Keywords: gyroid, mechanical metamaterials, topology optimization, finite element method, artificial neural network, adaptive materials.

This study presents a comprehensive approach for the topology optimization of gyroid-based mechanical metamaterials, aiming to develop intelligent, adaptive materials that respond effectively to mechanical loads. Experimental validation was performed by fabricating 3D-printed samples using carbon fibre-reinforced nylon via fused deposition modelling (FDM) technology. Key mechanical properties, including Young's modulus and Poisson's ratio, were determined to benchmark the performance of the metamaterials. A parametric numerical model was developed using the finite element method (FEM) and rigorously validated against experimental data. A substantial dataset capturing the mechanical behaviour under varied loading conditions was generated. This dataset served as the basis for training an artificial neural network (ANN), which facilitated rapid evaluation of design variations. To navigate the complex, the trained ANN was integrated with a genetic algorithm (GA) for topology optimization. The GA iteratively explored candidate solutions through selection, mutation, and recombination processes, specifically targeting parameters such as porosity, wall thickness, and effective density. The combination of ANN and GA provided an efficient framework for the intelligent design and optimization of gyroid-based mechanical metamaterials with tuneable properties. A significant contribution of this work is the demonstration of an integrated experimental–numerical–AI workflow for the intelligent design and topology optimization of mechanical metamaterials.

download the full-text of this paper (PDF, 2349 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description