Parameterised non-intrusive reduced-order model for general unsteady flow problems using artificial neural networks

Sugar-Gabor, O

Abstract

A non-intrusive reduced-order model for nonlinear parametric flow problems is developed. It
is based on extracting a reduced-order basis from high-order snapshots via proper orthogonal
decomposition and using multi-layered feedforward artificial neural networks to approximate
the reduced-order coefficients. The model is a generic and efficient approach for the reduction
of time-dependent parametric systems, including those described by partial differential
equations. Since it is non-intrusive, it is independent of the high-order computational method
and can be used together with black-box solvers. Numerical studies are presented for steadystate isentropic nozzle flow with geometric parameterisation and unsteady parameterised
viscous Burgers equation. An adaptive sampling strategy is proposed to increase the quality
of the neural network approximation while minimising the required number of parameter
samples and, as a direct consequence, the number of high-order snapshots and the size of
the network training set. Results confirm the accuracy of the non-intrusive approach as well
as the speed-up achieved compared with intrusive hyper reduced-order approaches.

Citation

Sugar-Gabor, O. (2021). Parameterised non-intrusive reduced-order model for general unsteady flow problems using artificial neural networks. International Journal for Numerical Methods in Fluids, 93(5), 1309-1331. https://doi.org/10.1002/fld.4930

• Journal Article Type: Article
• Acceptance Date: Oct 17, 2020
• Online Publication Date: Oct 23, 2020
• Publication Date: May 1, 2021
• Deposit Date: Oct 23, 2020
• Publicly Available Date: Oct 23, 2021
• Journal: International Journal for Numerical Methods in Fluids
• Print ISSN: 0271-2091
• Electronic ISSN: 1097-0363
• Publisher: Wiley
• Volume: 93
• Issue: 5
• Pages: 1309-1331
• DOI: https://doi.org/10.1002/fld.4930
• Publisher URL: https://doi.org/10.1002/fld.4930
• Related Public URLs: http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1097-0363
• Additional Information: Access Information : This is the peer reviewed version of the following article: Şugar‐Gabor, O. Parameterized nonintrusive reduced‐order model for general unsteady flow problems using artificial neural networks. Int J Numer Meth Fluids. 2021; 93: 1309– 1331., which has been published in final form at https://doi.org/10.1002/fld.4930. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions.