The identification of continuous, spatiotemporal systems
arXiv:chao-dyn/9907010 · doi:10.1103/PhysRevE.57.2820
Abstract
We present a method for the identification of continuous, spatiotemporal dynamics from experimental data. We use a model in the form of a partial differential equation and formulate an optimization problem for its estimation from data. The solution is found as a multivariate nonlinear regression problem using the ACE-algorithm. The procedure is successfully applied to data, obtained by simulation of the Swift-Hohenberg equation. There are no restrictions on the dimensionality of the investigated system, allowing for the analysis of high-dimensional chaotic as well as transient dynamics. The demands on the experimental data are discussed as well as the sensitivity of the method towards noise.
References in corpus (2)
Cited by in corpus (14)
- Sparse Identification of Nonlinear Dynamics for Rapid Model Recovery
- Parametric, nonparametric and parametric modelling of a chaotic circuit time series
- Inferring the dynamics of oscillatory systems using recurrent neural networks
- Parameter estimation in spatially extended systems: The Karhunen-Loeve and Galerkin multiple shooting approach
- Additive Nonparametric Reconstruction of Dynamical Systems from Time Series
- Iterative Symbolic Regression for Learning Transport Equations
- Stability selection enables robust learning of partial differential equations from limited noisy data
- Curriculum learning for data-driven modeling of dynamical systems
- Information-geometry of physics-informed statistical manifolds and its use in data assimilation
- Machine Discovery of Partial Differential Equations from Spatiotemporal Data
- Inferring nonlinear parabolic field equations from modulus data
- Estimation of Lyapunov spectra from space-time data
- Inferring the time-dependent complex Ginzburg-Landau equation from modulus data
- Hyperparameter Optimization in the Estimation of PDE and Delay-PDE models from data