A Viscoelastic-Plastic Model for the Core of Various Close- Packings of Multifilament Polyamide-6 Yarns

Abstract

The number of fibers in the cross-section of a yarn determines its physical and mechanical properties. Among the parameters that affect fiber placement (packing) in the yarn core structure, the number of fibers, the number of twists, and variations in fiber length in various layers of yarn during the fabrication process are of prime importance. Theoretically, yarn formation is associated with two different packings: open-packing and close-packing with a single fiber in the core. The packing of fibers has been modeled using various approaches, including analytical, finite element, and experimental methods. In this study, an analytical approach combining theoretical methods and artificial neural networks (ANNs) is proposed to predict the tensile behavior of yarns with different numbers of fibers in the core regions (ranging from 2 to 5 fibers). The novelty of this method lies in its ability to predict the time-dependent stress-strain curve of multifilament yarns composed of monofilaments with circular cross-sections. The proposed method extends beyond the linear region, incorporating both structural and material properties. It leverages the strong mapping capabilities of ANNs rather than relying solely on elastic models. Validation of the method showed excellent agreement between experimental and theoretical values. Numerical simulation indicated results consistent with the theoretical ones, proving the model's precision in predicting the tensile behavior of multifilament yarns. This modeling procedure has the potential to revolutionize the approach to modeling textile structures.