Crashworthiness of conical shells is known to depend on various factors. This study sets out to determine the extent to which the cross-sectional diameter contributes to their energy-absorbing properties. The object of the study was thin-walled aluminium tubes varying in upper diameter and wall thickness. The components were subjected to dynamic axial crushing kinetic energy equal to 1700 J. The numerical analysis was performed using Abaqus 6.14 software. The specific aim of the study was to determine the extent to which variable wall thickness affects the energy absorption capacity of the components under study. From the simulations, we have managed to establish a relationship between total energy absorption capacity and wall thickness. The results from the conducted analyses and the purpose-specific neural networks could provide the base for the future methodology for forecasting and optimisation of energy-absorbing systems.
Keywords: artificial neural networks; crashworthiness; thin-walled structure.