Abstract:Polymers are generally composed of randomly distributed macromolecular chains. The distribution, entanglement, and cross-linking of molecular chains significantly affect the mechanical and physical properties of the polymer. In this paper, a data-driven method was used to establish the relationship between the microstructure and mechanical properties of polymer molecular chains. Using the finite element method, the random microstructure models of two molecular chains were established and their mechanical properties were obtained. A data set was established based on the relationship between microstructure and mechanical properties, with the random molecular chain microstructure of the polymer as input, and the elastic stiffness of the polymer as the response output, to train and verify the data-driven model. We have obtained the analysis results of the microstructure-mechanical properties relationship with satisfactory accuracy. The result shows that it is reliable to study the elastic stiffness of polymers through data-driven methods.