A scalable parallel finite element framework for growing geometries. Application to metal additive manufacturing
share
This work introduces an innovative parallel, fully-distributed finite element framework for growing geometries and its application to metal additive manufacturing. It is well-known that virtual part design and qualification in additive manufacturing requires highly-accurate multiscale and multiphysics analyses. Only high performance computing tools are able to handle such complexity in time frames compatible with time-to-market. However, efficiency, without loss of accuracy, has rarely been the focus in the numerical community. Here, in contrast, the framework is designed to adequately exploit the resources of high-end distributed-memory machines. It is grounded on three building blocks: (1) Hierarchical adaptive mesh refinement with octree-based meshes; (2) a parallel strategy to model the geometry growth; (3) the customization of a parallel iterative linear solver, which leverages the so-called balancing domain decomposition by constraints preconditioning approach for fast convergence and high parallel scalability. Computational experiments consider the part-scale thermal analysis of the printing process by powder-bed technologies. After verification against a 3D benchmark, a strong scaling analysis is carried out for a simulation of 48 layers printed in a cuboid. The cuboid is adaptively meshed to model a layer-by-layer metal deposition process and the average global problem size amounts to 10.3 million unknowns. An unprecedented scalability for problems with growing domains is achieved, with the capability of simulating the printing and recoat of a single layer in 8 seconds average on 3,072 processors. Hence, this framework contributes to take on higher complexity and/or accuracy, not only of part-scale simulations of metal or polymer additive manufacturing, but also in welding, sedimentation, atherosclerosis, or any other physical problem with growing-in-time geometries.
READ FULL TEXT
