Prediction of impact performance of fiber reinforced polymer composites using finite element analysis and artificial neural network

Abstract In this study, a methodology combining finite element analysis (FEA) and artificial neural network (ANN) through multilayer perceptron architecture was utilized to predict the impact resistance behavior of hybrid non-hybrid fabric reinforced polymer (FRP) composites. A projectile at 250 m s −1 velocity considered for high simulations. The Kevlar, carbon glass fabric-based epoxy composites were modelled tests performed residual results from FEA used as training data ANN prediction. predicted in good agreement with maximum variation about 6.6%. terms resistance, composite laminates more Kevlar layers exhibited enhanced performance compared other samples. Neat Kevlar/epoxy (K/K/K) best lowest highest energy absorption 101.84 222.86 J, respectively. Whereas, neat glass/epoxy (G/G/G) specimens registered (165.13 ) (158.99 J) all specimens. 2-fabric sandwich K/G/K low 115.27 218.53 which is second among Comparatively, 3-fabric intermediate lower than that rich specimens, but significantly higher G/G/G composite, thus, proving effectiveness hybridization enhancement composite. Overall, chosen yielded accurate prediction FRP