.png){kind=icon}
.png){kind=icon}
Return
Analysis and optimization of process parameters affecting core-spun ring yarn properties using a soft computing-based approach
.png){kind=icon}
OA
AI
H
M
S
M
DOI：10.1038/s41598-026-66672-1.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This study presents a soft computing-based framework for the analysis, modeling, and optimization of cotton/polyester core-spun yarn properties produced on a ring spinning system. The effects of pre-tension, core multi-filament linear density, twist level, and spindle speed on yarn breaking strength and cover factor were investigated using a reduced factorial design of experiments comprising 36 runs. Multiple linear regression (MLR), random forest regression (RF) and artificial neural network (ANN) models were developed and compared for predicting the response variables. The ANN models exhibited superior predictive performance, achieving higher accuracy and generalization capability than RF and MLR models, and were therefore selected for further analysis and optimization. The trained ANN models were coupled with a genetic algorithm (GA) to perform simultaneous optimization of breaking strength and cover factor. The optimal process conditions were identified as 50 cN pre-tension, 50 den core multi-filament, twist of 650 TPM, and spindle speed of 11,500 rpm. Under these conditions, breaking strength increased from 726.68 to 740.11 cN and cover factor improved from 93.72 to 96.91%, while the objective function value decreased from 0.5212 to 0.3962. Sensitivity analysis revealed that core multi-filament linear density was the most influential parameter affecting breaking strength (52.18%), whereas pre-tension had the greatest impact on cover factor (48.72%). The results demonstrate that the proposed ANN–GA framework is an effective tool for both understanding process–property relationships and optimizing the production parameters of core-spun yarns.
Keywords:
Core spun ring yarn
Tensile strength
Cover factor
Artificial neural network
Genetic algorithm
Journal
IF:
3.9
Papers:
27.1W
Citations:
83.5W
Organization
Cited Papers