Mechanical behaviour and energy absorption performance of modified anti-tri chiral novel auxetic structures through experimental and numerical analysis subjected to surrogate metamodeling

• Three novel auxetic structures based on anti-tri chiral lattice were proposed. • Experimental, numerical, and metamodeling investigation of EA. • Superior EA and NPR of MATC1 structure was observed. • Robustness of RF in predicting experiment outcomes was demonstrated. Considering the advanced properties of meta-structures in absorbing energy and high crashworthiness performance, in this work three novel auxetic-lattice structures are proposed and developed based on anti-tri chiral structure by applying the element transformation approach. The novel auxetics were made of polylactic acid plus (PLA+) and additively manufactured by fused deposition modelling. The presented designs were numerically simulated considering elastic-plastic isotropic material properties, using C3D8R brick elements with second-order accuracy and by implementing mesh size 0.6 mm based on sensitivity analysis through Abaqus/Explicit 2020. The results were further validated by experimental investigation under quasi-static condition subjected to compression load. In addition, a hybrid surrogate metamodel using random forest (RF) and multilayer perceptron (MLP) deep learning has been developed and validated against experimental results. All three novel structures demonstrated auxetic behaviour by achieving negative Poisson’s ratio (NPR). The best auxetic characteristics were achieved by the first modified anti-tri chiral (MATC1) structure, with a considerable NPR of -0.42. Furthermore, the numerical simulation results aligned with experimental results and metamodeling prediction has been accomplished. Yet, slightly higher prediction performance was confirmed by RF model over MLP model.