Journal of Theoretical and Applied Vibration and Acoustics

Journal of Theoretical and Applied Vibration and Acoustics

Hybrid Koopman-neural network approach for robust parameter estimation and prediction in duffing oscillators

Document Type : Invited by Abdolreza Ohadi

Authors
Yassin Riyazi 1
Navidreza Ghanbari 2
Arash Bahrami 2
1 M.Sc. Student, School of Mechanical Engineering, College of Engineering, University of Tehran, Tehran, Iran
2 Assistant Professor, School of Mechanical Engineering, College of Engineering, University of Tehran, Tehran, Iran
10.22064/tava.2024.2029513.1241
Abstract
Nonlinear dynamical system research is essential in science and engineering because it could be potentially employed to represent real-world phenomena.  Traditional methods rely on pre-defined models or computationally expensive simulations, limiting their applicability to only numerical data. In the present research, without any prior knowledge of the system, we suggest a novel way to build a linear representation of the Duffing oscillator by fusing the capabilities of deep neural networks with the Koopman operator. This recently established methodology makes it easier to estimate system parameters effectively and accurately predict the oscillator's future behavior. Our approach incorporates a modified training procedure that restricts the Koopman operator to a single linear layer within the neural network, improving interpretability and potentially reducing training complexity. This methodology not only simplifies nonlinear system analysis but also paves the way for advancements in predictive modeling across various fields. Notably, our method yields distinctive eigenvalues of the Koopman generator matrix, enabling the Koopman operator to exhibit robustness against noise and capture a spectrum of Duffing equation behaviors. This includes the precise prediction of periodic oscillations and the capturing of period-doubling bifurcation, all while maintaining tractability within the neural network framework.

Highlights

  • An autoencoder, incorporating Koopman operator within a neural network is developed.
  • Duffing periodic responses and period-doubling bifurcations are captured.
  • A dual rescaling technique to address optimization challenges is implemented.
  • Strong resiliency to noise while maintaining accurate predictions is achieved.

Keywords
Koopman operator
parameter estimation
nonlinear dynamical systems
neural networks, period-doubling
Subjects
[1] B.O. Koopman, Hamiltonian systems and transformation in Hilbert space, Proceedings of the National Academy of Sciences, 17 (1931) 315-318.
[2] G.E. Hinton, R.R. Salakhutdinov, Reducing the dimensionality of data with neural networks, science, 313 (2006) 504-507.
[3] B. Lusch, J.N. Kutz, S.L. Brunton, Deep learning for universal linear embeddings of nonlinear dynamics, Nature communications, 9 (2018) 4950.
[4] K. Champion, B. Lusch, J.N. Kutz, S.L. Brunton, Data-driven discovery of coordinates and governing equations, Proceedings of the National Academy of Sciences, 116 (2019) 22445-22451.
[5] E. Kaiser, J.N. Kutz, S.L. Brunton, Sparse identification of nonlinear dynamics for model predictive control in the low-data limit, Proceedings of the Royal Society A, 474 (2018) 20180335.
[6] A. Krizhevsky, I. Sutskever, G.E. Hinton, ImageNet classification with deep convolutional neural networks, Communications of the ACM, 60 (2017) 84-90.
[7] K. Simonyan, A. Zisserman, Very deep convolutional networks for large-scale image recognition, arXiv preprint arXiv:1409.1556, (2014).
[8] K.He, X.Zhang, S.Ren,J.Sun,Deep residual Learning for image recognition, in: Proceeding of the IEEE conference on computer vision and pattern recognition,(2016)770-778.
[9] S.L. Brunton, M. Budišić, E. Kaiser, J.N. Kutz, Modern Koopman Theory for Dynamical Systems, SIAM Review, 64 (2022) 229-340.
[10] C.W. Rowley, I. Mezić, S. Bagheri, P. Schlatter, D.S. Henningson, Spectral analysis of nonlinear flows, Journal of fluid mechanics, 641 (2009) 115-127.
[11] P.J. Schmid, L. Li, M.P. Juniper, O. Pust, Applications of the dynamic mode decomposition, Theoretical and computational fluid dynamics, 25 (2011) 249-259.
[12] M.O. Williams, I.G. Kevrekidis, C.W. Rowley, A data–driven approximation of the koopman operator: Extending dynamic mode decomposition, Journal of Nonlinear Science, 25 (2015) 1307-1346.
[13] F. Moon, P.J. Holmes, A magnetoelastic strange attractor, Journal of Sound and Vibration, 65 (1979) 275-296.
[14] K.A. Kumar, S.F. Ali, A. Arockiarajan, Magneto-elastic oscillator: Modeling and analysis with nonlinear magnetic interaction, Journal of Sound and Vibration, 393 (2017) 265-284.
[15] G. Duffing, Ingenieur: Erzwungene Schwingungen bei veränderlicher Eigenfrequenz und ihre technische Bedeutung. Sammlung Vieweg. Heft 41/42, Braunschweig 1918. VI+ 134 S, Zamm‐zeitschrift Fur Angewandte Mathematik Und Mechanik, 1 (1918) 72.
[16] C. Runge, Über die numerische Auflösung von Differentialgleichungen, Mathematische Annalen, 46 (1895) 167-178.
[17] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, A. Rabinovich, Going deeper with convolutions, in:  Proceedings of the IEEE conference on computer vision and pattern recognition, 2015, pp. 1-9.
[18] Y. LeCun, L. Bottou, Y. Bengio, P. Haffner, Gradient-based learning applied to document recognition, Proceedings of the IEEE, 86 (1998) 2278-2324.
[19] J. Schmidhuber, S. Hochreiter, Long short-term memory, Neural Comput, 9 (1997) 1735-1780.
[20] Q. Li, F. Dietrich, E.M. Bollt, I.G. Kevrekidis, Extended dynamic mode decomposition with dictionary learning: A data-driven adaptive spectral decomposition of the Koopman operator, Chaos: An Interdisciplinary Journal of Nonlinear Science, 27 (2017).