Graph Neural Network for Multitask Prediction of Rheological and Microstructural Behavior in Suspensions

Armin Aminimajd, Joao Maia, Abhinendra Singh · 2026

Read Graph Neural Network for Multitask Prediction of Rheological and Microstructural Behavior in Suspensions by Armin Aminimajd, Joao Maia, Abhinendra Singh (2026) 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/arxiv-2602-07296.txt.

Description

Predicting how a suspension flows matters across industry, from paints to concrete, but direct simulation of dense particle contacts near jamming is costly. Armin Aminimajd and two co-authors train a graph neural network that learns the mapping from particle configurations to suspension behavior.

The multitask model predicts particle pressure, viscosity, and friction coordination at once in dynamic steady state, without explicit knowledge of interparticle forces. It covers regimes from continuous shear thickening through discontinuous thickening to jamming. Process builders get fast predictions where simulation is slow, and a microstructure-informed model that keeps the physics visible in its inputs.

Catalog record

CREATORArmin Aminimajd, Joao Maia, Abhinendra Singh
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE37 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2602-07296

Browse the full library · About the Tower of Babel