Physics-informed neural networks and time-series transformer for modeling of chemical reactors

Giacomo Lastrucci, Maximilian F. Theisen, Artur M. Schweidtmann*

*Corresponding author for this work

Research output: Chapter in Book/Conference proceedings/Edited volumeChapterScientificpeer-review

3 Downloads (Pure)

Abstract

Multiscale modeling of catalytical chemical reactors typically results in solving a system of partial differential equations (PDEs) or ordinary differential equations (ODEs). Despite significant progress, the numerical solution of such PDE or ODE systems is still a computational bottleneck. In the past, deep learning techniques have gained attention for developing surrogate models in chemical engineering. Also, hybrid models and physics-informed neural networks (PINNs) have been developed to integrate physical knowledge and data-driven approaches. However, it is often unclear how such modeling approaches compare for specific case studies. In this study, we investigate and compare state-of-the-art surrogate and hybrid models for the spatial evolution of the state variables in a packetbed reactor for methanol production. Firstly, we develop a tailored hybrid model based on PINNs, thereby seamlessly integrating physical knowledge and data. Secondly, we investigate a recently-developed time-series transformer model to learn the spatial evolution of the state variables. As a benchmark model, we train a traditional multilayer perceptron (MLP) and compare the models to a standard numerical integration technique. We achieve orders of magnitude in speedup using MLPs and PINNs when compared to classical ODE solvers, while maintaining high levels of accuracy in modeling the underlying system.
Original languageEnglish
Title of host publicationProceedings of the 34th European Symposium on Computer Aided Process Engineering
Subtitle of host publication15th International Symposium on Process Systems Engineering (ESCAPE34/PSE24)
EditorsFlavio Manenti, Gintaras V. Reklaitis
Place of PublicationAmsterdam/Kidlington/Cambridge, MA
PublisherElsevier
Pages571-576
Number of pages6
ISBN (Print)978-0-443-33897-7, 978-0-443-28824-1
DOIs
Publication statusPublished - 2024
Event34th European Symposium on Computer-Aided Process Engineering / 15th International Symposium on Process Systems Engineering - Palazzo dei Congressi - Villa Vittoria, Florence, Italy
Duration: 2 Jun 20246 Jun 2024
https://www.aidic.it/escape34-pse24

Publication series

NameComputer Aided Chemical Engineering
PublisherElsevier
Volume53
ISSN (Print)1570-7946

Conference

Conference34th European Symposium on Computer-Aided Process Engineering / 15th International Symposium on Process Systems Engineering
Abbreviated titleESCAPE34 - PSE24
Country/TerritoryItaly
CityFlorence
Period2/06/246/06/24
Internet address

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

  • hybrid modeling
  • physics-informed machine learning
  • reactor modeling
  • time-series transformer

Fingerprint

Dive into the research topics of 'Physics-informed neural networks and time-series transformer for modeling of chemical reactors'. Together they form a unique fingerprint.

Cite this