Koopman Reduced Order Modeling with Confidence Bounds

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mohr, Ryan, Fonoberova, Maria, Mezic, Igor
Format: Preprint
Published: 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912297811705856
author Mohr, Ryan
Fonoberova, Maria
Mezic, Igor
author_facet Mohr, Ryan
Fonoberova, Maria
Mezic, Igor
contents This paper introduces a reduced order modeling technique based on Koopman operator theory that gives confidence bounds on the model's predictions. It is based on a data-driven spectral decomposition of the Koopman operator. The reduced order model is constructed using a finite number of Koopman eigenvalues and modes, while the rest of spectrum is treated as a noise process. This noise process is used to extract the confidence bounds. Additionally, we propose a heuristic algorithm to choose the number of deterministic modes to keep in the model.
format Preprint
id arxiv_https___arxiv_org_abs_2209_13127
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Koopman Reduced Order Modeling with Confidence Bounds
Mohr, Ryan
Fonoberova, Maria
Mezic, Igor
Dynamical Systems
47B33, 15A18, 62F25, 62M20
This paper introduces a reduced order modeling technique based on Koopman operator theory that gives confidence bounds on the model's predictions. It is based on a data-driven spectral decomposition of the Koopman operator. The reduced order model is constructed using a finite number of Koopman eigenvalues and modes, while the rest of spectrum is treated as a noise process. This noise process is used to extract the confidence bounds. Additionally, we propose a heuristic algorithm to choose the number of deterministic modes to keep in the model.
title Koopman Reduced Order Modeling with Confidence Bounds
topic Dynamical Systems
47B33, 15A18, 62F25, 62M20
url https://arxiv.org/abs/2209.13127