Stability analysis of chaotic systems from data
arXiv:2210.06167 · doi:10.1007/s11071-023-08285-1
Abstract
The prediction of the temporal dynamics of chaotic systems is challenging because infinitesimal perturbations grow exponentially. The analysis of the dynamics of infinitesimal perturbations is the subject of stability analysis. In stability analysis, we linearize the equations of the dynamical system around a reference point, and compute the properties of the tangent space (i.e., the Jacobian). The main goal of this paper is to propose a method that learns the Jacobian, thus, the stability properties, from observables (data). First, we propose the Echo State Network (ESN) with the Recycle Validation as a tool to accurately infer the chaotic dynamics from data. Second, we mathematically derive the Jacobian of the Echo State Network, which provides the evolution of infinitesimal perturbations. Third, we analyse the stability properties of the Jacobian inferred from the ESN, and compare them with the benchmark results obtained by linearizing the equations. The ESN correctly infers the nonlinear solution, and its tangent space with negligible numerical errors. In detail, (i) the long-term statistics of the chaotic state; (ii) the covariant Lyapunov vectors; (ii) the Lyapunov spectrum; (iii) the finite-time Lyapunov exponents; (iv) and the angles between the stable, neutral, and unstable splittings of the tangent space (the degree of hyperbolicity of the attractor). This work opens up new opportunities for the computation of stability properties of nonlinear systems from data, instead of equations.
References in corpus (32)
- SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
- Dynamical Ensembles in Nonequilibrium Statistical Mechanics
- A Gallavotti-Cohen Type Symmetry in the Large Deviation Functional for Stochastic Dynamics
- Deep learning to represent sub-grid processes in climate models
- Machine learning accelerated computational fluid dynamics
- Dynamical ensembles in stationary states
- Using Machine Learning to Replicate Chaotic Attractors and Calculate Lyapunov Exponents from Data
- Predictability: a way to characterize Complexity
- Attractor Reconstruction by Machine Learning
- Data-Driven Forecasting of High-Dimensional Chaotic Systems with Long Short-Term Memory Networks
- Characterizing dynamics with covariant Lyapunov vectors
- Extensive Chaos in the Lorenz-96 Model
- Mathematical and Physical Ideas for Climate Science
- Energy transport in anharmonic lattices close and far from equilibrium
- Hyperbolicity and the effective dimension of spatially-extended dissipative systems
- Covariant Lyapunov vectors
- Statistical and Dynamical Properties of Covariant Lyapunov Vectors in a Coupled Atmosphere-Ocean Model - Multiscale Effects, Geometric Degeneracy, and Error Dynamics
- Detecting chaos in particle accelerators through the frequency map analysis method
- Physics-Informed Echo State Networks
- Violation of hyperbolicity in a diffusive medium with local hyperbolic attractor
- Dynamical Landscape and Multistability of a Climate Model
- Stability, sensitivity and optimisation of chaotic acoustic oscillations
- Dynamical Analysis of Blocking Events: Spatial and Temporal Fluctuations of Covariant Lyapunov Vectors
- Gradient-free optimization of chaotic acoustics with reservoir computing
- Covariant Lyapunov Vectors of a Quasi-geostrophic Baroclinic Model: Analysis of Instabilities and Feedbacks
- Critical Transitions and Perturbation Growth Directions
- Effective models and predictability of chaotic multiscale systems via machine learning
- Lyapunov analysis of strange pseudohyperbolic attractors: angles between tangent subspaces, local volume expansion and contraction
- Efficient Cross-Validation of Echo State Networks
- Estimating covariant Lyapunov vectors from data
- Guidelines for data-driven approaches to study transitions in multiscale systems: the case of Lyapunov vectors
- Data-driven prediction and control of extreme events in a chaotic flow