Space Decomposition Based Parallelisation Solutions for the Combined Finite-Discrete Element Method in 2D.

Abstract

The Combined Finite-Discrete Element Method (FDEM), originally invented by Munjiza, has become a tool of choice for problems of discontinua, where particles are deformable and can fracture or fragment. The downside of FDEM is that it is CPU intensive and, as a consequence, it is difficult to analyse large scale problems on sequential CPU hardware and parallelisation becomes necessary. In this work a novel approach for parallelisation of the combined finite-discrete element method (FDEM) in 2D aimed at clusters and desktop computers is developed. Dynamic domain decomposition-based parallelisation solvers covering all aspects of FDEM have been developed. These have been implemented into the open source Y2D software package by using a Message-Passing Interface (MPI) and have been tested on a PC cluster. The overall performance and scalability of the parallel code has been studied using numerical examples. The state of the art, the proposed solvers and the test results are described in the thesis in detail. 3