Neural Model of the Permeability Features of Woven Fabrics

Abstract

A method of modeling woven fabric permeability features by means of neural nets is presented. The experiment is based on a steady-system design. A formal dependence is found, describing the relationship of the integral of a function that is a real pressure impulse formed on the woven fabric under impact air flow conditions, with its apparent density, warp twist, and weft twist. The model is assessed by verifying the structure of the neural net.

Study of Quilted Fabrics Used in Outdoor Clothing

“…Air permeability was measured according to the Polish Standard [8]. Air permeability determines the resistance of fabrics (woven, knitted and nonwoven) to the passage of air [9]. e air permeability of clothing directly a ects the exchange of gas between a human and their surroundings and, thus the physiological comfort of the clothing user.…”

A Review of Literature About the Use of Artificial Neural Networks Systems on Prediction of Clothing Comfort

“…Bu hipotez yapay sinir ağları yöntemi ile doğrulanmıştır. Ancak aralarındaki ilişkiyi görebilmek için daha detaylı araştırmalar gerektiği belirtilmektedir [26].…”

“…Şekil 4'te görülen çok katmanlı ve tek yönlü geri yayılımlı bir yapay sinir ağı modeli tasarlanmıştır. Yapay sinir ağında girdi veri seti olarak kumaş sıklığı, çözgü ipliğinin ve atkı ipliğinin bükümü, çıktı veri seti olarak ise kumaşın hava geçirgenliği kullanılmıştır [26]. Hava geçirgenliği, insan vücudundan ortama doğru giden gaz akışını ve ortamdan vücuda doğru gelen temiz hava akışını etkileyen hijyenik bir özelliktir [27].…”

Artificial Neural Networks and Their Applications in the Engineering of Fabrics

“…The prediction of bursting using ANNs for knitted fabrics (Ertugrul & Ucar, 2000) as well as for woven fabrics (Vassiliadis et al, 2010) has been achieved with satisfactory results. The permeability of the woven fabrics has been modelled using ANNs as well (Tokarska, 2004;Çay et al, 2007). Further on, the impact permeability has been studied (Tokarska & Gniotek, 2009) and the quality of the neural models has been assessed (Tokarska, 2006).…”