Collaborative optimization design of process parameter and structural topology for laser additive manufacturing

Abstract

High-resolution laser additive manufacturing (LAM) significantly releases design freedom, promoting the development of topology optimization (TO) and advancing structural design methods. In order to fully take advantage of voxelated forming methods and establish the quantitative relationship between the mechanical properties of printing components and multiple process factors (laser- and process- parameters), the concurrent optimization design method based on LAM should cover the process-performance relationship. This study proposes a novel artificial intelligence-facilitated TO method for LAM to concurrently design microscale material property and macroscale structural topology of 3D components by adopting heuristic and gradient-based algorithms. The process–structure–property relationship of selective laser sintering is established by the back propagation neural network, and it is integrated into the TO algorithm for providing a systematic design scheme of structural topology and process parameter. Compared with the classical optimization method, numerical examples show that this method is able to improve the mechanical performance of the macrostructure significantly. In addition, the collaborative design method is able to be widely applied for complex functional part design and optimization, as well as case studies on artificial intelligence-facilitated product evaluation.