Data-driven design and morphological analysis of conical six-fold origami structures

Abstract

In engineering design, origami structures are frequently characterized by their geometry-shifting behavior, mechanical properties, lightness, and appealing aesthetic qualities. In particular, conical origami structures have attracted increasing attention in recent years owing to their excellent energy absorption performance. However, designing conical structures with customizable geometry and folding morphology has remained a less explored research area. In this paper, an artificial intelligence (AI)-based framework is proposed for the customizable design of conical origami structures. To this end, first, the geometric relationships between the 2D crease patterns and 3D structures have been derived to determine the critical parameters. Subsequently, a data-driven method based on a deep neural network (DNN) is developed to train and predict the relationship among these critical parameters to facilitate the parametric design of conical origami structures. Importantly, although the DNN model is trained based on a limited set of data, error analyses demonstrate that the predicted crease patterns are in satisfactory agreement with analytical solutions, verified by physical models. Moreover, numerical simulations are performed to reveal the folding morphology and ensure the smooth deployment of the origami structures. Finally, the findings of this study indicate that conical origami structures are promising choices for further research and development in the area of lightweight energy-absorption structures.