Application of Soft Computing Techniques for Modeling Electrical Conductivity in Carbon Nanotube-Based Nanocomposites

This study presents a soft computing-based approach for modeling the electrical conductivity of single-walled carbon nanotube (SWCNT) reinforced polymer nanocomposites. An artificial neural network (ANN) was developed and trained using normalized experimental data to capture the nonlinear relationship between material parameters and electrical conductivity. The predictive performance of the ANN was evaluated against a conventional linear regression model. Results indicate that the ANN provides significantly improved accuracy and robustness, particularly in regimes governed by percolation and interfacial effects. Furthermore, the model successfully describes the nonlinear evolution of conductivity with increasing SWCNT weight fraction. The proposed methodology demonstrates the effectiveness of soft computing techniques as powerful tools for the design and optimization of CNT-based conductive nanocomposites.

کلیدواژه ها:

Carbon nanotubes ، Electrical conductivity ، Soft computing ، Artificial neural network ، Nanocomposites ، Percolation behavior