docs.tclb.io/tutorials/additional/1.-porous-media-flow/
Dear Travis Mitchell. Thank you very much for your contribution on this discussion. It seems that TCLB would indeed be a good choice. It seems to have also a Python interface which makes it more flexible in many cases.
Currently we are testing Walberla. It has also a Python interface.
Hello everyone, is there any forum exists in which we can share our LBM-related problems and could get some useful suggestions?
Hello Syed Waqar Ali Shah ! You can have a look at the Palabos Forum: https://palabos-forum.unige.ch/ . You can post about Palabos, but also about the lattice Boltzmann theory. If you are a Palabos user, and you like to share your research with the community, there is also the Palabos Online Seminars: https://palabos.unige.ch/community/palabos-online-seminar-series/.
In general, I can recommend choosing Palabos as LBM library mainly for three reasons: 1. The community, 2. The robustness, 3. The future potential (check out the ongoing project to port it to GPU https://palabos.unige.ch/community/cpu-gpu-80-days/).
Dear all. I have now used Walberla for a while in the python environment, lbmpy, for simulations of fluid flow in x-ray tomographic images of filters. The environment is user friendly and it can be learned quickly. The people behind the code have been friendly and helpful. However, I am sure there are other excellent choices for this purpose.
- Badges
- Science topic
- Similar topics
- Correction