A review on artificial intelligence-enabled mechanical analysis of 3D printed and FEM-modelled auxetic metamaterials
Aman Garg, Anshu Sharma, Weiguang Zheng, Li Li
Abstract
Auxetic structures, characterised by a negative Poisson’s ratio, exhibit the unique property of expanding in all directions under tensile loading and contracting in all directions when compressed. The behaviour of these structures is highly influenced by the arrangement of the material within the unit cell and the interactions between these cells. Due to the complexity of their behaviour, understanding and modelling auxetic materials pose significant challenges, requiring extensive experimentation and sophisticated modelling frameworks. As a result, machine learning (ML) techniques have emerged as a powerful tool for predicting the mechanical properties and studying the behaviour of auxetic metamaterials under various loading conditions. This review provides a comprehensive analysis of the use of ML algorithms in predicting the properties and performance of auxetic metamaterials. It discusses the details of the datasets, the algorithms employed, the 3D printing technologies used for sample fabrication, and the geometries of the auxetic structures explored in the literature. In addition to 3D-printed structures, this review also includes findings from finite element-based models. Finally, the paper outlines future research directions, highlighting opportunities to advance the understanding and application of auxetic metamaterials using machine learning techniques.