Abstract: In the assumption at mesoscopic level, concrete materials are assumed as three-phase composites consisting of aggregate, mortar and the bonding interface between mortar and aggregate. In order to determine elastic modulus of concrete (EMC), on the basis of the assumption, we establish a random aggregate model in the two-dimensional plane with Monte Carlo method and Walraven formula. Meanwhile, we predict EMC by using finite element method and compare the numerical calculated results with test results to verify the effectiveness of this mesoscopic finite element model. Furthermore, we discussthe impacts of several parameters (elastic modulus, aggregate’s volume content , aggregate’s maximum size, aggregate’s gradation , interface thickness and pores) of mesoscopic component on EMC. Test results show that 1) as for mesoscopic components of concrete, impact of elastic modulus of mortar on EMC is the biggest; 2) under given conditions, EMC with continuous gradation is bigger than that with uncontinuous gradation; 3) EMC decreases with the increasing of interface thickness and the existence of pores. The research results offer reference for the design of concrete’s mix proportion and optimization of its mechanical properties.

Key words: concrete, composite material, random aggregate model, elastic modulus, prediction