Model Order Reduction with Neural Networks: Application to Laminar and Turbulent Flows

• Published in: SN computer science Vol. 2; no. 6; p. 467
• Main Authors: Fukami, Kai, Hasegawa, Kazuto, Nakamura, Taichi, Morimoto, Masaki, Fukagata, Koji
• Format: Journal Article
• Language: English
• Published: Singapore Springer Singapore 01.11.2021; Springer Nature B.V
• Subjects: Artificial neural networks; Channel flow; Computer Imaging; Computer Science; Computer Systems Organization and Communication Networks; Data Structures and Information Theory; Dynamical systems; Fluid dynamics; Fluid flow; Information Systems and Communication Service; Laminar flow; Machine learning; Methods; Model reduction; Multilayer perceptrons; Multilayers; Neural networks; Original Research; Parameter sensitivity; Pattern Recognition and Graphics; Reynolds number; Sea surface temperature; Software Engineering/Programming and Operating Systems; Turbulent flow; Two dimensional flow; Vision; Turbulence; Reduced order model; Autoencoder; Wake
• ISSN: 2662-995X; 2661-8907
• DOI: 10.1007/s42979-021-00867-3

We investigate the capability of neural network-based model order reduction, i.e., autoencoder (AE), for fluid flows. As an example model, an AE which comprises of convolutional neural networks and multi-layer perceptrons is considered in this study. The AE model is assessed with four canonical fluid flows, namely: (1) two-dimensional cylinder wake, (2) its transient process, (3) NOAA sea surface temperature, and (4) a cross-sectional field of turbulent channel flow, in terms of a number of latent modes, the choice of nonlinear activation functions, and the number of weights contained in the AE model. We find that the AE models are sensitive to the choice of the aforementioned parameters depending on the target flows. Finally, we foresee the extensional applications and perspectives of machine learning based order reduction for numerical and experimental studies in the fluid dynamics community.