Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Development of a lattice Boltzmann solver with adaptive grid refinement
The lattice Boltzmann method (LBM) is a state-of-the-art computational fluid dynamics (CFD) model used to simulate fluid flow based on the solution of the Boltzmann equation. LBM has considerable advantages in solving low Mach number flows as compared to conventional Navier-Stokes solvers, mainly due to the locality and explicitness of operations. This results in a huge potential for massive parallel computing on modern distributed memory machines.
Adaptive mesh refinement (AMR) is a commonly employed computational technique in CFD to enhance the ratio between efficiency and accuracy of simulations by dynamically adjusting the resolution of the computational grid based on some local indicators of the flow. By adaptively refining the grid in regions of interest, e.g. shocks in compressible flows or interfaces in multiphase flows, AMR can provide high resolution where it is most needed, while reducing computational effort and memory footprints in regions where coarser resolution is sufficient.
The group recently developed a parallel AMR solver based on a finite-volume discrete velocity Boltzmann method (FV-DVBM). The solver is strictly conservative and showed promising results for compressible flows with moderate Mach numbers and discontinuities. The advantages of AMR, however, are not restricted to the regime of compressible flows. Therefore, the group is currently developing an AMR framework for general purpose kinetic models, including LBM.
The goal of this project is to incorporate a standard lattice Boltzmann model for isothermal, low Mach number flows (D2Q9, possibly D3Q27) into the AMR framework and to validate it with (2D, possibly 3D) test cases. For a master’s thesis, the scope shall be extended to include work on boundary conditions, complex geometries, as well as models for compressible, turbulent, or multiphase flows, depending on discussed preferences.
Student_Project_AMR_for_LBM_short.pdf
