Geometry-informed dynamic mode decomposition in origami dynamics

Shuaifeng Li,Yasuhiro Miyazawa,Koshiro Yamaguchi,Panayotis G.Kevrekidis,Jinkyu Yang

2023-03-08

Abstract:Origami structures often serve as the building block of mechanical systems due to their rich static and dynamic behaviors. Experimental observation and theoretical modeling of origami dynamics have been reported extensively, whereas the data-driven modeling of origami dynamics is still challenging due to the intrinsic nonlinearity of the system. In this study, we show how the dynamic mode decomposition (DMD) method can be enhanced by integrating geometry information of the origami structure to model origami dynamics in an efficient and accurate manner. In particular, an improved version of DMD with control, that we term geometry-informed dynamic mode decomposition~(giDMD), is developed and evaluated on the origami chain and dual Kresling origami structure to reveal the efficacy and interpretability. We show that giDMD can accurately predict the dynamics of an origami chain across frequencies, where the topological boundary state can be identified by the characteristics of giDMD. Moreover, the periodic intrawell motion can be accurately predicted in the dual origami structure. The type of dynamics in the dual origami structure can also be identified. The model learned by the giDMD also reveals the influential geometrical parameters in the origami dynamics, indicating the interpretability of this method. The accurate prediction of chaotic dynamics remains a challenge for the method. Nevertheless, we expect that the proposed giDMD approach will be helpful towards the prediction and identification of dynamics in complex origami structures, while paving the way to the application to a wider variety of lightweight and deployable structures.

Dynamical Systems

What problem does this paper attempt to address?

### Problems the Paper Attempts to Solve

This paper aims to address the challenges in dynamic modeling of origami structures. Specifically, it focuses on how to efficiently and accurately model the dynamic behavior of origami structures using data-driven methods. Due to the inherent nonlinear characteristics of origami systems, traditional experimental observation and theoretical modeling methods have limitations in dealing with such problems. Therefore, the authors propose an improved dynamic mode decomposition (DMD) method, called geometry-informed dynamic mode decomposition (giDMD), which integrates geometric information to enhance the accuracy and interpretability of the model.

### Main Contributions

1. **Method Innovation**: The giDMD method is proposed, which integrates geometric information to improve the accuracy and interpretability of DMD in modeling origami dynamics.
2. **Application Validation**: The effectiveness of giDMD is validated on a single origami unit, an origami chain, and a double Kresling origami structure, demonstrating its predictive capability at different frequencies.
3. **Topological Edge State Identification**: Successfully identified topological edge states in the origami chain and revealed the importance of geometric parameters in dynamics.
4. **Nonlinear Dynamics Analysis**: In the double Kresling origami structure, giDMD accurately predicts periodic motion and partially captures the characteristics of chaotic motion.

### Specific Content

1. **giDMD Method**:
- By introducing geometric information, an enhanced DMDc model is constructed.
- Geometric parameters are added to the state vector to form augmented states and control variables.
- Sparse regression methods are used to optimize matrix K, balancing accuracy and the number of control elements.

2. **Application in Origami Chain**:
- In the origami chain, giDMD accurately predicts axial and rotational displacements at different frequencies.
- Successfully identified topological edge states and demonstrated their characteristics in the spectrum.

3. **Application in Double Kresling Origami Structure**:
- In the double Kresling origami structure, giDMD accurately predicts periodic motion and partially captures the characteristics of chaotic motion.
- Further validated the identification of different dynamic states by analyzing the singular value spectrum of the K matrix.

### Conclusion

The giDMD method performs excellently in modeling origami dynamics, not only improving prediction accuracy but also enhancing model interpretability. Although there are still some challenges in dealing with chaotic dynamics, this method provides a new approach for dynamic prediction and identification of complex origami structures and is expected to be applied to the design and analysis of more lightweight and deployable structures.