New statistic techniques for structure evaluation of particle packing

An application of statistic techniques for structure evaluation of packing of equal or distributed spherical particles is presented. The packing of particles generated by computer simulation is cut by a series of equally separated parallel planes, and the area densities of the cross-sections of particles on the cutting planes are used to analyse the packing structure. It is shown that as the number of the cutting planes increases, the mean of the area densities approaches the packing density. Then a time series analysis technique is applied to examine the randomness of the packing. The homogeneity of the packing is evaluated by testing the hypothesis that the particles are uniformly distributed within the packing space. The isotropy of the packing is evaluated both at micro-level, the distribution of projections of centre-to-centre lines between touching particles, and at macro-level, the variances of area densities on the cutting planes perpendicular to different directions. As case studies, the above techniques are applied to evaluate the packing structures of both equal and distributed particles obtained by a collective rearrangement simulation model.