Predicting multi-parametric dynamics of externally forced oscillator using reservoir computing and minimal data

TL;DR

This paper demonstrates that reservoir computing can learn and predict complex bifurcation behaviors of a forced oscillator from minimal data, including chaotic regimes, outperforming traditional methods in resource efficiency.

Contribution

It introduces a reservoir computing framework capable of predicting multi-parametric oscillator dynamics and bifurcations from limited training data, including unseen regimes.

Findings

RC accurately predicts bifurcations and chaos outside training data

Minimal training data suffices for complex dynamical predictions

Framework outperforms traditional resource-intensive methods

Abstract

Mechanical systems exhibit complex dynamical behavior from harmonic oscillations to chaotic motion. The dynamics undergo qualitative changes due to changes to internal system parameters like stiffness and changes to external forcing. Mapping out complete bifurcation diagrams numerically or experimentally is resource-consuming, or even infeasible. This study uses a data-driven approach to investigate how bifurcations can be learned from a few system response measurements. Particularly, the concept of reservoir computing (RC) is employed. As proof of concept, a minimal training dataset under the resource constraint problem of a Duffing oscillator with harmonic external forcing is provided as training data. Our results indicate that the RC not only learns to represent the system dynamics for the external forcing seen during training, but it also provides qualitatively accurate and robust system response predictions for completely unknown multi-parameter regimes outside the training data. Particularly, while being trained solely on regular period-2 cycle dynamics, the proposed framework correctly predicts higher-order periodic and even chaotic dynamics for out-of-distribution forcing signals.