Identification of the mechanical properties of the carbon fiber and the interphase region based on computational micromechanics and Kriging metamodel

Computational micromechanics has been widely used to predict the mechanical properties of fiber reinforced composite. The accuracy of the prediction depends largely on the precise definition of the mechanical parameters of the constituents, the fiber, the matrix and the interphase, which has been increasingly recognized as a transition region between the fiber and the matrix. The mechanical properties of the matrix and the axial modulus of the carbon fiber can be measured directly. However, the interphase effect and other properties of the carbon fiber are difficult to be obtained through experimental methods. This paper deals with the parameter identification of the mechanical properties of the T300 carbon fiber and the interphase region by means of numerical methods. Firstly, a microstructure based three-phase finite element model in which interphase region has been simulated by cohesive elements was developed to predict the elastic properties of the unidirectional carbon fiber reinforced composite. By combining the finite element model with the Latin Hypercube Sampling method, metamodels based on Kriging method were then developed to explicitly parameterize the relationship between the elastic properties of the composite and the input parameters. The mechanical properties of the T300 carbon fiber and the interphase region of the studied composite were finally obtained by an inverse parameter identification procedure based on the constructed Kriging metamodel and the tested mechanical properties of the composite. Benefiting from the proposed procedure, more reliable input parameters were extracted for the T300 carbon fiber and the interphase modeling, resulting in more precise prediction of the elastic properties of the carbon fiber reinforced composites through finite element simulation.