Spatial tuning of the positive and negative Poisson's ratio of metamaterials through gradient design

The design of negative Poisson's ratio materials has received much attention in recent years. However, there are few focus on tuning spatial negative and positive Poisson's ratio behavior, which has potential applications in engineering such as mimicking the mechanics behavior of natural tissue. In this study, a two-dimensional rectangular perforated material is used as an example, and the positive-negative Poisson's ratio (PPR-NPR) behavior is successfully obtained by gradient design. Deep learning combined with meta-heuristic algorithms is used to material design for targeted deformation behaviors, enabling precise control over material deformation, customization of constitutive models, and constrained single-objective material optimization. The gradient design is expected to be extended to different configurations to realize the tunable positive-negative Poisson's ratio of the materials, which will provide broader possibilities for the design and fabrication of new negative Poisson's ratio materials.