TY - JOUR
T1 - A multiscale topology optimisation framework for hollow spheres as cellular materials
AU - Christoff, Bruno G.
AU - Almeida, Humberto
AU - Cardoso, Eduardo Lenz
AU - Tita, Volnei
PY - 2023/6/1
Y1 - 2023/6/1
N2 - Cellular materials can be designed to achieve mechanical properties with low weight. Among them, hollow spheres have been investigated since modern manufacturing methods can accurately produce them. This work aims to develop an optimisation procedure to distribute hollow spheres to minimise the compliance of a structure through a multiscale approach. The model has two geometrical parameters: internal and external diameters, in which the asymptotic homogenisation method (AHM) is employed to predict the effective properties of the material. Equations relating the effective properties to geometrical parameters are obtained through the least square method. Then, topology optimisation is employed to minimise compliance constrained by an admissible relative density. The optimised structures are compared to homogeneous distributions of hollow spheres, and compliance reductions up to 79% are reached. The optimum distributions of hollow spheres are validated against the classical solid isotropic with material penalisation (SIMP) approach. Moreover, some structures are 3D-printed to show the feasibility and capabilities of the proposed approach. Experimental tests are conducted on the structures through additive manufacturing. It is shown that optimised structures outperform the ones with homogeneous distributions of hollow spheres with the same relative density.
AB - Cellular materials can be designed to achieve mechanical properties with low weight. Among them, hollow spheres have been investigated since modern manufacturing methods can accurately produce them. This work aims to develop an optimisation procedure to distribute hollow spheres to minimise the compliance of a structure through a multiscale approach. The model has two geometrical parameters: internal and external diameters, in which the asymptotic homogenisation method (AHM) is employed to predict the effective properties of the material. Equations relating the effective properties to geometrical parameters are obtained through the least square method. Then, topology optimisation is employed to minimise compliance constrained by an admissible relative density. The optimised structures are compared to homogeneous distributions of hollow spheres, and compliance reductions up to 79% are reached. The optimum distributions of hollow spheres are validated against the classical solid isotropic with material penalisation (SIMP) approach. Moreover, some structures are 3D-printed to show the feasibility and capabilities of the proposed approach. Experimental tests are conducted on the structures through additive manufacturing. It is shown that optimised structures outperform the ones with homogeneous distributions of hollow spheres with the same relative density.
UR - https://www.scopus.com/pages/publications/85151282130
U2 - 10.1016/j.engstruct.2023.115990
DO - 10.1016/j.engstruct.2023.115990
M3 - Article
SN - 0141-0296
VL - 284
JO - Engineering Structures
JF - Engineering Structures
M1 - 115990
ER -