Part 1. Predicting the Pilling Tendency of the Cotton Interlock Knitted Fabrics by Regression Analysis

摘要

This study, it was aimed to determine the equations and models for estimating the pilling propensity of interlock knitted fabrics. Seven different cotton blends supplied from different spinning mills, yarns in 3 different yarn counts (Ne 20, Ne 30 and Ne 40) and in 3 different twist coefficients (αe=3.2, 3.6 and 4.0) were produced. Interlock knitted fabrics were manufactured in three different fabric tightness values from each of the produced yarns. The pilling tendencies of the fabrics were tested according to EN ISO 12945-2 standard by a Martindale pilling and abrasion device. The PillGrade Objective Pilling Grading System, based on the image analysis principle, was used for evaluating the pilling propensity of the fabrics. By using this system, the pilling degree of the fabrics, total pill number, total weighted pill number, average pill area, and average pill height of the fabrics were measured. Fiber features determined by an AFIS PRO 2 instrument with the samples taken from the cotton roving were used as independent variables for the regression analysis. Moreover, yarn unevenness, yarn twist, yarn count, yarn hairiness, and fabric cover factor values were included in the equations as independent variables; and by considering each of the pilling features measured by PillGrade as a dependent variable, multivariate linear regression equations were determined and the availability of the equations was investigated in detail statistical analyses.