Comparison of the neural network model and linear regression model for predicting the intermingled yarn breaking strength and elongation


ÖZKAN İ., KUVVETLİ Y., Baykal P. D., EROL R.

JOURNAL OF THE TEXTILE INSTITUTE, vol.105, no.11, pp.1203-1211, 2014 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 105 Issue: 11
  • Publication Date: 2014
  • Doi Number: 10.1080/00405000.2014.882041
  • Journal Name: JOURNAL OF THE TEXTILE INSTITUTE
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.1203-1211
  • Keywords: feed forward neural network, intermingling, yarn elongation, yarn strength, partially oriented yarn, ROTOR SPUN YARNS, TENSILE PROPERTIES, BUSINESS, FLOW
  • Çukurova University Affiliated: Yes

Abstract

In this study, the effects of selected intermingling process parameters on yarn breaking strength and elongation were predicted using artificial neural network. For this aim, partially oriented polyester yarn with 283 dtex linear density and three different numbers of filaments (34, 68, and 100) were used for producing interlaced yarn under different process parameters (speed and pressure). Yarns' elongation and strength values measured with Uster Tensorapid test device and the number of filaments are input variables of the artificial neural networks. Feed forward neural network (FFNN) is used as the network structure. All FFNN computations were performed by MATLAB software package. The comparison results show that the FFNN has a better prediction performance than linear regression.