3D Anisotropic Tunable Mechanical Metamaterials

ABSTRACT On‐demand tunable mechanical performance after fabrication is crucial for intelligent and adaptive materials and structures. Yet toward the requirements in realistic scenarios involving multi‐axial loading, the anisotropic tunable mechanical responses remain largely unexplored. Here, we propose a 3D anisotropic tunable mechanical metamaterial that enables tunable triaxial mechanical performance through a local stiffening‐sensitivity mechanism. By selectively applying stiffening points within the metamaterial, the triaxial macroscopic moduli can be adjusted over a wide range without altering the constitutive material or the overall geometry. The uniaxial modulus can be increased by approximately 0.9 times the initial state modulus for each additional stiffening point (2% of available points), and the anisotropy degree can be tuned over a range from 1 up to 35.5. Finite element analysis results are employed with a convolutional neural network to establish the structure‐performance relationships of the metamaterial, and a genetic algorithm is used to achieve sequential inverse design of anisotropic mechanical behavior, including uniaxial, biaxial and triaxial stress–strain curves. The proposed approach provides an effective solution for tuning multiaxial mechanical behaviors of metamaterials and is expected to promote their potential applications in robots, adaptive structures, and systems requiring direction‐dependent stiffness.