Prediction of the mechanical behavior of composite materials based on image processing analysis and multi-scale finite element method

In recent years, with the development of data acquisition equipment, the use of image processing methods to simulate the objects as well as the reverse engineering of different parts has been widely considered. In this regard, the main purpose of the present study is to develop a simple method based on the image processing method for finite element simulation of the composite materials consisting of several separated parts. A multi-scale finite element model has been developed for modeling composite materials using ABAQUS finite element software to study the mechanical properties of composite materials without the need for empirical tests. To evaluate the efficiency of the proposed method, concrete cylindrical specimens were first created in the laboratory, and further image processing was performed on the CT scan images of these samples. After extracting the critical points and primal sketch, a Cad model was created, and mechanical analysis was carried out using finite element software. To verify the results of the finite element method, the experimental compression test results were used on concrete specimens. The results show that the proposed model can predict the mechanical behavior of concrete. Therefore, it is possible to predict the mechanical properties of composite materials by providing a sample using only one image as input and processing it without the need for costly and time-consuming experimental tests.