Empirical modelling of tensile strength of woven fabrics |
| |
Authors: | Abhijit Majumdar Anindya Ghosh Shib Sankar Saha Ayan Roy Subir Barman Dhrubajyoti Panigrahi Anjan Biswas |
| |
Institution: | (1) Government College of Engineering & Textile Technology, Berhampore, West Bengal, 742101, India |
| |
Abstract: | Aesthetic properties of fabrics have been considered as the most important fabric attribute for years. However, recently there
has been a paradigm shift in the domain of textile material applications and consequently more emphasis is now being given
on the mechanical and functional properties of fabrics rather than its aesthetic appeal. Moreover, in certain woven fabrics
used for technical applications, strength is a decisive quality parameter. In this work, tensile strength of plain woven fabrics
has been predicted by using two empirical modelling methods namely artificial neural network (ANN) and linear regression.
Warp yarn strength, warp yarn elongation, ends per inch (EPI), picks per inch (PPI) and weft count (Ne) were used as input
parameters. Both the models were able to predict the fabric strength with reasonably good precision although ANN model demonstrated
higher prediction accuracy and generalization ability than the regression model. The warp yarn strength and EPI were found
to be the two most significant factors influencing fabric strength in warp direction. |
| |
Keywords: | Artificial neural network Ends per inch Fabric assistance Fabric strength Picks per inch Regression Warp yarn strength |
本文献已被 SpringerLink 等数据库收录! |
|