Echo state networks for modeling turbulent convection
Mohammad Sharifi Ghazijahani, Christian Cierpka · 2024
Read Echo state networks for modeling turbulent convection by Mohammad Sharifi Ghazijahani, Christian Cierpka (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11612492.txt.
Description
Mohammad Sharifi Ghazijahani and Christian Cierpka test echo state networks on one of the classic chaotic flows. Turbulent Rayleigh-Bénard convection is a prominent example of chaos in fluid dynamics with wide relevance in nature. Echo state networks are a basic machine learning tool for sequential data. The authors build a reduced order model of experimental convection data with such a network.
The network modeled the flow well in qualitative terms. Even for this highly turbulent flow, predictions were hard to tell apart from the ground truth. Accuracy extended past raw velocity to spatial and temporal derivatives and vortices. Best hyperparameter values also tracked the flow dynamics, a clue for future network design.
Catalog record
| CREATOR | Mohammad Sharifi Ghazijahani, Christian Cierpka |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |