Framework for global stability analysis of dynamical systems
arXiv:2304.12786 · doi:10.1063/5.0159675
Abstract
Dynamical systems, that are used to model power grids, the brain, and other physical systems, can exhibit coexisting stable states known as attractors. A powerful tool to understand such systems, as well as to better predict when they may ``tip'' from one stable state to the other, is global stability analysis. It involves identifying the initial conditions that converge to each attractor, known as the basins of attraction, measuring the relative volume of these basins in state space, and quantifying how these fractions change as a system parameter evolves. By improving existing approaches, we present a comprehensive framework that allows for global stability analysis on any dynamical system. Notably, our framework enables the analysis to be made efficiently and conveniently over a parameter range. As such, it becomes an essential complement to traditional continuation techniques, that only allow for linear stability analysis. We demonstrate the effectiveness of our approach on a variety of models, including climate, power grids, ecosystems, and more. Our framework is available as simple-to-use open-source code as part of the DynamicalSystems.jl library.
References in corpus (8)
- Tipping points in open systems: bifurcation, noise-induced and rate-dependent examples in the climate system
- Short-term synaptic plasticity in the deterministic Tsodyks-Markram model leads to unpredictable network dynamics
- Potentials and Limits to Basin Stability Estimation
- Effortless estimation of basins of attraction
- Multistability and Variations in Basin of Attraction in Power-grid Systems
- Homoclinic puzzles and chaos in a nonlinear laser model
- Monte Carlo Basin Bifurcation Analysis
- Analysis of a bistable climate toy model with physics-based machine learning methods