The Nonlinear Mechanical Constitutive Model and Deep Learning Method to Inverse Design Dual‐Feature‐Integrated Lattice Metamaterial

Abstract

The bio‐inspired lattice structure of a regularly fibrous organization is a kind of structural material with practical value in flexible bio‐integrated electronics. The recently proposed dual‐feature‐integrated lattice structure can accurately customize the nonlinear mechanical curve of biological issues. However, it is still lacking the constitutive models to inverse design the desirable mechanical properties. Herein, a nonlinear mechanical constitutive model for the dual‐feature‐integrated metamaterial is established by introducing the equilibrium equation and deformation coordination conditions. The experimental and numerical results show that the proposed constitutive model can predict accurately the stress–strain curves of dual‐feature‐integrated lattice structure. In addition, the machine learning‐whale optimization algorithm method is used to inverse design the dual‐feature‐integrated lattice structure, which can quickly find the target mechanical responses (chicken skin and human skin) in a board design space. The dual‐feature‐integrated mechanical metamaterial has a higher structural design option in comparison to the pure horseshoe and chiral metamaterial. The finding of this work contributes to the designs of lattice structures with flexibility and stretchable electronics.