Abstract:In this paper, a deep learning method using Koopman operator is presented for modeling nonlinear multiscale dynamical problems. Koopman operator is able to transform a non- linear dynamical system into a linear system in a Koopman invariant subspace. However, it is usually very challenging to choose a set of suitable observation functions spanning the Koopman invariant subspace when only data is available for the model. It is practically important for us to predict the evolution of the state of the dynamical system from the Koopman invariant subspace. To this end, we introduce a reconstruction operator that maps the observation function space to the model's state space. Incorporating measurement data, a set of neural networks are constructed to learn the Koopman invariant subspace and the reconstruction operator. The loss function not only considers the properties of Koopman invariant subspace, but also reflects the prediction of future state, which makes the proposed method can realize the prediction of future state for a relatively long time. It may be experimentally expensive to collect the fine-scale data. It will be challenging to use limited computational resources to generate sufficient fine-scale data for neural network training. To overcome this difficulty, we use the data in a coarse-scale and learn effective coarse models for the nonlinear multiscale dynamical problems. In order to make the learned coarse model effectively capture fine-scale information, the loss functions for the neural networks are constructed using a set of multiscale basis functions, which are assumed to be given as a prior. In this case, an accurate fine-scale model can be derived by downscaling the learned coarse model. The deep learning multiscale models using Koopman operator can achieve a relatively long-time prediction for the evolution of the state of the nonlinear multiscale dynamical problems. A few numerical examples are presented to show that the effectiveness of learning multiscale models and the long-time prediction. The numerical results also demonstrate the advantage of the proposed learning method over some other similar learning methods.

Learning Hamiltonian neural Koopman operator and simultaneously sustaining and discovering conservation law

Learning Hamiltonian Neural Koopman Operator and Simultaneously Sustaining and Discovering Conservation Laws

Physics-informed Deep Koopman Operator for Lagrangian Dynamic Systems

Learning Hamiltonian dynamics with reservoir computing

Learning Deep Neural Network Representations for Koopman Operators of Nonlinear Dynamical Systems

Deep Learning for Structure-Preserving Universal Stable Koopman-Inspired Embeddings for Nonlinear Canonical Hamiltonian Dynamics

Learning Koopman-based Stability Certificates for Unknown Nonlinear Systems

Deep Learning Nonlinear Multiscale Dynamic Problems Using Koopman Operator

KoopmanLab: Machine Learning for Solving Complex Physics Equations.

Data-driven optimal control of unknown nonlinear dynamical systems using the Koopman operator

Distributed Deep Koopman Learning for Nonlinear Dynamics

Learning Neural Hamiltonian Dynamics: A Methodological Overview

Koopman neural operator as a mesh-free solver of non-linear partial differential equations

Learning Koopman Operators with Control Using Bi-level Optimization

On Learning Hamiltonian Systems from Data

Learning Noise-Robust Stable Koopman Operator for Control with Hankel DMD

Forecasting Dynamics by an Incomplete Equation of Motion and an Auto-Encoder Koopman Operator

Online Real-time Learning of Dynamical Systems from Noisy Streaming Data: A Koopman Operator Approach

Limits and Powers of Koopman Learning

Self-tuning moving horizon estimation of nonlinear systems via physics-informed machine learning Koopman modeling