Abstract

The current work presents the latest industrial simulation solutions and workflows applied for Additive Manufacturing (AM) process simulation and designing with a special focus on topology and sizing optimization.

The latest developments obtained by integrating adjoint sensitivities into a general non-linear FE-solver will be shown on AM applications for topology and non-parametric sizing optimization. It is demonstrated that realistic simulation including pre-loading of the assemble process, stiffness of the bolt connections and contacts can be included for obtaining realistic AM optimization results. Additionally, we consider an additive manufacturing constraint for topology optimization in the form of an overhanging constraint. Commonly, the present geometrical constraint is combined with various objective functions and constraints typically applied in industrial applications as minimizing mass with stiffness, strength and modal eigenfrequency requirements. Lately sizing optimization using the adjoint sensitivities for design responses of finite element results including modeling non-linearities for shell elements is implemented. To the best of our knowledge, this is the first work which shows results for non-linear sizing of shell thicknesses using adjoint sensitivities including simultaneously the three modeling non-linearities as large deformations, plasticity as material non-linearity and contact.

With continuous growth in the AM machines, faster processes and diverse materials, it is critical for companies investing in AM manufacturing to justify their investment by taking structural designs fast to production without doing a lot of expensive trial-and-error manufacturing adjustments. The produced AM parts needs to meet tolerance, performance and durability requirements. Thus, we present a highly customizable general simulation framework for a wide spectrum of AM processes based on a thermal-stress general purpose finite element code for eliminating trial-and-error manufacturing adjustments as well as ensuring that produced AM parts needs meet tolerance, performance and durability requirements. The present AM simulation framework allows for: arbitrary meshes of CAD representations; exact specification in time and space of machine tooling (e.g., powder addition, laser trajectories, dwell times, etc.) as would be used on an actual machine; precise tracking of the progressive raw material addition to each element in the mesh via complex geometric computations; precise integration of the moving energy sources (e.g., laser, electron beams, arc welds, high temperature polymer extrusion); automatic computation of the continuously evolving convection and radiation surfaces, and simulation of a wide spectrum of AM processes such as laser and electron beam powder bed fabrication, direct energy deposition, arc welding, polymer extrusion, ink jetting, etc. Several industrial applications and benchmarks from the aerospace, automotive and consumer goods industry will be shown.