The choice of representative volumes in the approximation of effective properties of random materials (Q2322729)

This paper focus on analysis of selection approaches of representative volumes for representation of large scale properties with random heterogeneities on a small scale. The representative volume is a chosen sample of the random material whese effective properties are computed by a cell formula. For a fixed sample size a strategy of selection of representative volumes was suggested by [\textit{C. Le Bris} et al., Monte Carlo Methods Appl. 22, No. 1, 25--54 (2016; Zbl 1334.35450)]. In this paper the rigorous analysis of this selection approach is presented. It is studied in the context of stochastic homogenization of linear elliptic equations. In particular the method is proven to never essentially perform worse than a random selection of the material sample. Moreover it is shown that its performance may be much better if the selection criterion for the material samples is chosen suitably.