A Contact Detection Algorithm for Three-Dimensional Discontinuous Deformation Method

An efficient and accurate three-dimensional (3-D) contact detection algorithm is of vital importance in large-scale computing of discontinuous deformation analysis. The contact detection algorithm takes up a large proportion of the whole calculation time because of the increasing calculation number in three-dimensional case. For block systems which consist of convex polyhedrons, the efficiency and accuracy of the contact detection algorithm is a great challenge. The presented contact detection algorithm in discontinuous deformation analysis (DDA) method is carried out in three phases: the global neighbour searching, the local contact type searching and the contact transferring. A contact detection algorithm for 3-D DDA is presented with some improvements on the three phases. A grid-based global neighbour searching algorithm is used to eliminate unnecessary neighbour blocks to the exact number. The local contact type searching includes two steps. In the first step, the common plane method is used to determine the contact type between convex polyhedrons. In the second step, the contact pairs detected by the common plane method are divided into four basic types. In the third phase, the four contact types are transferred into two basic contact types, vertex-to-face contact type and edge-to-edge contact type, which are managed in a uniform procedure. Three verification examples are calculated, the results show that the proposed algorithm based on the three contact detection phases is robust and efficient.