Topology Optimization with Void Accessibility Constraints for Additive Manufacturing

Abstract

The innovation of additive manufacturing has allowed for the production of complex geometries not previously able to be manufactured easily or at all. While additive manufacturing was formerly thought of as only a rapid prototyping tool, recent improvements have increased the applications of additive manufacturing in a number of fields. One of these uses is manufacturing components designed using computational optimization methods like topology optimization. However, while additive manufacturing has less constraints than traditional manufacturing, it is not completely without constraints. In order to additively manufacture components designed with topology optimization, optimization results generally must be modified to remove voids from which support material and unused powder or resin can not be removed. These modifications cause a reduction in the optimized performance. This reduction in performance can be minimized by considering the void accessibility during the optimization itself. This work presents a novel methodology for void restriction in topology optimization considering void accessibility. The constraint is formulated in such a way that it can identify accessibility paths of varying widths with no restriction on finite element mesh type. Additionally, the constraint allows the designer to choose the number of accessibility directions used, giving them more control over the trade-off between increased design freedom and performance, and decreased computational efficiency and ease of manufacturing. The constraint is demonstrated on 2D and 3D academic test cases which demonstrate the effectiveness of the methodology. In all cases, fully accessible designs were generated with compliance increases up to only 55%.