Simultaneous Topology Optimization of Differentiable and Non‐Differentiable Objectives via Morphology Learning: Stiffness and Cell Growth on Scaffold

• DOI: 10.1002/aidi.202400015
• Publication Date: 2025-03-14T04:22:16Z
• Authors: Weiming Wang; Yanhao Hou; Renbo Su; Weiguang Wang; Charlie C. L. Wang
• Keywords: Topology optimization; Morphology

Topology optimization (TO) of microstructures plays a critical role in optimizing functional performance across diverse engineering applications. While metamaterials with enhanced mechanical properties—such as hyperelasticity, energy absorption, and thermal efficiency—are commonly designed using complex microstructural geometries multi‐physics simulations, achieving the simultaneous non‐differentiable objectives remains significant challenge. In this work, novel framework is proposed for TO differentiable via data‐driven morphology learning approach. The extracts shape patterns from curated dataset recognized their superior specific To showcase versatility approach, it applied to scaffolds bone tissue engineering, cell growth representative objective. By integrating learned into process, method generates that effectively balance stiffness biological performance, such proliferation. As case study, scaffold design improves by 29.69% 37.05% on day 7 33.30% 14 demonstrated. This approach highlights general applicability broad range challenges, beyond growth.