Artificial neural network based prediction on tribological properties of polycarbonate composites reinforced with graphene and boron carbide particle

Abstract

Polycarbonate (PC) composites filled with Graphene (up to 10 vol%) and Boron carbide particles (up to 5 vol%) are fabricated by Injection molding and subsequently compressed and quenched. An interactive outcome of incorporated Graphene (C) and Boron carbide particles (B4C) are reported. The lowest wear was prevailed for the composition of PC with 10 vol% C and 5 vol% of B4C. Artificial neural network (ANN) proficiency is employed to predict wear properties of polycarbonate based composites. Frictional coefficient and wear are successfully computed by well trained ANN from the experimental database of Graphene and Boron carbide particles reinforced polycarbonate composite. 3-D plots for the predicted Frictional coefficient and wear as a function of testing conditions and material compositions were constituted. The final results are very much in accord with the measured data. A well trained ANN is anticipated to aid for optimum design of composite materials for consistent parameter studies.