Developing accurate yet fast computational tools to simulate complex physical phenomena is a long-standing problem. Recent advances in machine learning have revolutionized the way simulations are approached, shifting from a purely physics- to AI-based paradigm. Although impressive achievements have been reached, efficiently predicting complex physical phenomena in materials and structures remains a challenge. Here, we present an AI-based general framework, implemented through graph neural networks, able to learn complex mechanical behavior of materials from a few hundreds data. Harnessing the natural mesh-to-graph mapping, our deep learning model predicts deformation, stress, and strain fields in various material systems, like fiber and stratified composites, and lattice metamaterials. The model can capture complex nonlinear phenomena, from plasticity to buckling instability, seemingly learning physical relationships between the predicted physical fields. Owing to its flexibility, this graph-based framework aims at connecting materials’ microstructure, base materials’ properties, and boundary conditions to a physical response, opening new avenues towards graph-AI-based surrogate modeling.

Manipulating the architecture of materials to achieve optimal combinations of properties (inverse design) has always been the dream of materials scientists and engineers. Lattices represent an efficient way to obtain lightweight yet strong materials, providing a high degree of tailorability. Despite massive research has been done on lattice architectures, the inverse design problem of complex phenomena (such as structural instability) has remained elusive. Via deep neural network and genetic algorithm, we provide a machine-learning-based approach to inverse-design non-uniformly assembled lattices. Combining basic building blocks, our approach allows us to independently control the geometry and topology of periodic and aperiodic structures. As an example, we inverse-design lattice architectures with superior buckling performance, outperforming traditional reinforced grid-like and bio-inspired lattices by ~30–90% and 10–30%, respectively. Our results provide insights into the buckling behavior of beam-based lattices, opening an avenue for possible applications in modern structures and infrastructures.

3D-printing technology is opening up new possibilities for the co-printing of sensory elements. While quasi-static research has shown promise, the dynamic performance has yet to be researched. This study researched smart 3D structures with embedded and printed sensory elements. The embedded strain sensor was based on the conductive PLA (Polylactic Acid) material. The research was focused on dynamic measurements of the strain and considered the theoretical background of the piezoresistivity of conductive PLA materials, the temperature effects, the nonlinearities, the dynamic range, the electromagnetic sensitivity and the frequency range. A quasi-static calibration used in the dynamic measurements was proposed. It was shown that the temperature effects were negligible, the sensory element was linear as long as the structure had a linear response, the dynamic range started at ∼ 30 $\mathsf{\mu }\mathsf{ϵ}$ and broadband performance was in the range of few kHz (depending on the size of the printed sensor). The promising results support future applications of smart 3D-printed systems with embedded sensory elements being used for dynamic measurements in areas where currently piezo-crystal-based sensors are used.