A comparison of homogenization and standard mechanics analyses for periodic porous composites (Q1194781)

The paper is devoted to the comparison of various determinations of the effective elastic properties of composite materials. All the methods are based on the concept of a representative volume element (RVE) and estimate the effective elastic moduli when the ratio \(\eta\) of the RVE size to the size of the global region of interest goes to zero. The paper gives in first a review of the RVE methods for composite elastic materials, and the theoretical backgrounds of the standard mechanics approach and of the homogenization theory for periodic structures. It is proved that the two methods are equivalent for a restricted class of one- dimensional periodic composites. The main purpose of the paper is to compare the homogenization and standard mechanics estimates of effective stiffness and local strain energy density distributions for two- dimensional linear elastic porous materials with periodic microstructures. Both methods are implemented using a displacement based finite element formulation. It appears that the effective stiffness predicted by standard mechanics approaches converges towards the homogenization effective stiffness as \(\eta\to 0\). Furthermore, homogenization estimates of local strain energy density are within \(30\%\) of direct analyses while standard mechanics approaches generally differ from direct analyses by more than \(70\%\). These results indicate that homogenization theory is preferable over standard mechanics approaches for analyzing locally periodic composites.