A Physics-Informed Neural Network for Potential Energy Prediction and Inverse Design of Origami Structures

Origami structures have the advantages of foldability and adjustability, with applications spanning numerous engineering fields. However, there remains a dearth of intelligent and convenient methods that can effectively tackle both potential energy prediction and design problems on origami structures. This study proposes a novel physics-informed neural network (PINN) for predicting potential energy and performing design of origami structures. A sorting operation is developed for the PINN to address the challenge of the model converging to local optima. Given the boundness of the design variables, constraints on them are enforced during the design process. Two loss functions with physical connotation are customized for prediction and design problems, respectively. The accuracy of the potential energy curves predicted by the PINN with the sorting operation is demonstrated through comparison with a reference and the exhaustive method. Furthermore, two design cases for Kresling origami structures, matching a target potential energy curve and a set of target potential energy points, are performed to show the applicability of the model in inverse problems. The presented physics-driven approach without labelled data offers an innovative tool with learning ability to predict and design origami structures. In addition, the code of the PINN is shared online.