Investigating the effect of textural properties on CO2 adsorption in porous carbons via deep neural networks using various training algorithms
Pardis Mehrmohammadi, Ahad Ghaemi · 2023
Read Investigating the effect of textural properties on CO2 adsorption in porous carbons via deep neural networks using various training algorithms by Pardis Mehrmohammadi, Ahad Ghaemi (2023) 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-10692134.txt.
Description
Pardis Mehrmohammadi and Ahad Ghaemi published this modeling study in 2023 on capturing carbon dioxide with porous carbons. How much gas these materials soak up depends on their texture, above all their tiniest pores. Rather than run endless lab tests, the authors trained deep neural networks to predict uptake from textural properties, and they compared many training algorithms for the job.
One algorithm, Levenberg-Marquardt, gave the most accurate network, with extremely low error, and it also converged in the fewest passes. A rival was fastest per run, while another took the longest. The team also tuned a second network type, finding a nine-neuron hidden layer worked best. On fresh data, the top networks predicted uptake with the highest fidelity. The study maps how pore size rules uptake: micropores dominate at low pressure and larger pores at high pressure.
Catalog record
| CREATOR | Pardis Mehrmohammadi, Ahad Ghaemi |
|---|---|
| YEAR | 2023 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 73 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |