Page 202 - The Combined Finite-Discrete Element Method
P. 202

DYNAMICS OF IRREGULAR DISCRETE ELEMENTS SUBJECT        185

           5.2 DYNAMICS OF IRREGULAR DISCRETE ELEMENTS SUBJECT TO
                 FINITE ROTATIONS IN 3D


           Very often a combined finite-discrete element system also comprises rigid discrete ele-
           ments. In such cases, no forces due to deformation are present, while discretisation of
           rigid discrete elements is necessary only for a description of the geometry of discrete ele-
           ments and processing of contact interaction. In these types of problems, once the contact
           forces and external loads are known, the governing equations can be integrated.
             For systems comprising rigid bodies in 2D, in general this is achieved through solving
           equations for translation and rotation about the centre of mass, i.e. assigning to each
           discrete element three degrees of freedom – two translations in the direction of the coor-
           dinate axes and one rotation about the z-axis. Thus, in 2D problems the central difference
           time integration scheme as described in the previous section is directly applicable.
             In 3D problems this situation is complicated by the presence of finite rotations about the
           centre of mass of the discrete element. The problem with simply extending 2D algorithms
           into 3D is that the angular velocity describing such rotation in general does not coincide
           with the principal axes of the discrete element. Simple extension of 2D algorithms into 3D
           would therefore not work. For instance, in 3D problems a description of spatial orientation
           is not a trivial task.



           5.2.1   Frames of reference

           To describe the motion of a particular rigid discrete element, two reference frames are
           introduced. The first is an inertial frame which does not move with the discrete element,
           and in the following text it is referred to as the ‘inertial frame’ or ‘fixed frame’. The ori-
           entation of the inertial frame is defined by a triad of unit vectors parallel to the respective
           axes of a Cartesian coordinate system, (Figure 5.3).

                                              ˜ ˜ ˜
                                             (i, j, k)                          (5.36)
           The second frame is fixed to the discrete element. The origin of this frame coincides with
           the centre of mass of the discrete element. This frame is assumed to move (translate and



                                                           j

                                                                        i
                                ~
                                j
                                                r

                                        ~
                                         i                 k
                        ~
                        k

                             Figure 5.3 Moving and fixed frames of reference.
   197   198   199   200   201   202   203   204   205   206   207