This study presents an intelligent analytical tool for evaluating the effect of carbon black type and loading on the mixing dynamics of rubber compounds using an artificial neural network. A Generalized Regression Neural Network (GRNN) model was developed to simulate the time evolution of torque and temperature in natural rubber-based compounds. The model's input parameters included the type of carbon black, its concentration in the compound, and the mixing time, while the outputs were the instantaneous values of torque and temperature recorded throughout the mixing process. Based on experimental data from laboratory mixing trials, the simulation results of the well-trained GRNN model demonstrated high accuracy in capturing complex, nonlinear relationships between the inputs and the system’s rheological response. The proposed neural model offers a novel approach to modeling the time-dependent rheological behavior of rubber compounds during processing. It enables more accurate predictions and provides a practical tool for optimizing compound formulations and improving the control of dynamic processes in rubber manufacturing and related industries.

Acknowledgement: This research work has been supported by the Operational Programme Integrated Infrastructure - project CEDITEK II., ITMS2014+ code 313011W442.