Search

Inverse Optimization of Mechanical Metamaterials Based on the GA-SA Model

Author(s)

Binglan Lin1,*, Yao Feng1, Xiaoxin Yun1, Xueying Wang2

Affiliation(s)

1Yisheng Innovation and Entrepreneurship Education Base, North China University of Science and Technology, Tangshan, Hebei, China 2School of Emergency Management and Safety Engineering, North China University of Science and Technology, Tangshan, Hebei, China *Corresponding Author

Abstract

To address the challenges of high-dimensional non-convex optimization and computational cost bottlenecks encountered in the inverse design of mechanical metamaterials, this paper proposes a GA-SA hybrid inverse optimization framework driven by both data and algorithms. This framework couples the global search capabilities of genetic algorithms with the local exploration capabilities of simulated annealing, whilst introducing surrogate models to replace traditional high-fidelity finite element analysis. Validation results demonstrate that this method effectively overcomes local optima traps. Whilst strictly controlling the approximation error of macroscopic mechanical responses within 1.5% thereby providing an efficient theoretical foundation and technical pathway for the large-scale customised design of microstructures.

Keywords

Mechanical Metamaterials; Inverse Optimization; GA-SA Hybrid Algorithm; Surrogate Models

References

[1] Barhemat R, Bao Y. Multi-objective optimization of mechanical metamaterials for enhanced mechanical strength, energy absorption, and negative Poisson’s ratio [J]. Structural and Multidisciplinary Optimization, 2026, 69(4):89-89. DOI:10.1007/S00158-026-04297-8. [2] Dilgen B C, Dilgen B S, Aage N, et al. Topology optimization of acoustic mechanical interaction problems: a comparative review [J]. Structural and Multidisciplinary Optimization, 2019, 60(2):779-801. DOI:10.1007/s00158-019-02236-4. [3] Cao X, Zhao Z, Wang P, et al. Inverse Design of Highly Deformable Mechanical Metamaterial Based on Partitional Semi-Random Optimization. [J]. Advanced science (Weinheim, Baden-Wurttemberg, Germany), 2025, 12(28):e2415935. DOI:10.1002/ADVS. 202415935. [4] Goudarzi S, Hassan H W, Anisi H M, et al. Comparison between hybridized algorithm of GA–SA and ABC, GA, DE and PSO for vertical-handover in heterogeneous wireless networks [J]. Sādhanā, 2016, 41(7):727-753. DOI:10.1007/s12046-016-0509-4. [5] Doksoo L, Wayne W C, Liwei W, et al. Data-Driven Design for Metamaterials and Multiscale Systems: A Review. [J]. Advanced materials (Deerfield Beach, Fla.), 2023, 36(8):e2305254-e2305254. DOI:10.1002/ADMA. 202305254.