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.

Koopman operator learning using invertible neural networks

Extended Dynamic Mode Decomposition with Invertible Dictionary Learning

Invertible neural network combined with dynamic mode decomposition applied to flow field feature extraction and prediction

Invertible Koopman Network and Its Application in Data-Driven Modeling for Dynamic Systems

Diffeomorphically Learning Stable Koopman Operators

Koopman-based Deep Learning for Nonlinear System Estimation

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

Deep Learning Nonlinear Multiscale Dynamic Problems Using Koopman Operator

Learning Invariant Subspaces of Koopman Operators--Part 1: A Methodology for Demonstrating a Dictionary's Approximate Subspace Invariance

Koopman-Based Learning of Infinitesimal Generators without Operator Logarithm

Koopman Operator Theory for Nonlinear Dynamic Modeling using Dynamic Mode Decomposition

Learning Koopman Invariant Subspaces for Dynamic Mode Decomposition

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

Koopman dynamic-oriented deep learning for invariant subspace identification and full-state prediction of complex systems

Koopman analysis of nonlinear systems with a neural network representation

Learning Koopman-based Stability Certificates for Unknown Nonlinear Systems

Extracting Koopman Operators for Prediction and Control of Non-linear Dynamics Using Two-stage Learning and Oblique Projections

The Deep Input-Koopman Operator for Nonlinear Systems

NN-ResDMD: Learning Koopman Representations for Complex Dynamics with Spectral Residuals

Learning Compositional Koopman Operators for Model-Based Control