Identification Of The Tensile Properties Of Hybrid Composite Material By Artificial Neural Network

Composite materials are made from two or more constituent materials with significantly different physical or chemical properties which remain separate and distinct on a macroscopic level within the finished structure. In the present work a silica-styrene-butadiene rubber hybrid composite material is developed with 1 and 2 wt% of nano sized silica particle and 0.25 to 1.5 wt% of styrene-butadiene rubber mixed in resin. Nonlinear relation between mechanical properties and filler material have been obtained. An artificial neural network (ANN) model with one hidden layer and two neurons seems to be approximate for the prediction of mechanical characteristic from the various weight percentage of filler material. Neural networks trained with the back-propagation algorithm are applied to predict the future values of tensile data with different percentage of constitute material. The influencing indicators of constituent percentage of composite material and strain rate are taken into consideration. The design and implementation of a neural network forecasting system is described that has been developed to estimate mechanical properties with different combination of constituent of developed composite material. The performance of the networks is evaluated by comparing them to the experimental data. The comparison shows that neural networks perform the conventional techniques with regard to the prediction quality.