Prediction Model for Tensile Modulus of Carbon Nanotube–Polymer Composites

Abstract

This work develops a Neural Network (NN) model for the prediction of the tensile modulus of carbon nanotube (CN)/polymer nanocomposites, based on experimental database. A data set composed of 282 configurations is collected from available resources. Considered input variables of the dataset are such as mechanical properties of separated phases, density of polymer matrix, processing method, geometry of CN, modification method at the CN surface, etc. while the problem output is the tensile modulus of nanocomposite. Parametric studies have been performed in finding optimum architecture of the proposed NN model.