This paper presents a density-based topology optimization method for designing 3D thin-walled structures with adaptive meshing. Uniform wall thickness is achieved by simultaneously constraining the minimum and maximum feature sizes using Helmholtz partial differential equations (PDE). The PDE-based constraints do not require information about neighbor cells and therefore can readily be integrated with an adaptive meshing scheme. This effectively enables the 3D topology optimization of thin-walled structures with a desktop PC, by significantly reducing computation in large void regions that appear during optimization. The uniform feature size constraint, when applied to 3D structures, can produce thin-walled geometries with branches and holes, which have previously been difficult to obtain via topology optimization. The resulting thin-walled structures can provide valuable insights for designing thin-walled lightweight structures made of stamping, investment casting and composite manufacturing.
@article{arxiv.1908.10825,
title = {Topology optimization for 3D thin-walled structures with adaptive meshing},
author = {Yuqing Zhou and Tsuyoshi Nomura and Ercan M. Dede and Kazuhiro Saitou},
journal= {arXiv preprint arXiv:1908.10825},
year = {2019}
}