Modern Koopman theory for dynamical systems

DOI10.1137/21M1401243zbMATH Open1497.37105arXiv2102.12086OpenAlexW3129637124MaRDI QIDQ5075835FDOQ5075835

S. L. Brunton, J. N. Kutz, Marko Budišić, Eurika Kaiser

Publication date: 12 May 2022

Published in: SIAM Review (Search for Journal in Brave)

Abstract: The field of dynamical systems is being transformed by the mathematical tools and algorithms emerging from modern computing and data science. First-principles derivations and asymptotic reductions are giving way to data-driven approaches that formulate models in operator theoretic or probabilistic frameworks. Koopman spectral theory has emerged as a dominant perspective over the past decade, in which nonlinear dynamics are represented in terms of an infinite-dimensional linear operator acting on the space of all possible measurement functions of the system. This linear representation of nonlinear dynamics has tremendous potential to enable the prediction, estimation, and control of nonlinear systems with standard textbook methods developed for linear systems. However, obtaining finite-dimensional coordinate systems and embeddings in which the dynamics appear approximately linear remains a central open challenge. The success of Koopman analysis is due primarily to three key factors: 1) there exists rigorous theory connecting it to classical geometric approaches for dynamical systems, 2) the approach is formulated in terms of measurements, making it ideal for leveraging big-data and machine learning techniques, and 3) simple, yet powerful numerical algorithms, such as the dynamic mode decomposition (DMD), have been developed and extended to reduce Koopman theory to practice in real-world applications. In this review, we provide an overview of modern Koopman operator theory, describing recent theoretical and algorithmic developments and highlighting these methods with a diverse range of applications. We also discuss key advances and challenges in the rapidly growing field of machine learning that are likely to drive future developments and significantly transform the theoretical landscape of dynamical systems.

Full work available at URL: https://arxiv.org/abs/2102.12086

Recommendations

• Data-driven approximations of dynamical systems operators for control
• Applied Koopmanism
• The Koopman operator in systems and control. Concepts, methodologies and applications
• On dynamic mode decomposition: theory and applications
• Generalizing Koopman Theory to Allow for Inputs and Control

zbMATH Keywords

dynamical systemscontrol theoryembeddingsspectral theoryoperator theorydynamic mode decompositionKoopman operatordata-driven discovery

Mathematics Subject Classification ID

Nonlinear ordinary differential equations and systems (34A34) Dynamics induced by flows and semiflows (37C10) Dynamical systems in control (37N35) Linear composition operators (47B33) Approximation methods and numerical treatment of dynamical systems (37M99) Ergodic theory of linear operators (47A35)

Cites Work

• Title not available (Why is that?)
• An Introduction to Statistical Learning
• Online Dynamic Mode Decomposition for Time-Varying Systems
• Title not available (Why is that?)
• Model predictive control: Theory and practice - a survey
• Diffusion maps
• Nonlinear oscillations, dynamical systems, and bifurcations of vector fields
• Statistical modeling: The two cultures. (With comments and a rejoinder).
• Title not available (Why is that?)
• Data assimilation: Mathematical and statistical perspectives
• Finding structure with randomness: Probabilistic algorithms for constructing approximate matrix decompositions
• Title not available (Why is that?)
• The Future of Data Analysis
• Discovering governing equations from data by sparse identification of nonlinear dynamical systems
• Compressive Sensing and Low-Rank Libraries for Classification of Bifurcation Regimes in Nonlinear Dynamical Systems
• Dynamic mode decomposition of numerical and experimental data
• Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information
• Spectral analysis of nonlinear flows
• Analysis of Fluid Flows via Spectral Properties of the Koopman Operator
• Turbulence, Coherent Structures, Dynamical Systems and Symmetry
• Compressed sensing
• Multiresolution dynamic mode decomposition
• Variable Projection Methods for an Optimized Dynamic Mode Decomposition
• Data-Driven Sparse Sensor Placement for Reconstruction: Demonstrating the Benefits of Exploiting Known Patterns
• Manifolds, tensor analysis, and applications.
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• The Inverse Scattering Transform‐Fourier Analysis for Nonlinear Problems
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Ergodic theory of chaos and strange attractors
• Title not available (Why is that?)
• Title not available (Why is that?)
• Integrals of nonlinear equations of evolution and solitary waves
• On a quasi-linear parabolic equation occurring in aerodynamics
• Chaos, fractals, and noise: Stochastic aspects of dynamics.
• Diffusion maps, spectral clustering and reaction coordinates of dynamical systems
• Hydrodynamic Stability Without Eigenvalues
• Diffusion Maps, Reduction Coordinates, and Low Dimensional Representation of Stochastic Systems
• Title not available (Why is that?)
• Title not available (Why is that?)
• Estimating the domain of attraction of nonlinear feedback systems
• On the Approximation of Complicated Dynamical Behavior
• On the Early History of the Singular Value Decomposition
• Proof of the Quasi-Ergodic Hypothesis
• Geometric diffusions as a tool for harmonic analysis and structure definition of data: Diffusion maps
• Title not available (Why is that?)
• Physics constrained nonlinear regression models for time series
• Mathematical Description of Linear Dynamical Systems
• The partial differential equation u_t + uu_x = μ_xx
• A survey of projection-based model reduction methods for parametric dynamical systems
• On the use of Fourier averages to compute the global isochrons of (quasi)periodic dynamics
• Title not available (Why is that?)
• Introduction to Applied Nonlinear Dynamical Systems and Chaos
• Operator Theoretic Aspects of Ergodic Theory
• Chaos, fractional kinetics, and anomalous transport
• Optimal control of the Liouville equation
• Nonlinear controllability and observability
• Chaos, Scattering and Statistical Mechanics
• Substitution dynamical systems. Spectral analysis
• Definition and properties of Lagrangian coherent structures from finite-time Lyapunov exponents in two-dimensional aperiodic flows
• Operator theoretic framework for optimal placement of sensors and actuators for control of nonequilibrium dynamics
• Actuator and sensor placement in linear advection PDE with building system application
• Zur Operatorenmethode in der klassischen Mechanik
• Title not available (Why is that?)
• The ensemble Kalman filter for combined state and parameter estimation
• Non-linear autonomous systems of differential equations and Carleman linearization procedure
• Lagrangian coherent structures from approximate velocity data
• Title not available (Why is that?)
• 4SID: Subspace algorithms for the identification of combined deterministic-stochastic systems
• Comparison of systems with complex behavior
• Variants of dynamic mode decomposition: boundary condition, Koopman, and Fourier analyses
• Spectral properties of dynamical systems, model reduction and decompositions
• On dynamic mode decomposition: theory and applications
• Dynamic mode decomposition with control
• Analysis of unsteady behaviour in shockwave turbulent boundary layer interaction
• Applied Koopmanism
• Compressive sensing based machine learning strategy for characterizing the flow around a cylinder with limited pressure measurements
• Compressed sensing and dynamic mode decomposition
• A kernel-based method for data-driven Koopman spectral analysis
• Title not available (Why is that?)
• Reduced-order models for control of fluids using the eigensystem realization algorithm
• Applications of the dynamic mode decomposition
• An eigensystem realization algorithm for modal parameter identification and model reduction
• Detecting and Locating Near-Optimal Almost-Invariant Sets and Cycles
• Geometry of the ergodic quotient reveals coherent structures in flows
• A method for visualization of invariant sets of dynamical systems based on the ergodic partition
• Dynamical Systems of Continuous Spectra
• Title not available (Why is that?)
• Ergodic theory and visualization. I. Mesochronic plots for visualization of ergodic partition and invariant sets
• Koopman-mode decomposition of the cylinder wake
• Selected topics from the metric theory of dynamical systems
• A data-driven approximation of the koopman operator: extending dynamic mode decomposition
• On the rate of mixing of Axiom A flows
• Finite approximation for the Frobenius-Perron operator. A solution to Ulam's conjecture
• A subdivision algorithm for the computation of unstable manifolds and global attractors
• Applied and computational measurable dynamics
• Data Driven Modal Decompositions: Analysis and Enhancements
• Sufficient conditions for controllability
• Koopman Operator Family Spectrum for Nonautonomous Systems
• On Matching, and Even Rectifying, Dynamical Systems through Koopman Operator Eigenfunctions
• Time-Delay Observables for Koopman: Theory and Applications
• Learning data-driven discretizations for partial differential equations
• Koopman mode expansions between simple invariant solutions
• Linearly Recurrent Autoencoder Networks for Learning Dynamics
• Optimal Construction of Koopman Eigenfunctions for Prediction and Control
• Extraction and prediction of coherent patterns in incompressible flows through space-time koopman analysis
• Discovery of Nonlinear Multiscale Systems: Sampling Strategies and Embeddings
• Dynamic mode decomposition for continuous time systems with the Liouville operator
• Koopman operator-based model reduction for switched-system control of PDEs
• Data-Driven Model Predictive Control using Interpolated Koopman Generators
• Tensor-based computation of metastable and coherent sets
• Existence and uniqueness of global Koopman eigenfunctions for stable fixed points and periodic orbits
• Global linearization and fiber bundle structure of invariant manifolds
• Centering Data Improves the Dynamic Mode Decomposition
• Kernel Analog Forecasting: Multiscale Test Problems
• Consistent Dynamic Mode Decomposition
• Feedback control of nonlinear PDEs using data-efficient reduced order models based on the Koopman operator
• Deep learning models for global coordinate transformations that linearise PDEs
• Title not available (Why is that?)
• From Bypass Transition to Flow Control and Data-Driven Turbulence Modeling: An Input–Output Viewpoint
• An incremental approach to online dynamic mode decomposition for time-varying systems with applications to EEG data modeling
• Model and data reduction for data assimilation: particle filters employing projected forecasts and data with application to a shallow water model
• The Koopman Operator in Systems and Control
• Data-driven resolvent analysis
• Dynamic mode decomposition in vector-valued reproducing kernel Hilbert spaces for extracting dynamical structure among observables
• Fréchet differentiable drift dependence of Perron–Frobenius and Koopman operators for non-deterministic dynamics
• Dynamic mode decomposition for analytic maps
• Delay-coordinate maps, coherence, and approximate spectra of evolution operators
• Title not available (Why is that?)
• Koopman operator and its approximations for systems with symmetries
• On the Koopman Operator of Algorithms
• A data-driven phase and isostable reduced modeling framework for oscillatory dynamical systems
• Carleman Linearization of Linearly Observable Polynomial Systems
• Title not available (Why is that?)
• Zero-dimensional singular continuous spectrum for smooth differential equations on the torus
• Invariant Sets in Quasiperiodically Forced Dynamical Systems
• Sparsity-promoting algorithms for the discovery of informative Koopman-invariant subspaces
• Estimation of the Koopman Generator by Newton's Extrapolation
• On Koopman mode decomposition and tensor component analysis
• On Least Squares Problems with Certain Vandermonde--Khatri--Rao Structure with Applications to DMD
• Sparse identification of multiphase turbulence closures for coupled fluid–particle flows
• Koopman Analysis of Isolated Fronts and Solitons
• Pseudogenerators of Spatial Transfer Operators
• On the numerical approximation of the Perron-Frobenius and Koopman operator
• Estimating Long-Term Behavior of Flows without Trajectory Integration: The Infinitesimal Generator Approach
• A Variational Approach to Modeling Slow Processes in Stochastic Dynamical Systems
• Data-driven approximations of dynamical systems operators for control
• Componentwise energy amplification in channel flows
• The Lorenz attractor is mixing
• Flow structures around a high-speed train extracted using proper orthogonal decomposition and dynamic mode decomposition
• Uncertainty propagation in dynamical systems
• Computing Lagrangian coherent structures from their variational theory
• Experimental investigation of nonlinear instabilities in annular liquid sheets
• Statistically accurate low-order models for uncertainty quantification in turbulent dynamical systems
• Fourier asymptotics of fractal measures
• Title not available (Why is that?)
• The rate of convergence in ergodic theorems
• Harmonic Analysis and Ergodic Theory
• Title not available (Why is that?)
• The dynamics of control. With an appendix by Lars Grüne
• The role of the rigged Hilbert space in quantum mechanics
• Time-series analysis
• Mode decomposition on surface-mounted cube
• Coherent swing instability of power grids
• A critical-layer framework for turbulent pipe flow
• Volterra series and geometric control theory
• APPROXIMATIONS IN ERGODIC THEORY
• Approximation of measure preserving transformations
• Linear system identification via an asymptotically stable observer
• A trajectory-free framework for analysing multiscale systems
• Application de la théorie des équations intégrales linéaires aux systèmes d'équations différentielles non linéaires
• Escape rates and Perron-Frobenius operators: open and closed dynamical systems
• Spectral characterization of WienerWintner dynamical systems
• Identification of observer/Kalman filter Markov parameters - Theory and experiments
• Exploring invariant sets and invariant measures
• On the Approximation of Koopman Spectra for Measure Preserving Transformations
• Turbulence Modeling in the Age of Data
• Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders
• Recursive dynamic mode decomposition of transient and post-transient wake flows
• Data-driven operator inference for nonintrusive projection-based model reduction
• Tensor-based dynamic mode decomposition
• Data-driven model reduction and transfer operator approximation
• A set oriented definition of finite-time Lyapunov exponents and coherent sets
• Conceptual dynamical models for turbulence
• Spectral analysis of the Koopman operator for partial differential equations
• Data-driven spectral analysis of the Koopman operator
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Delay-coordinate maps and the spectra of Koopman operators
• Data-driven spectral decomposition and forecasting of ergodic dynamical systems
• Data-Driven Science and Engineering
• Randomized Dynamic Mode Decomposition
• Data Assimilation
• Reynolds averaged turbulence modelling using deep neural networks with embedded invariance
• Koopman spectra in reproducing kernel Hilbert spaces
• On convergence of extended dynamic mode decomposition to the Koopman operator
• Isostables, isochrons, and Koopman spectrum for the action-angle representation of stable fixed point dynamics
• Global Stability Analysis Using the Eigenfunctions of the Koopman Operator
• Ergodic Theory, Dynamic Mode Decomposition, and Computation of Spectral Properties of the Koopman Operator
• Ruelle-Pollicott resonances of stochastic systems in reduced state space. Part I: Theory
• Approximation Theories for Measure Preserving Transformations
• Linear predictors for nonlinear dynamical systems: Koopman operator meets model predictive control
• Machine Learning for Fluid Mechanics
• Dynamic mode decomposition for financial trading strategies
• Spectral proper orthogonal decomposition and its relationship to dynamic mode decomposition and resolvent analysis
• Singular continuous spectrum and quantitative rates of weak mixing
• Data-driven approximation of the Koopman generator: model reduction, system identification, and control
• Data-driven nonlinear stabilization using Koopman operator
• Extended dynamic mode decomposition with dictionary learning: A data-driven adaptive spectral decomposition of the Koopman operator
• Data Driven Koopman Spectral Analysis in Vandermonde--Cauchy Form via the DFT: Numerical Method and Theoretical Insights
• Koopman operator theory for nonautonomous and stochastic systems
• Rigged Hilbert spaces for chaotic dynamical systems
• Resonances and the extension of dynamics to rigged Hilbert spaces
• Generalized spectral decomposition and intrinsic irreversibility of the Arnold cat map
• Ergodic theorem, ergodic theory, and statistical mechanics
• Reproducing kernel Hilbert space compactification of unitary evolution groups
• Applied Nonautonomous and Random Dynamical Systems
• Sparsity enabled cluster reduced-order models for control
• A new selection operator for the discrete empirical interpolation method -- improved a priori error bound and extensions
• Deep learning in fluid dynamics
• Applied Koopman theory for partial differential equations and data-driven modeling of spatio-temporal systems
• Infinite-dimensional Carleman linearization, the Lie series and optimal control of non-linear partial differential equations
• Sparse Sensing and DMD-Based Identification of Flow Regimes and Bifurcations in Complex Flows
• Linearization in the large of nonlinear systems and Koopman operator spectrum
• Variational approach for learning Markov processes from time series data
• Koopman-Based Lifting Techniques for Nonlinear Systems Identification
• On reduced input-output dynamic mode decomposition
• Generalizing Koopman Theory to Allow for Inputs and Control
• Title not available (Why is that?)
• Optimized Sampling for Multiscale Dynamics
• Subgrid modelling for two-dimensional turbulence using neural networks
• Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations
• Eigendecompositions of transfer operators in reproducing kernel Hilbert spaces
• Sparse identification of nonlinear dynamics for model predictive control in the low-data limit
• Ergodic theory and visualization. II. Fourier mesochronic plots visualize (quasi)periodic sets
• Operator inference for non-intrusive model reduction of systems with non-polynomial nonlinear terms
• Koopman operator spectrum for random dynamical systems
• On robust computation of Koopman operator and prediction in random dynamical systems
• Spectrum of the Koopman operator, spectral expansions in functional spaces, and state-space geometry
• Data-driven Koopman operator approach for computational neuroscience
• Geometric considerations of a good dictionary for Koopman analysis of dynamical systems: cardinality, ``primary eigenfunction, and efficient representation
• Lift \& learn: physics-informed machine learning for large-scale nonlinear dynamical systems
• Detecting isolated spectrum of transfer and Koopman operators with Fourier analytic tools

Cited In (30)

• Design of the monodomain model by artificial neural networks
• Applied Koopmanism
• Finite-data error bounds for Koopman-based prediction and control
• A study on data-driven identification and representation of nonlinear dynamical systems with a physics-integrated deep learning approach: Koopman operators and nonlinear normal modes
• Parsimony as the ultimate regularizer for physics-informed machine learning
• On the Approximability of Koopman-Based Operator Lyapunov Equations
• Koopman neural operator as a mesh-free solver of non-linear partial differential equations
• On principles of emergent organization
• Multiscale model reduction for incompressible flows
• From snapshots to manifolds – a tale of shear flows
• Projected data assimilation using sliding window proper orthogonal decomposition
• The Adaptive Spectral Koopman Method for Dynamical Systems
• A LAPACK implementation of the dynamic mode decomposition
• Approximation of translation invariant Koopman operators for coupled non-linear systems
• Auxiliary functions as Koopman observables: data-driven analysis of dynamical systems via polynomial optimization
• Koopman Operator Inspired Nonlinear System Identification
• Ensemble forecasts in reproducing kernel Hilbert space family
• Real algebraic geometry with a view toward Koopman operator methods. Abstracts from the workshop held March 12--17, 2023
• Residual dynamic mode decomposition: robust and verified Koopmanism
• Another look at residual dynamic mode decomposition in the regime of fewer snapshots than dictionary size
• Koopman form of nonlinear systems with inputs
• EDMD for expanding circle maps and their complex perturbations
• Koopman Linear Quadratic Regulator Using Complex Eigenfunctions for Nonlinear Dynamical Systems
• Dictionary-free Koopman model predictive control with nonlinear input transformation
• The mpEDMD Algorithm for Data-Driven Computations of Measure-Preserving Dynamical Systems
• Koopman analysis of the singularly perturbed van der Pol oscillator
• Probabilistic forecast of nonlinear dynamical systems with uncertainty quantification
• Deep neural network based adaptive learning for switched systems
• Non-intrusive data-driven reduced-order modeling for time-dependent parametrized problems
• Reduced-order variational mode decomposition to reveal transient and non-stationary dynamics in fluid flows

Uses Software

• EnKF
• GAIO
• rsvd
• odmd
• VAMPnets
• PyDMD
• DeepONet
• LANDO

This page was built for publication: Modern Koopman theory for dynamical systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5075835)