Machine learning assisted design and optimization of plate-lattice structures with superior specific recovery force

Abstract

In load carrying structures and devices, there is a growing need for shape memory polymer (SMP) metamaterials that are lightweight and have superior strength, remarkable flexibility, and substantial specific recovery force (SFR). One of the challenges is to find optimum lightweight structures with high SFR. To address this challenge, we propose a novel inverse design framework to design plate-lattice structures (PLSs) with user-defined optimum specific maximum compression strength. Consisting of three sub-frameworks, the performance of the inverse design framework was validated before it was utilized to optimize PLSs. The optimum PLSs developed are fabricated with 3D printing using a novel SMP. In addition, we have printed a solid cylinder and Cubic+Octet (control) PLSs to compare their structural capacity with the predicted structures. The optimized PLSs display 30 ∼ 170 % greater SFR compared to the control PLS and solid cylinder. These findings suggest a promising strategy for enhancing the effectiveness of actuators based on SMP mechanical metamaterials. The inverse design framework has the potential to be utilized for generating structures with user-defined optimum mechanical properties.

Keywords

• Metamaterials
• Plate-lattice structures
• Conditional generative adversarial networks
• Machine learning
• Shape memory polymer