What is it about?
Can the equations describing diffusion be improved? This work explores the use of symbolic regression machine learning to identify new empirical equations describing diffusion in simple fluids and mixtures. The performance of these new models is directly compared to existing empirical equation from literature as well as results from more complex artificial neural networks.
Featured Image
Photo by Chaozzy Lin on Unsplash
Why is it important?
Our findings show that diffusion equations obtained using symbolic regression have improved predictive performance in comparison to existing relations from literature, but actually employ a smaller number of adjustable parameters.
Perspectives
I hope this article will lead other researchers to consider using symbolic regression methods when exploring in other applications of machine learning.
Todd Alam
ACC Consulting New Mexico
Read the Original
This page is a summary of: Symbolic regression development of empirical equations for diffusion in Lennard-Jones fluids, The Journal of Chemical Physics, July 2022, American Institute of Physics,
DOI: 10.1063/5.0093658.
You can read the full text:
Contributors
The following have contributed to this page
