Abstract

Predicting pull-out force of loop pile of woven terry fabrics using artificial neural network algorithm In this study, the capability of artificial neural network (ANN) method based on scale conjugate gradient learning algorithm for modeling the pull-out force of pile yarn in terry towel was investigated. Accordingly, 24 different samples were produced. 21 samples were selected as training set and 3 ones as testing set. The developed models were assessed by verifying mean square error (MSE) of prediction of testing data. The obtained results reveal that the artificial neural network model is very effective for predicting the pull-out force of pile yarn of terry fabrics based on process parameters. The R2 value of artificial neural network model was 0.998. The relative importance of process parameters on pull-out force of pile loop has also been investigated. The count of pile yarn was found as the major contributing parameter.

Key-words: terry fabric, pull-out force, artificial neural network, scale conjugate gradient