Optimization of cotton two-ply yarn parameters to enhance tensile strength using the grey wolf optimization algorithm

Optimizing the structural parameters of two-ply yarn is essential for achieving superior mechanical performance in textile applications. In this study, an intelligent hybrid framework combining Artificial Neural Networks (ANN) with the Grey Wolf Optimizer (GWO) was developed to systematically optimize cotton two-ply yarn parameters with the aim of maximizing tenacity. Comparative modeling results demonstrated that the ANN significantly outperformed the Multiple Linear Regression (MLR) model in terms of predictive accuracy and generalization ability, as confirmed by higher TGF values, thereby validating its suitability for integration within the ANN-GWO optimization scheme. The optimization process identified the optimal configuration as 1000 TPM twist for both single yarns, 790 TPM for the two-ply yarn twist, and a Z twist direction. Under these conditions, the cost function value improved markedly from - 0.4876 to - 0.9814, while tenacity increased substantially from 28.72 cN/tex to 35.81 cN/tex. Sensitivity analysis further revealed that the twist direction of the two-ply yarn was the most influential parameter, accounting for 32.61% of the variation in tenacity. Overall, the proposed ANN-GWO framework provides a robust, data-driven, and industry-relevant approach for the precise regulation and performance enhancement of two-ply yarns.