This study proposes a novel methodology to combine topology optimization and ply draping simulation to partition composite structures, improve structural performance, select materials, and enable more accurate representations of cost- and weight-efficient manufacturable designs. The proposed methodology is applied to a structure as a case study to verify that the methodology is effective. One design concept is created by subjecting the structure to a kinematic ply draping simulation to inform the partitioning of the structure, improve drapability and performance, and reduce structural defects. A second design concept is created that assumes that plies are draped over the entire structural geometry, forming an integral design. The two design concepts' topologies are subsequently optimized to specify ideal material and ply geometries to minimize mass and reduce costs. The results indicate that the partitioned structure has a 19% lower mass and 15% lower material costs than the integral design. The two designs produced with the new methodology are also compared against two control designs created to emulate previously published methodologies that have not incorporated ply draping simulations. This demonstrates that neglecting the effects of ply draping produces topology optimization solutions that under-predict the mass of a structure by 26% and costs by 38%.
Keywords: carbon fiber; composites; draping simulation; glass fiber; structural partitioning; topology optimization.