Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: właściwości fizyczne przędzy

Sortuj według:

Ogranicz wyniki do:

Sztuczne sieci neuronowe i kierunki ich zastosowań w przemyśle. Przykłady zastosowań sztucznych sieci neuronowych wykorzystanych do poprawy jakości procesów przędzalniczych

Lewandowski S., Drobina R., Grecka A.

Przegląd Włókienniczy - Włókno, Odzież, Skóra

2016

nr 9

32--36

Artificial neural networks have been used in all stages of the manufacturing process of textiles, from fibers, and even at the stage of forming a fiber-forming polymers, starting and ending with finished products. This article presents som examples ofapplications of artificial neural networks used to improve the qualit of spinning processes. Artificial neural multilayer perceptron type learned usin a back propagation algorithm and the algorithm of Marquardt are inter alii to predict the course of the spinning process as well as predicting the physicc properties of yarns, ensuring sufficient accuracy.

Regressional Estimation of Cotton Sirospun Yarn Properties from Fibre Properties

Bedez Üte T., Kadoğlu H.

Autex Research Journal

2014

Vol. 14, no. 3

161--167

In this paper, it is aimed at determining the equations and models for estimating the sirospun yarn quality characteristics from the yarn production parameters and cotton fibre properties, which are focused on fibre bundle measurements represented by HVI (high volume instrument). For this purpose, a total of 270 sirospun yarn samples were produced on the same ring spinning machine under the same conditions at Ege University, by using 11 different cotton blends and three different strand spacing settings, in four different yarn counts and in three different twist coefficients. The sirospun yarn and cotton fibre property interactions were investigated by correlation analysis. For the prediction of yarn quality characteristics, multivariate linear regression methods were performed. As a result of the study, equations were generated for the prediction of yarn tenacity, breaking elongation, unevenness and hairiness by using fibre and yarn properties. After the goodness of fit statistics, very large determination coefficients (R2 and adjusted R2) were observed.