Small, Год журнала: 2025, Номер unknown

Опубликована: Май 8, 2025

Abstract Multi‐target inverse design, which involves designing multiple targets with different optimization objectives, becomes a key focus in mechanical metamaterials (MMs) design. Specifically, many practical applications impose varying requirements for sections. For instance, the heel of sole demands to provide support, while arch should be comfortable, adequately supportive, and lightweight. However, existing approaches, such as topology optimization, typically on optimizing MMs specific e.g., high strength. Worse, these approaches are often inaccurate time‐consuming, even just single target. In this work, based graded triply periodic minimal surface (TPMS) architectures, machine‐learning‐powered approach is proposed rapid, accurate, multi‐target MM design by employing six‐parallel pipeline network architecture utilizing deep networks map structural parameters curves. The most suitable results selected target curves other performance requirements, can derived from parameters. achieves normalized root‐mean‐square error (NRMSE) 2.49% test dataset outputs corresponding within seconds, simultaneously meeting targets. Finally, an demonstrated soles various gait scenarios foot deformity treatments.

Язык: Английский