About this Abstract |
Meeting |
TMS Specialty Congress 2024
|
Symposium
|
2nd World Congress on Artificial Intelligence in Materials & Manufacturing (AIM 2024)
|
Presentation Title |
Application of Graph Neural Network in Prediction of Mesoscale Structure in Dense Slurries |
Author(s) |
Joao Maia, Armin Aminimajd, Abhinendra Singh |
On-Site Speaker (Planned) |
Abhinendra Singh |
Abstract Scope |
Understanding the rheological response of particulate system is relevant to industry and nature. The mesoscale force chain network (FCN) structure holds the key for structure-property relationship in these disordered materials. While traditional simulation methods are expensive, recent deep learning techniques were found to be a powerful tool to predict properties of particulate systems. Herein, we train the deep graph convolutional neural network (GCNN) model using datasets obtained from lubrication flow discrete element modeling to accurately predict (above 98%) FCN in suspensions for different control variables. Our machine learning model goes beyond recent research on granular materials; it not only predicts FCN with higher accuracy but also interpolates and extrapolates to conditions far from control parameters. The method used in this study can be utilized for prediction of rheological and characterization of complex particulate systems in the future. |
Proceedings Inclusion? |
Definite: Other |