▎ 摘　　要

The incorporation of graphene and its derivatives as inclusions for electrically conductive cement composites is currently under extensive investigations due to their potential applications in structural health monitoring. Experiments have shown that the electrical conductivity of the cement composites is highly dependent on the porosity and the fluid contained in the pores. Therefore, it is necessary to quantitatively predict the overall electrical conductivity of the graphene reinforced cement composites while identifying the effects of the pores and the fluid contained. Taking graphene nanoplatelets (GNPs) as inclusions, samples are prepared by wet method, and the electrical conductivity of the samples with different saturations are tested by four-probe method. Constitutive equations for electron tunneling and interface resistance are established while deter-mining the electrical conductivity of GNP reinforced cement composites (GNPRCC), in which the effects of pore and saturation are incorporated. A hybrid micromechanical model (HMM) with a two-step homogenization frame work and a functionally graded electric double layer (FGEDL) simulating saturated pores is developed for the first time. Our model shows excellent agreement with experimental data for different scenarios compared to other models. It is found for the HMM-FGEDL the enhancement of GNPs needs to be considered before incor-porating pores for dry samples, whereas pores need to be considered first for saturated ones. Setting the aspect ratio of pores to be 0.01, which is the consistent with the internal-cracks observed in experiments, is preferred while modelling the electrical conductivity of the GNPRCCs.