A Gaussian statistics-based unit-cell modeling method for the mechanical behavior prediction of 2.5D shallow straight-link-shaped woven composites

Abstract

The shallow straight-link-shaped structure of 2.5D woven composites (2.5D-SSLS-WC) is a novel material with a wide range of promising applications, but the randomness of its internal structure has brought challenges to performance prediction and engineering design. The gap of existing unit-cell models in capturing the internal randomness of the 2.5D-SSLS-WC material limits the accurate prediction of properties and the optimization of engineering design. In this work, the modeling method for random unit-cell model based on Gaussian distribution is proposed for 2.5D-SSLS-WC structure considering random distribution property of weft yarns and extrusion effects between structural surfaces and internal yarns during the molding process, which can more effectively reflect the practical mesoscopic structure of the material, reveal the micro mechanical behavior inside materials and provide theoretical basis for material design and performance optimization. The mechanical property and damage evolution of the proposed model is predicted and analyzed using finite element (FE) analysis and progressive damage method. The maximum errors of equivalent stiffness and strength between prediction and experimental values are 6.7% and 2.3%, respectively. The results show that the proposed model has high prediction accuracy and can reasonably reflect the damage extension and ultimate failure mode of the material during the loading process, which verifies the rationality of the proposed unit-cell modeling strategy. The modeling method proposed in this paper provides a new perspective for the in-depth understanding and description of the microstructure and mechanical behavior of 2.5D-SSLS-WC materials, which is of significant theoretical and engineering value and can be extended to the modeling of other woven composites.