| The prediction of the packing density for a granular mixture is a problem of relevance to cement concrete. It is also relevant to many other industrial sectors (e.g., compacting pills for pharmaceutical industry, packing and storing seeds for agriculture, deposition of natural sand, etc.) Many composite materials like concrete are made of granular material embedded in a binding matrix. The aim is often to combine grains in order to minimize the porosity, which allows the use of the least amount of binder.; This research included the development of three-dimensional computational simulation method for random particle packing as first phase. The second phase applied the developed model on concrete aggregates.; The developed computational method consists of two steps of simulation: kinematics simulation and dynamics simulation. The kinematics simulation uses a 3-D cellular model for particle deposition. The dynamics simulation models a matrix of particles in a three-dimensional space using the Discrete Element Method (DEM). The numerical simulation models the packing process of particles at a microscopic level including every particle in the matrix. Many details omitted in other models can now be included to increase the fidelity, allowing one to consider random sizes and locations of the aggregates. Application examples show that this approach works effectively, compared with experimental results. |
