Modeling the effect of carbon black type and dosage on the mixing dynamics of rubber blends using a general regression neural network
Abstract
This study presents the application of a general regression neural network to model the effects of carbon black type and content on the mixing dynamics of rubber blends.The neural network model, trained on data from 16 tests (four carbon black types with contents of 45-60 phr), accurately reconstructs the torque and temperature profiles during blend mixing.Model optimization was performed using 10-fold cross-validation.The simulation results (average R ≈ 1 and RMSE ≈ 10 -4 ) confirm the practical applicability of the presented model for predicting the time evolution of nonlinear processes in complex technological applications.
Fields of Study & Subjects
Neural Networks and Applications
Machine Learning in Materials Science
Polymer Science and Applications
Publication details
- Journal / source
- Journal of Applied Mathematics and Computational Mechanics
- Volume
- 25
- Issue
- 1
- Pages
- 86–97
- Publication date
- 1 March 2026
- Language
- EN