Stochastic Simulations of Two-Dimensional Composite Packings

摘要: In recent years, dense packings of two- and three-dimensional objects have been studied intensely in the context computational physics material sciences. For example, computer simulations disordered solids usually employ a two-dimensional model which is based on hexagonal networks elastic rigid bonds or arrangements mixed soft hard disks, respectively. Both types bonds/disks are distributed randomly. Large systems equations to be solved at any simulation step for calculation local displacements particle velocities. The start from equidistant nodes network centers respectively, which, general, may not an equilibrium state. We suggest extension where first near-equilibrium packing randomly calculated. Then, we can compute displacement caused by external forces this initial elementary units. To end, propose stochastic impact incorporating computation states as well specific boundary conditions. Our methodology two-step approach consisting preprocessing stage, physical properties different particles analyzed numerical methods, second stage (annealing-based) exploits approximate formulas interactions. implemented two cooling schedules with expected serial run-timen? ln2nandn3/2? ln5/2n, reach forndisks. algorithms were parallelized 20-processor machine, sufficiently large number speedup close processors. parallel run-time computing about 212h 449 using schedule, 37 h 1068 schedule. performed calculating average regular, packings. underlying our implementations very similar used analysis granular composites involves disks. However, emphasis aspects rather than particular interactions, because substituting system interactions another one does affect significantly overall approach.