Choose a model and generate a realization. It is kept in the workspace, from which it can be sent to the simulation, the export or the optimization.
Generate a morphology, then simulate it here: its apparent property by finite elements on a conforming mesh of its level set (MMG), or by FFT on its voxels, with its fields. The result is kept in the workspace.
Files of a morphology or a simulation of the workspace, with an archive of them all: figures, videos (the colours of a field following its load), tables, data and the configuration to rerun a simulation with morphogen run.
Densities on a grid, from a void ball, noise, or a morphology of the workspace (the fraction of its solid phase in each voxel; from a simulation, its morphology and solid material). Planar grids (nz = 1) take the middle section of a 3D morphology; the symmetry chosen averages the start. A projection of 32 makes stiff designs black and white; auxetic designs need a softer one (8) and a filter radius of 2 to 3 voxels.
Inverse homogenization with the FFT solver: the cell evolves live. The design is kept in the workspace as a morphology (its level set: the densities about 1/2), to simulate on a conforming mesh or to export.
Combinatorial metamaterials: a box of tiles whose faces must match those of their neighbours (polarized blocks, after Coulais, Teomy, de Reus, Shokef & van Hecke, Nature 2016; or pipes, networks without dead ends). The exact number of compatible assemblies is counted, and one assembly is drawn by wave function collapse. To simulate such structures, generate them with the model tile_assembly.
Exact number of compatible assemblies, and one drawn by wave function collapse.
Normalized charts: a relative property (P / Ps) against the relative density; n = 1 is the rule of mixtures (stretching-dominated lattices), n = 2 bending-dominated foams. Three families in colour; the others are grouped.
Read a database of results, then draw its records.