I’m new to the LBM and am currently trying to understand how to validate mesh convergence within this framework. My understanding is that convergence might depend on the number of particles in the fixed domain. However, in FluidX3D, it seems that a grid-based approach is used if the particle extension option remains commented out. Could someone clarify how LBM simulations can be performed without explicitly representing particles?
If mesh convergence depends on the computational grid resolution, does each pixel effectively act as a single point in the domain? If so, would increasing the structure size result in finer resolution near its surface? Additionally, by keeping the dimensionless domain size (Lx/D, Ly/D, Lz/D) constant, could I achieve finer mesh resolution and validate mesh convergence?
Any insights or recommendations would be greatly appreciated!
Hi all,
I’m new to the LBM and am currently trying to understand how to validate mesh convergence within this framework. My understanding is that convergence might depend on the number of particles in the fixed domain. However, in FluidX3D, it seems that a grid-based approach is used if the particle extension option remains commented out. Could someone clarify how LBM simulations can be performed without explicitly representing particles?
If mesh convergence depends on the computational grid resolution, does each pixel effectively act as a single point in the domain? If so, would increasing the structure size result in finer resolution near its surface? Additionally, by keeping the dimensionless domain size (Lx/D, Ly/D, Lz/D) constant, could I achieve finer mesh resolution and validate mesh convergence?
Any insights or recommendations would be greatly appreciated!
Best Regards, Hongcao