Structural topology optimization for additive manufacturing with free choice of self-supporting and infill-supporting structures

In the realm of non-powder printing additive manufacturing, the removal of supports from enclosed voids has long posed a formidable challenge. In response to this challenge, inspired by the traditional approach of designing either self-supporting or infill-supporting structures, this paper presents a novel topology optimization method that allows for the free choice between the two types of support within same enclosed voids. The infill material not only serves as a support during the additive manufacturing process but also becomes an integral component of the final design, never requiring to be removed. Firstly, a bi-directional nonlinear virtual temperature method (BN-VTM) is introduced to identify enclosed voids. The BN-VTM adds virtual heating material in regions below a predefined temperature threshold and heat-absorbing material in regions exceeding the threshold. This strategy ensures that the maximum temperature within enclosed voids adheres strictly to the specified range, enhancing the precision of enclosed void identification throughout the iterative process. On this basis, by combining the filtering/projection method, an interpolation function for enclosed void structures that encompasses both self-supporting and infill-supporting structures is developed. This function also is applied to identify overhanging interfaces within enclosed voids. Finally, by constraining the overhang angles, the proposed topology optimization method automatically determines whether to design self-supporting structures to achieve steeper boundaries or internal infill support structures within enclosed voids. The problem of minimizing compliance under material volume fraction constraints is investigated, and sensitivity analysis is derived. Several numerical and printing examples illustrate the effectiveness of the proposed method.