Deep learning or interpolation for inverse modelling of heat and fluid flow problems?
International Journal of Numerical Methods for Heat & Fluid Flow
ISSN: 0961-5539
Article publication date: 22 January 2021
Issue publication date: 26 August 2021
Abstract
Purpose
The purpose of this study is to compare interpolation algorithms and deep neural networks for inverse transfer problems with linear and nonlinear behaviour.
Design/methodology/approach
A series of runs were conducted for a canonical test problem. These were used as databases or “learning sets” for both interpolation algorithms and deep neural networks. A second set of runs was conducted to test the prediction accuracy of both approaches.
Findings
The results indicate that interpolation algorithms outperform deep neural networks in accuracy for linear heat conduction, while the reverse is true for nonlinear heat conduction problems. For heat convection problems, both methods offer similar levels of accuracy.
Originality/value
This is the first time such a comparison has been made.
Keywords
Acknowledgements
H. Antil is partially supported by NSF Grants DMS-1818772, DMS-1913004, the Air Force Office of Scientific Research (AFOSR) under Award No.: FA9550-19–1-0036, and Department of Navy, Naval Postgraduate School under Award No.: N00244-20–1-0005.
Perumal Nithiarasu acknowledges partial support from Ser Cymru III – Tackling Covid-19 fund, Welsh Government Project number 095.
The authors would also acknowledge the very thorough review and comments of an anonymous reviewer, which was very constructive and helped to improve the archival quality of the paper.
Citation
Löhner, R., Antil, H., Tamaddon-Jahromi, H., Chakshu, N.K. and Nithiarasu, P. (2021), "Deep learning or interpolation for inverse modelling of heat and fluid flow problems?", International Journal of Numerical Methods for Heat & Fluid Flow, Vol. 31 No. 9, pp. 3036-3046. https://doi.org/10.1108/HFF-11-2020-0684
Publisher
:Emerald Publishing Limited
Copyright © 2020, Emerald Publishing Limited