Robust topology optimization for thermoelastic hierarchical structures with hybrid uncertainty

• Published in: Journal of thermal stresses Vol. 44; no. 12; pp. 1458 - 1478
• Main Authors: Jiang, H. X., Wei, B. L., Zhou, E. L., Wu, Yi, Li, X. K.
• Format: Journal Article
• Language: English
• Published: Philadelphia Taylor & Francis 07.12.2021; Taylor & Francis Ltd
• Subjects: BESO; Cost analysis; hierarchical structure; hybrid uncertainty; Optimization; perturbation analysis; Perturbation methods; robust topology optimization; Robustness (mathematics); Structural hierarchy; Thermal analysis; thermoelastic structure; Topology optimization; Uncertainty
• ISSN: 0149-5739; 1521-074X
• DOI: 10.1080/01495739.2021.1999877

This paper presents a robust topology optimization (RTO) framework for thermoelastic hierarchical structures with hybrid uncertainty. Firstly, the thermoelastic concurrent optimization model is established and the uncertainties with interval random parameters are integrated into the thermoelastic hierarchical structure. Then, a reliable and cost-effective hybrid uncertainty perturbation analysis method (HUPAM) is derived for a quick estimate of the robust objective function subject to the mechanical and thermal loads. Finally, by calculating the design variables sensitivities of macroscale and microscale, the robust topological design can be generated efficiently. To obtain clear and optimal topologies for both macro- and micro- structures, the bi-directional evolutionary structural optimization (BESO) method is adopted. Some 2D and 3D numerical examples are presented to demonstrate the influences of the hybrid uncertainties on the final designs. The results also show that the proposed method can effectively improve the thermoelastic structural performance when it comes to uncertainties.