Prediction on macroscopic elastic properties of interphase-contained long-fiber-reinforced composites and multiple nonlinear regression analysis

Abstract

A convenient method to predict the macroscopic elastic performance of composite containing interphase was developed in this paper based on a new modified random sequential adsorption method and multiple nonlinear regression analysis. First, a three-phase micromechanical model with randomly distributed fibers was established with a new modified random sequential adsorption method named the Moving Window Method. Second, the macroscopic elastic behaviors of T300/914C were predicted based on micromechanical parameters of constituent materials using energy method, and the influence of the randomness of fiber distribution on the prediction was analyzed. Third, the effects of interphase thickness, interphase modulus and Poisson’s ratio on the macroscopic elastic properties of the composite were studied by changing these interphase micromechanical parameters gradually within certain ranges. Finally, the multiple nonlinear regression models that describe the relationship between the macroscopic elastic properties of T300/914C and micromechanical characteristic parameters of the interphase were established using a limited number of data from numerical simulation. Results indicate that the relative errors for the longitudinal modulus and the major Poisson's ratio are within ±1% while for the transverse modulus, shear modulus and Poisson’s ratio at cross section, the relative errors are within 5%.