Level-set based topology optimization considering milling directions via fictitious physical model

Abstract

Topology optimization is the most flexible structural optimization method that allows the topological modification as well as the shape changes. Topology optimization techniques have recently been utilized in engineering applications regarding multi-physics or multi-disciplinary optimization. The manufacturing evaluation is an important factor for the practical application to the field of industry. For example, conventional manufacturing processes such as molding and casting, require fulfillment of specific geometric conditions due to the path of the tools. To obtain the optimal structures satisfying such conditions, a new level set-based topology optimization method, based on the convection-diffusion equations with fictions fluxes, are utilized to impose the geometric constrains regarding manufacturability, has been developed. In this method, handling geometric constraints for the manufacturability depends on the fictitious heat fluxes. The propagation of the fictitious fluxes distinguishes the non-manufacturable domain. However, in previous approaches, the manufacturing directions are regarded as design parameters where the values are fixed before the optimization procedure. As the non-manufacturable domain is dominated by the path of the manufacturing tools, the appropriate selection of these parameter values is necessary to obtain the optimal design with the best manufacturability constraints. Thus, these parameters should be appropriately set during the optimization process. In this method, we develop a new topology optimization method where the manufacturing directions are also considered as design variables. A sensitivity analysis based on the adjoint equation is introduced to obtain optimized manufacturing directions. Several engineering applications are introduced and optimized through the new method and the previous methods to confirm the availability of the proposed method.