Applications of Neural Networks in the Hairiness of Ring Spinning Polyester/Cotton Yarn Prediction

Abstract

In this paper, back-propagation (BP) neural network model is introduced and established. This work describes the application of the BP artificial neural network model for the purpose of predicting the polyester/cotton yarn hairiness. This approach has been developed and evaluated with the use of multiple sets of data, comprising of a range of processing parameters. The yarn hairiness of ring spinning is strongly related to the processing parameters. However, it is difficult to establish physical models on the relationship between the processing parameters and the yarn hairiness. Due to the artificial neural network can fully approximate any complex nonlinear system and study dynamic behavior of any serious undetermined system. It has a highly parallel calculation ability, strong robustness and fault tolerance. So using the artificial neural network to predict the polyester/cotton yarn hairiness of ring spinning is a very effective way. The experimental results and corresponding analysis show that the BP neural network model is an efficient technique for the yarn hairiness of ring spinning prediction and has wide prospect in the application of ring spinning yarn production system.