Shifted proper orthogonal decomposition and artificial neural networks for time-continuous reduced order models of transport-dominated systems

0560836 - ÚT 2023 CZ eng C - Konferenční příspěvek (zahraniční konf.)
Kovárnová, A. - Krah, P. - Reiss, J. - Isoz, Martin
Shifted proper orthogonal decomposition and artificial neural networks for time-continuous reduced order models of transport-dominated systems.
Topical Problems of Fluid Mechanics 2022. Praha: Ústav termomechaniky AV ČR, v. v. i., 2022 - (Šimurda, D.; Bodnár, T.), s. 111-118. ISBN 978-80-87012-77-2. ISSN 2336-5781.
[Topical problems of fluid mechanics 2022. Praha (CZ), 16.02.2022-18.02.2022]
Grant CEP: GA MŠMT(CZ) EF15_003/0000493
Institucionální podpora: RVO:61388998
Klíčová slova: model order reduction * shifted POD * CFD-DEM * OpenFOAM
Obor OECD: Applied mechanics
http://www2.it.cas.cz/fm2015/im/admin/showfile/data/my/Papers/2022/16-TPFM2022.pdf

Transport-dominated systems are pervasive in both industrial and scientific applications. However, they provide a challenge for common mode-based model order reduction (MOR) approaches, as they often require a large number of linear modes to obtain a sufficiently accurate reduced order model (ROM). In this work, we utilize the shifted proper orthogonal decomposition (sPOD), a methodology tailored for MOR of transport-dominated systems, and combine it with an interpolation based on artificial neural networks (ANN) to obtain a time-continuous ROM usable in engineering practice. The resulting MOR framework is purely data-driven, i.e., it does not require any information on the full order model (FOM) structure, which extends its applicability. On the other hand, compared to the standard projection-based approaches to MOR, the dimensionality reduction utilizing sPOD and ANN is significantly more computationally expensive since it requires a solution of high-dimensional optimization problems.
Trvalý link: https://hdl.handle.net/11104/0334779