Efficient elastic characterisation of various closed-cell composites via scaled boundary finite element method

This is the authors' abstract. We don't add key points for this paper.

One of the key difficulties in assessing lightweight and load-carrying composite structures, made of closed-cell foams, lies in the lack of efficient examination methods, induced by their irregular geometries with high cellular randomness. The present work innovatively adopts scaled boundary finite element method to address this problem by conducting efficient elastic characterisation on various closed-cell composite foams. These foam samples are generated via Voronoi decomposition and B-spline, where representative volume element models with periodic boundary conditions are constructed and discretised using quadtree meshing. The outcomes are focused on the elastic responses under extensional and shear deformations. A direct comparison between computational accuracy and time consumption of the developed framework and commercial software highlights its effectiveness and efficiency. The properties of closed-cell foams with diverse cellular morphologies are investigated through 7125 samples and discussed to quantify the impact of geometrical traits. Porosity clustering with non-uniform gradation patterns and high-density foams achieved by removing internal pores are proposed for performance tailoring. The quantitative evaluation on result divergence indicates a higher requirement on cell number for shear responses and high-density foams. This work provides a pathway for efficient large-data generation of closed-cell foam materials and novel designs of porous composite structures with tailored properties.