A novel numerical method for stochastic study of fiber-reinforced composites with nanoparticles under impact loading

Abstract

Nanoparticle-reinforced composites have caught attention due to their enhanced properties and wide potential use. However, the uncertainty in the distribution of nanoparticles during the manufacturing poses several challenges when deterministic numerical methods are used to model their mechanical behavior, following a macro homogeneous approach. In this work, a stochastic numerical method based on finite element modeling is proposed to simulate the inhomogeneous distribution of nanoparticles inside the composite, focusing on low-velocity impact. Each element in the finite element model is regarded as a unit cell with a constant weight fraction of nanoparticles, while the random distribution of nanoparticles is described by a variable weight fraction among elements following a Gaussian distribution. This distribution is related to the mechanical behavior of the unit cell inside the model and it was defined based on material characterization tests for different nanoparticles weight fractions. To validate the proposed model, experimental low-velocity impact tests, at different energy levels, on composites reinforced with nanoparticles (0.5 wt.%) were carried out. The numerical method was assessed by comparing experimental results in terms of mechanical response and damage phenomena. The current numerical methodology was found to contribute to uncertainty investigation based on a mixed experimental-numerical analysis.