Topology optimization of additively manufactured components using artificial intelligence: A method for the integration of composite lattice structures

Powder bed fusion of metals using a laser beam enables the fabrication of complex geometries for high-performance applications in the machine tool industry. In particular, additively manufactured composite lattice structures combine the high stiffness of metallic lattices with the superior damping properties of polymeric filler materials. By tailoring the volume fractions, these structures offer promising potential for dynamically loaded components, such as milling tools. However, determining an optimal material distribution remains challenging, as it typically requires iterative and computationally expensive topology optimization (TO) algorithms. The homogenization method enables the assignment of the material properties of composite lattice structures to the material properties of an element according to its relative density. This supports the integration of the composite lattice structures into TO methods and reduces the computational effort. On this basis, artificial intelligence models for image generation can speed up the TO process. In this work, a methodology for an efficient design of tools with an optimal distribution of composite lattice structures under dynamic loads was derived. For this purpose, two different interpolation functions for the integration of homogenized composite lattice structures in a dynamic TO were calculated. The material curves were used to generate data sets for the training of the conditional generative adversarial network (cGAN) model. The model parameters were tuned, and various performance metrics were compared. The cGAN model reached a mean squared error of 0.0020 after the training. For manufacturing, a design option for transferring the two-dimensional structures into three dimensions was tested and discussed.