Over the past years, research has attempted to relate fiber properties with yarn prop erties, and many regression equations have been developed to accomplish this. The complexities of multiple regression equations put limits on their universal acceptance. Neural networks with better nonlinear mapping have also been used to develop such relationships. Our statistical data analysis of a few yarn properties will determine the suitability of neural networks for such textile applications.
Get full access to this article
View all access options for this article.
References
1.
Baum, E.B., and Hausia, D., What Size Net Gives Valid Generalization, Neural Comput.1, 151-160 (1988).
2.
Chen, Luo, and Adams, David L., Yarn Strength Prediction Using Neural Networks, Textile Res. J.65(9), 495-500 (1995).
3.
Devijer, P.A., and Kitter, J., "Pattern Recognition: A Statistical Approach," Prentice-Hall, NY, 1982.
4.
de Villiers, Jacques, and Barnard , Etienne, Back Propagation Neural Nets with One and Two Hidden Layers, IEEE Trans. Neural Nets4 (1993).
5.
Frydrych, Iwona , A. NewApproach for Predicting Strength Properties of Yarn, Textile Res. J.62(6), 340-348 (1992).
6.
Haykins, S., "Artificial Neural Networks," MacMillan College Publishing, 1994.
7.
Mirchandani, G., Cao, W., On Hidden Nodes for Neural Nets , IEEE Trans. Ckts. Syst.36, 661-664 (1989).
8.
Sette, S., Boullart, L., Vanlangehave, L., and Kiekens, P., Optimizing the Fiber-to-yarn Prediction Process with a Combined Neural Network/Genetic Algorithm Approach, Textile Res. J.67(2), 84-92 (1997).
