Multi-fidelity Learning of Reduced Order Models for Parabolic PDE Constrained Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Klein, Benedikt, Ohlberger, Mario
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916663852531712
author Klein, Benedikt
Ohlberger, Mario
author_facet Klein, Benedikt
Ohlberger, Mario
contents This article builds on the recently proposed RB-ML-ROM approach for parameterized parabolic PDEs and proposes a novel hierarchical Trust Region algorithm for solving parabolic PDE constrained optimization problems. Instead of using a traditional offline/online splitting approach for model order reduction, we adopt an active learning or enrichment strategy to construct a multi-fidelity hierarchy of reduced order models on-the-fly during the outer optimization loop. The multi-fidelity surrogate model consists of a full order model, a reduced order model and a machine learning model. The proposed hierarchical framework adaptively updates its hierarchy when querying parameters, utilizing a rigorous a posteriori error estimator in an error aware trust region framework. Numerical experiments are given to demonstrate the efficiency of the proposed approach.
format Preprint
id arxiv_https___arxiv_org_abs_2503_21252
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Multi-fidelity Learning of Reduced Order Models for Parabolic PDE Constrained Optimization
Klein, Benedikt
Ohlberger, Mario
Optimization and Control
Numerical Analysis
49M20, 49K20, 35J20, 65N30, 90C06
This article builds on the recently proposed RB-ML-ROM approach for parameterized parabolic PDEs and proposes a novel hierarchical Trust Region algorithm for solving parabolic PDE constrained optimization problems. Instead of using a traditional offline/online splitting approach for model order reduction, we adopt an active learning or enrichment strategy to construct a multi-fidelity hierarchy of reduced order models on-the-fly during the outer optimization loop. The multi-fidelity surrogate model consists of a full order model, a reduced order model and a machine learning model. The proposed hierarchical framework adaptively updates its hierarchy when querying parameters, utilizing a rigorous a posteriori error estimator in an error aware trust region framework. Numerical experiments are given to demonstrate the efficiency of the proposed approach.
title Multi-fidelity Learning of Reduced Order Models for Parabolic PDE Constrained Optimization
topic Optimization and Control
Numerical Analysis
49M20, 49K20, 35J20, 65N30, 90C06
url https://arxiv.org/abs/2503.21252