Topology optimization of coated structures with layer-wise graded lattice infill for maximizing the fundamental eigenfrequency. (15th October 2022)

• Record Type: Journal Article
• Authors: Hu, Tiannan; Wang, Yaguang; Zhang, Heng; Li, Hao; Ding, Xiaohong; Izui, Kazuhiro; Nishiwaki, Shinji
• Is Part Of: Computers & structures. Volume 271(2022)
• Publication Date: 2022-10-15
• Subjects: Multiscale topology optimization -- Coated structures -- Layer-wise graded lattice infill -- Velocity field-based level set method -- Fundamental eigenfrequency; Structural engineering -- Data processing -- Periodicals; Electronic data processing -- Structures, Theory of -- Periodicals; 624.171
• DOI: 10.1016/j.compstruc.2022.106861
• Languages: English
• ISSNs: 0045-7949

Abstract

Highlights: Coated structures with layer-wise graded lattice microstructures optimized using the proposed method show higher fundamental eigenfrequency than those with periodic and uniform lattice infill. Optimized microstructures are well-connectable between neighboring layers and the generated designs can be built out using the additive manufacturing (AM) technique. Owing to the general gradient-based algorithms, the proposed approach can handle multiple constraints efficiently. The modal assurance criterion (MAC) method in the proposed framework is able to treat the fundamental eigenfrequency and its eigenvector more accurately, which leads to great iteration convergence. Abstract: Combining the advantages of lattice infill and coating, coated structures with lattice infill are widely used in engineering to achieve light-weight structural characteristic or certain functionalities. This paper presents a novel concurrent multiscale topology optimization (TO) framework to maximize the fundamental eigenfrequency of structures with a uniform outer coating and layer-wise graded lattice infill microstructures. The macroscale topology optimization is conducted by using the velocity field-based level set method, which inherits the implicit geometrical representation and signed distance property of the conventional level set method (smooth and clear boundaries and well-maintenance of the uniform thickness of the coating). Besides, the employment of general mathematical programming algorithms enables the velocity field-based level set method to handle multiple constraints in an easier way. At microscale, the popular density-based method is used to design layer-wise graded lattice structures. The effective material properties of microstructures are computed by using the asymptotic homogenization method, which bridges macroscale and microscale designs. With higher design flexibility, it is found that coated structures with layer-wise graded lattice infill have higher eigenfrequencies than those with periodic and uniform lattice infill microstructures. The influence of layers of lattice infill and coating thickness on the concurrent optimization of macrostructures and microstructures are carefully studied by showcasing several numerical examples, and the effectiveness of the proposed method is also confirmed, as well as the manufacturability of optimization results.