Stochastic Elastic Properties of Composite Matrix Material with Random Voids Based on Radial Basis Function Network

Abstract

The stochastic characteristics of the voids result in the randomness of the equivalent elastic modulus. It is computationally expensive to quantify the stochastic characteristics of the equivalent elastic modulus using the representative volume element. This paper is focused on the stochastic elastic properties of composite material with random voids. The Karhunen–Loeve Transform is introduced to convert the random variables to a certain number of principal components; the approximate relationship is established between the principal components and the equivalent elastic modulus taking use of the radial basis function network to reduce the computational burden. The dispersion-based sampling is proposed to solve the overfitting problem. The cumulative distribution function (CDF) and probability density function (PDF) are obtained from the surrogate model, which has good agreement with the results from Monte Carlo simulations from the original model.