Comparison of artificial neural network and linear regression models for prediction of ring spun yarn properties. I. Prediction of yarn tensile properties

In this study artificial neural network (ANN) models have been designed to predict the ring cotton yarn properties from the fiber properties measured on HVI (high volume instrument) system and the performance of ANN models have been compared with our previous statistical models based on regression analysis. Yarn count, twist and roving properties were selected as input variables as they give significant influence on yarn properties. In experimental part, a total of 180 cotton ring spun yarns were produced using 15 different blends. The four yarn counts and three twist multipliers were chosen within the range of Ne 20–35 and α _e 3.8–4.6 respectively. After measuring yarn tenacity and breaking elongation, evaluations of data were performed by using ANN. Afterwards, sensitivity analysis results and coefficient of multiple determination (R²) values of ANN and regression models were compared. Our results show that ANN is more powerful tool than the regression models.     

Prediction of cotton yarn properties using support vector machine

This paper presents a support vector machine (SVM) regression approach to forecast the properties of cotton yarns produced on ring and rotor spinning technologies from the fibre properties measured by HVI and AFIS. Prediction performance of SVM models have been compared against those of the artificial neural network (ANN) models. A k-fold cross validation technique is applied to assess the expected generalization accuracies of both SVM and ANN models. The investigation indicates that the yarn properties can be predicted with a very high degree of accuracy using SVM models and the prediction performance of SVM models are better than that of ANN models.     

Compression properties of weft knitted fabrics consisting of shrinkable and non-shrinkable acrylic fibers

High-bulk worsted yarns with different shrinkable and non-shrinkable acrylic fibers blend ratios are produced and then single jersey weft knitted fabrics with three different structures and loop lengths are constructed. The physical properties of produced yarns and compression properties of produced fabrics at eight pressure values (50, 100, 200, 500, 1000, 1500 and 2000 g/cm²) were measured using a conventional fabric thickness tester. Then, weft-knitted fabric compression behavior was analyzed using a two parameters model. It is found that at 40% shrinkable fibre blending ratio the maximum yarn bulk, shrinkage, abrasion resistance and minimum yarn strength are obtained. It is also shown that high-bulk acrylic yarn has the highest elongation at 20% shrinkable fibre blend ratio. The statistical regression analysis revealed that the compression behavior of acrylic weft-knitted fabrics is highly closed to two parameter model proposed for woven fabrics. It is also shown that for weft-knitted structure, there is an incompressible layer (V′) which resists against high compression load. Acrylic weft-knitted fabrics with knit-tuck structure exhibit higher compression rigidity and lower softness than the plain and knitmiss structures. In addition, at 20% shrinkable fibre blend ratio, the high-bulk acrylic weft-knitted fabrics are highly compressible.     

A comparative study on performance properties of yarns and knitted fabrics made of biodegradable and conventional fibers

Biodegradable products are parts of a natural cycle. The biopolymers and the fibers that can be produced from them are very attractive on the market because of the positive human perception. Therefore, PLA being a well known biodegradable fiber and some conventional fibers were selected for the current study to examine the differences between them and to emphasize the importance of biodegradability beside fabric performance. 14.8 tex (Ne 40/1) combed ring spun yarns produced from biodegradable fiber PLA, new generation regenerated fibers Modal and Tencel, synthetic and blends 50/ 50 % cotton/polyester and 50/50 % viscose/polyester, polyester were selected as yarn types and by using these yarns, six knitted fabrics were produced and some important yarn and fabric properties were compared. In this context, moisture and the tensile behavior of yarns and pilling, bursting strength, air permeability and moisture management properties of the test fabrics are discussed.     

Dimensional and physical properties of plain knitted fabrics made from 50/50 bamboo/cotton blended yarns

In this study, the dimensional and some physical properties of plain knitted fabrics made from 50/50 bamboo/cotton blended yarns are investigated. In order to see the differences and similarities, the results are then compared with those for similar fabrics knitted from 50/50 conventional viscose/cotton and 50/50 modal/cotton blended yarns. Each fabric type was produced with three different stitch lengths. After all fabrics were dyed under identical dyeing conditions, they were subjected to dry and full relaxation treatments. For dimensional properties of fabrics, course, wale and stitch densities were measured. Then, by calculating statistically best-fit lines passing both through the experimental points and the origin, dimensional constants i.e. k values were predicted in terms of the fiber types. The result show that each fabric type knitted from bamboo/cotton, viscose/cotton and modal/cotton blended yarns behaves in a similar manner. However, in both dry and fully relaxed states, the modal/cotton knitted fabrics tend to have slightly higher k values than the bamboo/cotton and viscose/cotton knitted fabrics. For physical properties, fabric weight per unit area, thickness, bursting strength, air permeability and pilling were evaluated. The results show that the weight, thickness and air permeability values are independent of the fiber type. Plain knitted fabrics from modal/cotton blended yarns have the highest bursting strength values. Plain knitted fabrics from bamboo/cotton blended yarns tend to pill less.     

A comparative study of finer conventional and modified cotton yarns and their resultant woven fabrics

A modified ring spinning technique has been recently developed by incorporating false twisting devices into the conventional ring frame. Its application on the coarser yarn counts (7–32 Ne) showed notable advantages in modified yarn and fabric performance. More recently, it was noted that this technique can also be applied for producing finer cotton yarns. Thus this paper aims to carry out a systematic study of the physical properties of the finer modified yarns (80 Ne) and woven fabrics with respect to the conventional ones. Physical properties of conventional and modified single yarns were evaluated and compared. These two types of single yarn were used for the production of woven fabrics. Moreover, the above two types of single yarn were also plied and used for the production of woven fabrics under a commercial condition. All woven fabrics were assessed in terms of fabric tensile strength, tearing strength, abrasion resistance, fabric weight, and air-permeability as well as other fabric performance measured by the Kawabata Evaluation System (KES). Experimental results showed that finer modified yarns and fabrics exhibit higher strength, lower hairiness, and improved abrasion resistance, slightly better compression property, and smoother surface with relatively larger thickness.     

Modelling of ring yarn unevenness by soft computing approach

This paper demonstrates the application of two soft computing approaches namely artificial neural network (ANN) and neural-fuzzy system to forecast the unevenness of ring spun yarns. The cotton fiber properties measured by advanced fiber information system (AFIS) and yarn count have been used as inputs. The prediction accuracy of the ANN and neural-fuzzy models was compared with that of linear regression model. It was found that the prediction performance was very good for all the three models although ANN and neural-fuzzy models seem to have some edge over the linear regression model. The linguistic rules developed by the neural-fuzzy system unearth the role of input variables on the yarn unevenness.     

Using artificial neural network to predict colour properties of laser-treated 100% cotton fabric

In this paper, artificial neural network (ANN) model was used for predicting colour properties of 100 % cotton fabrics, including colour yield (in terms of K/S value) and CIE L, a, and b values, under the influence of laser engraving process with various combination of laser processing parameters. Variables examined in the ANN model included fibre composition, fabric density (warp and weft direction), mass of fabric, fabric thickness and linear density of yarn (warp and weft direction). The ANN model was compared with a linear regression model where the ANN model produced superior results in prediction of colour properties of laser engraved 100 % cotton fabrics. The relative importance of the examined factors influencing colour properties was also investigated. The analysis revealed that laser processing parameters played an important role in affecting the colour properties of the treated 100 % cotton fabrics.     

Investigation of air permeability of core spun cotton/spandex weft knitted structures under relaxation

In this research work, air permeability variations of core spun cotton/spandex single jersey and 1×1 rib knitted structures were studied under relaxation treatments. Results are compared with similar fabrics made from 100 % cotton material. Even though cotton/spandex fabrics knitted with same stitch lengths, their structural spacing and stitch densities vary with the progression of treatments. Similar behavior was also observed with 100 % cotton knitted structures. Under higher machine set stitch lengths (i.e., lower fabric tightness factor), higher structural spacing and lower stitch densities were resulted and those variations significantly affected on the air permeability variations of knitted structures. 1×1 rib knitted structures showed significantly higher air permeability than single jersey structures and it is more prominent with cotton rib structures. However, cotton/spandex 1×1 rib and single jersey structures have not showed such significant deviations. Air permeability of cotton/spandex and 100 % cotton rib and single jersey knitted structures decreased with lower machine set stitch lengths (i.e., at higher fabric tightness factors). There was a correlation with fabric tightness, air permeability, areal density and fabric thickness such as knitted fabrics became tighter, their weight and thickness were higher, while their air permeability was lower. Thus, fabric areal density and fabric thickness are positively correlates to machine set stitch length^?1 (fabric tightness factor). Air permeability of a knitted structure depends on material type, knitted structure, stitch length, relaxation treatment, structural spacing and stitch density.     

The effects of some machine parameters on the spirality in single jersey fabrics

In this work, the effects of machine parameters on the fabric spirality, which is an important quality problem of single jersey knitted fabrics, are investigated. For this aim, two circular knitting machines with the same gauge, but one of them revolving in the reverse direction, are chosen. Single jersey fabric samples with the same weight per square meter and the same yarn count (Ne 20 Cotton) are knitted on the chosen machines at four different numbers of knitting systems. The effects of the number of the knitting systems and the rotation directions of the machines on the spirality angles are investigated.     

Analysis on the pilling factors of cashmere knitted fabric

The effect of cashmere yarn twist, knitted fabric density, and cashmere properties on pilling rates of cashmere knitted fabric is investigated in this paper. The experimental results show that yarn twist and fabric density have little influence on pilling rates of cashmere knitted fabric for yarn 38.4 tex/2 when yarn twist varies from 234 T/m to 272 T/m, and the fabric density is 9.7, 10.7, and 11.2 yarns/inch, respectively. The length of cashmere fiber, in particular less than 7.5 mm, is responsible for the pilling rates of cashmere knitted fabric based on optimal scaling regression analysis.     

A comparative study of cotton knitted fabrics and garments produced by the modified low twist and conventional ring yarns

Spirality is one of the major potential problems in knitted fabrics and garments. It affects the aesthetics and physical properties of the garment produced, such as the seam displacement, shape retention, pattern distortion and sewing difficulties. In this paper, a comparative study has been carried out to evaluate the physical performance of 100 % cotton knitted fabrics and garments produced by the modified low twist and conventional ring yarns through the actual wearing and washing trials. Experimental results showed that the properties of side seam displacement, fabric spirality, dimensional stability and skewness change of the T-shirts and sweaters made by the modified single yarns are comparable to those of garments made from the control plied yarns but much improved when compared to those from the control single yarns. In addition, the pilling resistance and bursting strength of the knitted fabrics made by the modified single yarns can still maintain a reasonably high level at a low yarn twist.     

Spinning performance and antibacterial activity of SeaCell® active/cotton blended rotor yarns

Microorganisms can lead to functional, hygienic and aesthetic (e.g. deterioration, staining) problems on textile products. Natural fibers especially cotton are more easily affected by microorganisms. Blending of cotton fibers with antimicrobial fibers can enhance the protective properties of products against microorganisms. Demand of antimicrobial performance from the products changes depending on the application area. Therefore determination of suitable antimicrobial fiber quantity for the desired application is important. In this study the spinning performance of SeaCell Active/cotton blended open end rotor yarns and antibacterial activities of fabrics produced by these blended yarns were investigated. Five different cotton/SeaCell Active blended slivers with SeaCell Active content from 3 % up to 53 % were prepared on drawframe machine and all slivers were spun into yarns on open end rotor spinning machine at a yarn count of 20 tex with α_Tt=3827 twist coefficient. The effects of rotor speed, opening roller speed, rotor, opening roller and navel type on the quality parameters of SeaCell Active/cotton blended yarns were investigated. Tensile properties, hairiness, unevenness and IPI values of the yarns were reported. All types of cotton/SeaCell Active blended yarns were knitted on a circular knitting machine. Antibacterial activity of the fabrics was analyzed quantitatively. Antibacterial tests showed that good antibacterial activity can be achieved after several washings even with 3 % of SeaCell Active fibers in fabrics.     

Evaluation of the structural properties of plain fabrics woven from various fibers using Peirce’s model

Peirce’s fabric model has been widely used to predict the structural behavior of various plain woven fabrics. The structure of plain woven fabric can be defined in terms of the warp yarn number, weft yarn number, warp fabric density, weft fabric density, warp crimp, and weft crimp. The warp and weft yarn diameters are calculated from the warp and weft yarn numbers, and the effective coefficient of the yarn diameter is defined by using this model. We have investigated structural properties, such as the effective coefficient of the yarn diameter, yarn crimp, and fabric thickness for two different fabrics in which the constituent yarns are assumed to be either incompressible or compressible. This model is also applied to various plain fabrics woven from cotton, rayon, wool, linen, nylon, acetate, polyester, and silk yarns.     

Characterizing the tensile properties of spun yarns using bivariate normal distribution

The tensile properties of spun yarns decisively influence its performance in various mechanical processing stages. This study is primarily aimed at simultaneous analysis of two tensile properties of spun yarns namely tenacity and breaking strain, which play crucial role in determining the frequency of warping breaks. The threshold values of yarn tenacity and breaking strain required for 20’s Ne carded cotton yarn to sustain the imposed stresses and strains during warping process have been determined using a bivariate normal distribution model. This study opens up the possibility of minimizing end breakage rate in various manufacturing processes of textile industry by engineering of spun yarns devoid of potential weak spots which are responsible for breaks.     

Flexural stiffness of core spun cotton/spandex weft knitted structures under relaxation and machine washing treatments

In this research work, behavior of flexural stiffness of core spun cotton spandex single jersey, 1x1 rib and interlock fabrics was studied under relaxation and machine washing treatments. Results are compared with similar fabrics made from 100 % cotton. Fabric weight density increased with the progression of treatments and it is proportionate to the fabric tightness factor (stitch length^?1). Even though both types of fabrics had same machine set stitch lengths, cotton/spandex fabrics have shown the higher fabric weight densities than that of 100 % cotton fabrics. Although 1x1 rib and single jersey fabrics knitted with the same machine set stitch lengths, rib fabrics have given higher fabric weight densities than single jersey fabrics. Among the three knitted structures, interlock fabrics with higher machine set stitch lengths gave the higher fabric weights. Fabric stiffness and flexural rigidity have given higher values under the progression of treatments and it was found that higher values of stiffness have given by cotton/spandex knitted fabrics compared to their cotton fabrics. Fabric stiffness and flexural rigidity in wale direction were higher than that in course direction, but it is only observed in single jersey fabrics. However, 1x1 rib and interlock fabrics have shown an opposite behavior. It was also observed a positive correlation between TF (i.e.: stitch length^?1) and bending length/flexural rigidity in both fabric types. Lower flexural rigidities reported with single jersey structures and highest values gave with interlock structures of cotton/spandex and cotton fabrics.     

Manufacture technique and electrical properties evaluation of bamboo charcoal polyester/stainless steel complex yarn and knitted fabrics

There are derivative problems of electromagnetic wave radiation accompanying the advances of science and technology nowadays and secure protections are also emphasized gradually. To shield these electromagnetic wave radition jeopardizing people’s health, in this study, stainless steel wires were the core yarn and bamboo charcoal polyester textured yarns were the wrapped yarn. The bamboo charcoal polyester/stainless steel (BC/SS) complex yarns were manufactured using a rotor twister machine. The BC/SS complex knitted fabrics were woven with the complex yarns employing a circular knitting machine. Three manufacture parameters were the wrapped amount of the complex yarn (2 to 6 turns/cm), the lamination amount of the knitted fabrics (1 to 6 layers) and lamination angles of the knitted fabrics (0°/0°/0°/0°/0°/0°, 0°/45°/90°/−45°/0°/45°, and 0°/90°/0°/90°/0°/90°). The knitted fabric exhibited the lowest surface resistance 32.3 Ω/sq. Optimum electromagnetic shielding effectiveness (EMSE) was 45 dB when the knitted fabrics were with 0°/45°/90°/−45°/0°/45° laminating in 0.51 GHz.     

Color and whiteness properties of fabrics knitted from different hybrid core-spun yarns containing metal wire

In this study, the color and whiteness properties of fabrics knitted from ring, siro and compact core-spun yarns containing metal wire were investigated. In general, an increase in the metal ratio of yarns causes a decrease in the whiteness and color strengths of fabrics. In our study, spinning method had no statistically significant effect on either the T _w or K/S values of fabrics while the effect of spinning method on the whiteness index of fabrics was briefly observed. Fabrics knitted from siro spun yarns showed higher whiteness properties than those of fabrics knitted from compact and ring spun yarns. This might be explained by the superior covering effectiveness of the siro spinning method on metal wire.     

The Effects of Fabric and Conductive Wire Properties on Electromagnetic Shielding Effectiveness and Surface Resistivity of Interlock Knitted Fabrics

Our aim in this study was to investigate the effects of course density, yarn linear density and thickness and type of conductive wire on electromagnetic shielding effectiveness. Metal/cotton conductive composite yarns were produced by the core-spun technique on the ring spinning machine, involving stainless steel, copper and silver coated copper wires with 40 μm, 50 μm, 60 μm thicknesses and Ne10/1 and Ne20/1 count yarns. The interlock fabrics were knitted on a 7G flat knitting machine with the three different machine settings. The EMSE and the surface resistivity of knitted fabrics were measured by the co-axial transmission line method according to the ASTM-D4935-10 standard in the frequency range from 15 to 3000 MHz and by the ASTM D257-07 standard, respectively. It was observed that all fabrics shielded around 95 % of electromagnetic waves at low frequencies, 80 % at medium frequencies and 70 % at high frequencies. Increasing the course density and thickness of conductive wire in interlock knitted fabrics increased the EMSE correspondingly. The knitted fabrics that had been produced with high yarn count showed greater EMSE because there was less isolation. The effect of the metal wire type was highly significant between 15 and 600 MHz.