Application of Neural Network and Physical-mathematical Models for the Prediction of Fiber Diameter of Polyethylene Terephthalate (PET)Spunbonding Nonwovens Fabric
In this study,two models are developed and used to predict the fiber diameter of spunbonding nonwoven fabrics.Ten different spunbondlng nonwoven fabrics are manufactured with ten different parameters in spunbonding nonwoven machine.A back propagation multi layer perception (MLP) network and a physical-mathematical model are developed to predict the fiber diameter of spunbonding nonwoven fabrics. Both ANN and physical-mathematical model have given predictions. A physical-mathematicai model provides a useful insight into the air jet flow characteristics,air drawing model and formation processing of fibers in the drafting assembly of a spunbonding machine. By analyzing the results of the physicalmathematical model, the effects of the process parameters on fiber diameter can be predicted. An artificial neural network (ANN) model provides quantitative predictions of fiber diameter. The effects of parameters on fiber diameter are also determined by the neural network model However, the results show that the predictions of artificial neural network models give more reliable results than physical-mathematical model.
spunbonding machine diameter parameter physical-mathematical model artificial neural network model
Bo Zhao
College of Textiles Zhongyuan University of Technology Henan,Zhengzhou,450007,Peoples Republic of China
国际会议
The 10th International Conference on Intelligent Technologies(第十届智慧科技国际会议 InTech09)
桂林
英文
176-181
2009-12-12（万方平台首次上网日期，不代表论文的发表时间）