An investigation into the engineering of the drapability of fabric

George K. Stylios, Norman J. Powell, Lu Cheng

Research output: Contribution to journalArticle

Abstract

The drape attributes of fabrics, number of folds, depth of folds and evenness of folds were measured together with the drape coefficient. The relationship between these measurements and the subjective evaluation of the fabric drape were modelled for each end use on a neural network using back propagation, which can correctly predict the grades of 90% of the samples. The relationship between the drape attributes and fabric bending, shear and weight was also modelled using neural networks. It was found that using the natural logarithm of the material property divided first by the weight of the fabric produced the most predictive model. Together, these models provide a powerful predictive tool to determine both the drape attributes and the drape grade from the mechanical properties of a fabric. The accuracy of the prediction of this system was found to be 83% overall. Combining this with a novel feedback system (Stylios and Cheng, in preparation), the drape grade or drape attributes of a fabric can be modified to fit customer requirements and then the changes to the material properties required to achieve them can be determined.

Original languageEnglish
Pages (from-to)33-50
Number of pages18
JournalTransactions of the Institute of Measurement and Control
Volume24
Issue number1
DOIs
Publication statusPublished - 2002

Fingerprint

Materials properties
Neural networks
Backpropagation
Feedback
Mechanical properties

Keywords

  • Aesthetic drape
  • Drape meter
  • Material properties

Cite this

@article{85b26b64678e4fd68a6e5e4ec8cc02e6,
title = "An investigation into the engineering of the drapability of fabric",
abstract = "The drape attributes of fabrics, number of folds, depth of folds and evenness of folds were measured together with the drape coefficient. The relationship between these measurements and the subjective evaluation of the fabric drape were modelled for each end use on a neural network using back propagation, which can correctly predict the grades of 90{\%} of the samples. The relationship between the drape attributes and fabric bending, shear and weight was also modelled using neural networks. It was found that using the natural logarithm of the material property divided first by the weight of the fabric produced the most predictive model. Together, these models provide a powerful predictive tool to determine both the drape attributes and the drape grade from the mechanical properties of a fabric. The accuracy of the prediction of this system was found to be 83{\%} overall. Combining this with a novel feedback system (Stylios and Cheng, in preparation), the drape grade or drape attributes of a fabric can be modified to fit customer requirements and then the changes to the material properties required to achieve them can be determined.",
keywords = "Aesthetic drape, Drape meter, Material properties",
author = "Stylios, {George K.} and Powell, {Norman J.} and Lu Cheng",
year = "2002",
doi = "10.1191/0142331202tm049oa",
language = "English",
volume = "24",
pages = "33--50",
journal = "Transactions of the Institute of Measurement and Control",
issn = "0142-3312",
publisher = "SAGE Publications Ltd",
number = "1",

}

An investigation into the engineering of the drapability of fabric. / Stylios, George K.; Powell, Norman J.; Cheng, Lu.

In: Transactions of the Institute of Measurement and Control, Vol. 24, No. 1, 2002, p. 33-50.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An investigation into the engineering of the drapability of fabric

AU - Stylios, George K.

AU - Powell, Norman J.

AU - Cheng, Lu

PY - 2002

Y1 - 2002

N2 - The drape attributes of fabrics, number of folds, depth of folds and evenness of folds were measured together with the drape coefficient. The relationship between these measurements and the subjective evaluation of the fabric drape were modelled for each end use on a neural network using back propagation, which can correctly predict the grades of 90% of the samples. The relationship between the drape attributes and fabric bending, shear and weight was also modelled using neural networks. It was found that using the natural logarithm of the material property divided first by the weight of the fabric produced the most predictive model. Together, these models provide a powerful predictive tool to determine both the drape attributes and the drape grade from the mechanical properties of a fabric. The accuracy of the prediction of this system was found to be 83% overall. Combining this with a novel feedback system (Stylios and Cheng, in preparation), the drape grade or drape attributes of a fabric can be modified to fit customer requirements and then the changes to the material properties required to achieve them can be determined.

AB - The drape attributes of fabrics, number of folds, depth of folds and evenness of folds were measured together with the drape coefficient. The relationship between these measurements and the subjective evaluation of the fabric drape were modelled for each end use on a neural network using back propagation, which can correctly predict the grades of 90% of the samples. The relationship between the drape attributes and fabric bending, shear and weight was also modelled using neural networks. It was found that using the natural logarithm of the material property divided first by the weight of the fabric produced the most predictive model. Together, these models provide a powerful predictive tool to determine both the drape attributes and the drape grade from the mechanical properties of a fabric. The accuracy of the prediction of this system was found to be 83% overall. Combining this with a novel feedback system (Stylios and Cheng, in preparation), the drape grade or drape attributes of a fabric can be modified to fit customer requirements and then the changes to the material properties required to achieve them can be determined.

KW - Aesthetic drape

KW - Drape meter

KW - Material properties

UR - http://www.scopus.com/inward/record.url?scp=0036221772&partnerID=8YFLogxK

U2 - 10.1191/0142331202tm049oa

DO - 10.1191/0142331202tm049oa

M3 - Article

VL - 24

SP - 33

EP - 50

JO - Transactions of the Institute of Measurement and Control

JF - Transactions of the Institute of Measurement and Control

SN - 0142-3312

IS - 1

ER -