Integrating scientific knowledge with machine learning for engineering and environmental systems

J Willard, X Jia, S Xu, M Steinbach, V Kumar - ACM Computing Surveys, 2022 - dl.acm.org
There is a growing consensus that solutions to complex science and engineering problems
require novel methodologies that are able to integrate traditional physics-based modeling …

Data-driven modeling for unsteady aerodynamics and aeroelasticity

J Kou, W Zhang - Progress in Aerospace Sciences, 2021 - Elsevier
Aerodynamic modeling plays an important role in multiphysics and design problems, in
addition to experiment and numerical simulation, due to its low-dimensional representation …

[PDF][PDF] Integrating physics-based modeling with machine learning: A survey

J Willard, X Jia, S Xu, M Steinbach… - arXiv preprint arXiv …, 2020 - beiyulincs.github.io
There is a growing consensus that solutions to complex science and engineering problems
require novel methodologies that are able to integrate traditional physics-based modeling …

[HTML][HTML] POD-DL-ROM: Enhancing deep learning-based reduced order models for nonlinear parametrized PDEs by proper orthogonal decomposition

S Fresca, A Manzoni - Computer Methods in Applied Mechanics and …, 2022 - Elsevier
Deep learning-based reduced order models (DL-ROMs) have been recently proposed to
overcome common limitations shared by conventional reduced order models (ROMs)–built …

Physics-guided neural networks (pgnn): An application in lake temperature modeling

A Daw, A Karpatne, WD Watkins… - Knowledge Guided …, 2022 - taylorfrancis.com
This chapter introduces a framework for combining scientific knowledge of physics-based
models with neural networks to advance scientific discovery. It explains termed physics …

[HTML][HTML] Reduced-order modeling of advection-dominated systems with recurrent neural networks and convolutional autoencoders

R Maulik, B Lusch, P Balaprakash - Physics of Fluids, 2021 - pubs.aip.org
A common strategy for the dimensionality reduction of nonlinear partial differential equations
(PDEs) relies on the use of the proper orthogonal decomposition (POD) to identify a reduced …

[HTML][HTML] Physics-informed machine learning for reduced-order modeling of nonlinear problems

W Chen, Q Wang, JS Hesthaven, C Zhang - Journal of computational …, 2021 - Elsevier
A reduced basis method based on a physics-informed machine learning framework is
developed for efficient reduced-order modeling of parametrized partial differential equations …

Super-resolution analysis via machine learning: a survey for fluid flows

K Fukami, K Fukagata, K Taira - Theoretical and Computational Fluid …, 2023 - Springer
This paper surveys machine-learning-based super-resolution reconstruction for vortical
flows. Super resolution aims to find the high-resolution flow fields from low-resolution data …

A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized PDEs

S Fresca, L Dede', A Manzoni - Journal of Scientific Computing, 2021 - Springer
Conventional reduced order modeling techniques such as the reduced basis (RB) method
(relying, eg, on proper orthogonal decomposition (POD)) may incur in severe limitations …

Subgrid modelling for two-dimensional turbulence using neural networks

R Maulik, O San, A Rasheed, P Vedula - Journal of Fluid Mechanics, 2019 - cambridge.org
In this investigation, a data-driven turbulence closure framework is introduced and deployed
for the subgrid modelling of Kraichnan turbulence. The novelty of the proposed method lies …