Create: Lattice

The main lattice is a cuboid bounding box with dimensions that automatically adjust during model setup to contain all user-imported solid geometry. By default, the main lattice grid spacing is uniformly spaced, as defined by the baseline resolution.

The options in this menu are refinements used to reduce the total lattice count and therefore the total memory requirements of a simulation. These reductions fall into two categories: Lattice Modifications and Lattice Coarse Graining.

Lattice Modifications change the lattice spacing and bounding box via local spatial refinement or local lattice culling. These objects are included as children to the main lattice parent. Refinement zones can increase the memory efficiency of the code, but are less computationally efficient than uniform grid spacing. Similarly, inclusion zones cull unwanted sections from the lattice domain, but make the solver less efficient. They should be employed judiciously.

Lattice Coarse Graining uses semi-empirical models to describe the net effects of unresolved (e.g., sub-grid) solid structures on the overall flow field. These are new parent objects added to the model tree.

Lattice Modifications

  • Refinement Zone: Refinement zones are regions where the space resolution of the lattice is locally increased relative to the main lattice and the associated baseline resolution. These zones can be defined as boxes, spheres, and cylinders, or they can contour user-defined solid bodies.

  • Inclusion Zone: Inclusion zones are regions of the domain where the solver will explicitly place lattice points. Anything outside an inclusion zone is not considered as part of the simulation, which reduces the number of lattice points stored in the system memory.

Lattice Coarse Graining

  • Fluid Screen: A fluid screen models the effects of a 2D screen on fluid flow, linking local fluid velocity to local pressure drop across the screen. This avoids the need to resolve a fine screen explicitly using high resolution simulations. The screens are assumed to be 2D and thinner than the grid spacing.

  • Porous Medium: This function implicitly models a packed structure and predicts the relationship between flow rate and pressure drop. Like the pressure screen, this avoids the need to resolve a fine solid structure using high resolution simulations. Beyond the fluid screen, the packed structure can be arbitrary shapes and large relative to the grid spacing.

  • Particle Sink: This function causes all particles entering into the geometry of the sink to be removed from the system.