期刊
JOURNAL OF THE TEXTILE INSTITUTE
卷 97, 期 2, 页码 129-136出版社
TEXTILE INST
DOI: 10.1533/joti.2005.0135
关键词
artificial neural network; pilling prediction; wool; knitted fabrics
This work investigates the application of artificial neural network modeling (ANN) to model the relationships between fiber, yarn, and fabric properties and the pilling propensity of single jersey and rib pure wool knitted fabrics based on the ICI Pilling Box method. Validation of the model on an independent validation data set suggests that the accurate prediction of pilling propensity is possible with the best performing model achieving a correlation with the subjectively rated pilling grades of approximately 85%. Importantly, it is also illustrated that a larger training set can lead to a marked improvement in the accuracy of predictions.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据