Share

IIR document

Digital twin of a full-scale industrial heat pump producing steam above 140 °C.

Number: 1202

Author(s) : ISPIR A. C., OTERO RODRIGUEZ G., DE VRIES W., SPEETJENS M.

Summary

Industrial heat pumps (HPs) are an efficient and, if running on green electricity, sustainable alternative to fossil fuels for the production of process heat. One of the main challenges that limit end-user confidence in this technology is the lack of understanding of HP behaviour for the varying operating ranges and dynamic conditions in industry. Conventional approaches to investigate HP behaviour and performance under such conditions, such as high-fidelity numerical simulations or experimental testing can be expensive and impractical and are, therefore, often not an option. We address this challenge by proposing a digital-twin framework that enables efficient modelling of industrial HPs entirely from data by a machine-learning approach (i.e., Gaussian Process Regression). This concept involves creating a digital replica of the heat pump system under study by integrating machine learning approaches into the experimental simulation dataset and estimating performance based on user-defined independent parameters. The digital twining is demonstrated and validated for the steady-state operation of a full-scale industrial HP using n-Pentane (R601) as the working medium, which serves as a test facility for (future) high-temperature steam-producing HPs. This reveals that our digital twin can accurately predict key process variables and performance indicators in the operating regime. Moreover, its entirely data-driven nature makes the same modelling methodology readily applicable to other types of (industrial) HPs.

Available documents

Format PDF

Pages: 11 p.

Available

  • Public price

    20 €

  • Member price*

    Free

* Best rate depending on membership category (see the detailed benefits of individual and corporate memberships).

Details

  • Original title: Digital twin of a full-scale industrial heat pump producing steam above 140 °C.
  • Record ID : 30032715
  • Languages: English
  • Source: 16th IIR-Gustav Lorentzen Conference on Natural Refrigerants (GL2024). Proceedings. University of Maryland, College Park, Maryland, USA, August 11-14 2024
  • Publication date: 2024/08
  • DOI: http://dx.doi.org/10.18462/irr.gl2024.1202

Links


See other articles from the proceedings (135)
See the conference proceedings

Indexing