Predictıve Modelıng of Yarn Quality at Ring Spinning Machine using Resilient Back Propogation Neural Networks

Abstract

The final attenuation and twisting of fiber take place at ring spinning machine and hence its optimized performance is very crucial in terms of yarn quality. Drafting at ring spinning machine has a decisive effect on quality. There exist many influencing parameters in the spinning geometry that have to be optimized for manufacturing of quality yarn. The present research work was carried out to develop the Artificial neural networks (ANN) based prediction model for the polyester/cotton blended ring spun yarns by using these influencing parameters as inputs. ANN prediction model was developed using resilient backpropogation algorithm. Yarn quality parameters like yarn evenness, hairiness and tensile parameters were predicted. The low mean absolute error values for the yarn quality parameters proved that it is possible to predict the yarn quality on the basis of spinning geometry for cotton/polyester blended ring spun yarns using Resilient Back Propogation Neural Networks.