Prediction of polyester/cotton blended rotor-spun yarns hairiness based on the machine parameters

19-25Effect of rotor type, rotor diameter, doffing-tube nozzle, and torque-stop on polyester/cotton rotor-spun yarn hairiness have been studied. To model the hairiness of polyester/cotton blended yarn, artificial neural networks and regression models have been used. The results show that there are significant differences in performance of network with different architectures and training algorithms. The network with two hidden layers has the best performance and can predict hairiness with high accuracy. Relative importance of input variables is studied with partial derivatives method based on the optimum network. The results indicate that rotor type and rotor diameter have the greatest and least effect on the blended yarn hairiness