Artificial Neural Network based Prediction of Tensile Strength of Hybrid Composites

Abstract

The use of a composite material for any application involves the mechanical characterization. A lot of tests has to be conducted for obtaining satisfactory mechanical properties for the desired application. In this work, the ultimate tensile strength of hybrid short fiber composites made of glass fiber, carbon fiber, and epoxy resin is predicted using artificial neural network techniques. The data for the study has been obtained from tensile test results of certain volume fraction combinations of the constituents. This approach leverages the strengths of both experimental procedure and the mathematical modelling of neural networks for the prediction of unknown values.