I’m beginning a MS program in Computational Science and Engineering. As part of the program I’m going to have to do a thesis. The problem I have in mind for the thesis is to study the gas flow through a reactor during the pump down from atmospheric pressure to a vacuum condition of a few millitorr. At vacuum conditions the continuum conditions of the NS equations will not apply to the fluid flow.
From what I’ve read the lattice Boltzmann method (LBM) seems like it might be suitable for modeling this system.
Now the questions.
First, is this a problem the LBM would be a good method for solving?
Second, what are the limits of the LBM in solving this type of problem?
Finally, how do the speed and computational requirements of the LBM compare to other methods?
It is somewhat risky to answer this question in general terms, because the only way to know for sure is to try it out. Lattice Boltzmann is probably one of the better approaches to use for this type of problems, because it is based on the Boltzmann equation which can model hydrodynamic flows as well as rarefied gases.
In my humble opinion, your problem seems however to be a very subtle one, because it depends on the physics of a gas in many different regimes. In the framework of a master’s thesis, I wonder if it wouldn’t be best to chose a numerical method for which you find local experts, in your research group or in your surrounding, who can coach you appropriately. I don’t think that you will find a textbook which tells you in clear terms how exactly to model your problem with lattice Boltzmann, and you will probably be depending on highly qualified advice, or get prepared for a long-term research activity.
