Abstract: Mechanical properties prediction of short fiber reinforced composites is a difficult point in engineering application. In this paper, the construction method and the performance prediction analysis of representative volume element model based on micro-CT images (CT-RVE) of fiber reinforced composites are studied. The elastic properties of carbon fiber reinforced nylon 66 (PA66CF) were predicted based on the CT-RVE model of four sizes, and the results were compared with those of the Mori-Tanaka model (M-T). The PA66CF tensile specimen with fiber fraction of 10wt% was prepared by injection molding process. The fracture surface of tensile specimen was observed by scanning electron microscope (SEM). X-ray computed tomography (CT) images of CFRP was used to obtain internal microscopic structure information. The four dimensions of CT-RVE model which were established by using ImageJ, Photoshop, Mimics, Cero, HyperMesh and ABAQUS, predicted the elastic performance of PA66CF materials. And the prediction results were compared with the CT-RVE models and M-T models which had the same geometric parameters. The results show that the random distribution of fibers in the RVE model is regular, the number of fibers and the volume of RVE are linearly increasing, the volume fraction of fibers is about 3%~5%, and the orientation of fibers is mainly in the X direction. When the RVE size is 114.4 μm×114.4 μm×77 μm, the prediction results of the CT-RVE model and the M-T model on the elastic performance were similar. When the RVE size is too large or too small, M-T model can give reasonable prediction on the elastic performance, while CT-RVE model has significantly smaller prediction results on the longitudinal elastic performance of fiber. The research results can provide guidance for the accurate prediction of the properties of short fiber reinforced composites.

Key words: short carbon fiber; microstructure; CT scan; simulation model; Mori-Tanaka model; composites