Indian Journal of Fibre & Textile Research Vol. 31. December 2006, pp. 50 1 -506
Predicting abrasion behaviour of chenille fabric by fu zzy logic
E K (even & b bzdemir" Textile Engineering Department, Faculty of Engineering and Architecture, Uludag University. Gorukle, Bursa 16059, Turkey
and
L Dagkurs Vocational School of Textile, Gazioslllanpa�a University. Ta�IJ(,:iftlik. Toka! Turkey 60240. Received Septelllber 2005; revised received Decelllber 2005; accepted 2 Fe/Jrt/{//Y2006 29 26
The effects of chcnille yarn count. pile length and yarn twist level on abrasion resistance of chenille fabric have been studied using fuzzy logic system. Different chenille yarns arc produced with varying yarn count, pile length and twist levcl on a chenille yarn machine. The viscose and acrylic have been used as pile and core yarns respectively and the fabrics arc woven from these yarns by using them as fi lling yarns in the weaving construction. Abrasion rcsistancc of chen ille fabrics is measured with a Mart indale abrasion tester. Experimental data arc used in the establishment of the fu zzy logic model and the construction of basic principles. It is observed that the use of high twist levels and pile lengths brings about an 0.978) improvement in abrasion resistance and the yarn count has a significant efrect on mass loss. Correlation analysis (r = confirms strong linear relationship between measured :l Ild predicted mass loss values. It is practically possible to obtain positive results wilh ahrasion resistance optimization in a more economical way by fuzzy logic system.
Keywol'(ls: Abra,llIn resistance, Chenille yarn, Fuzzy logic, Pile length. Twist level, Yarn count
Int. CLK D02G3/42. D03D27/00, G06N7/02 IPC Code:
1 Introduction Chenille is a difficult yarn to manufacture, Chenille yarn is a kind of fancy yarns which is requiring great care in production and conversion fascinating about its texture and gleam. These yarns into fi nal articles. Due to the instability of its are produced with the intention of producing an construction, chenille is susceptible to wear in those enhanced aesthetic impression. They fi nd a wide areas of the garment where abrasion occurs. such as range of application area including outerwear fabrics. the underarm, neckline, and waist. Any removal of the fu rnishing fabrics. trimmings and knitwear. pi Ie yarn forming the beard. either during fu rther Chenille yarnshave a pile protruding all around at processing or during the eventual end-use. will expose right angles. These are constructed by twisting core yarns together in chenille yarn machines where pile yarnsare inserted at right angles and cut to within I or 2 mm length to create a surface in which the fibres contained in the pile yarns burst and form a soft pile surface to the yarn. The basic structure of a chenille yarn is shown in Fig. 1. The core yarn gi ves the strength to cheni lie and Pile yarns I constitutes 25-30 % of the relative mass. The pi e (a) yarn is responsible for the body and bulk of the yarn 70-75 and constitutes {Iii } of the relative mass. The pile yarn should provide the desired aesthetic effect for the resulting chenille. I The count and number of the pile yarns and how much of them are fed onto the core determine the count of the chenille yarn.24
'To whom all the correspondence should be addressed. Fig. I -Chenille yarn structure I(a) chenille yarn components and E-mail: [email protected] (b) chenille yarn picture] 502 INDIAN J. FIBRE TEXT. RES., DECEMBER 2006 the ground yarns, which, in turn, will result in a bare count 2011 Ne. The chenille yarns of the count 6 Nm appearance.5 were produced using two core yarns of the count 2411 Despite the fact that chenille yarns are used to Ne (yarn twist 580 turns/m in Z direction, staple produce special fabrics with high added value, there is acrylic fibre) and one pile yarn of the count 3011 Ne. limited research available on the abrasion behaviour Thereafter, woven fabrics were produced with of such yarns and fabrics. Recently, several these chenille yarns using them as fi lling in the fabric researchers attempted to determine the effect of yarn construction on a Dornier weaving machine. For all structure (pile length, twist level, pile material type) fabrics, construction was satin [1/4 3(S)] weave, warp on abrasion characteristics.2-6 They used conventional yarn was polyester (150 denier, punched) and warp statistical test methods for their assessments and density was 66 ends/cm. Weft yarns were: 4 Nm predictions. chenille yarn with weft density of 10 picks/cm and 6 An alternativemethod, fuzzy logic, can be defined Nm chenille yarn with weft density of 14 picks/cm. as a mathematical model to study and define 2.2 Measurement of Fabric Abrasion Properties uncertainties. The importance of fuzzy logic has Abrasion tests of the woven cheni lie fabric samples continually increased, especially in engineering 7 9 were conducted on Martindale wear and abrasion studies. . Fuzzy logic uses the fuzzy set theory and tester in accordance with ASTM 4158-82 (ref. 11). approximate reasoning to deal with imprecision and Before starting the abrasion tests, primary trials were ambiguity in decision making. It provides intuitive, made to determine the abrasion cycle that would be flexible ways to create fuzzy inference systems for suitable for the chenille fabric specimens. According solving complex control and classification problems. to the results of these trials, abrasion cycles were For classification applications, fuzzy logic is a limited with 10000 rubs and the cut samples were process of mapping an input space into an output weighed at the beginning and end of 10000 cycles. space using membership functions and linguistically 1o Mass loss ratios were obtained by dividing mass loss specifiedru les. after 10000 cycles to initial mass of the samples.12 The present study is aimed at analysing abrasion Measurements were repeated three times for each behaviour of chenille fabrics with fuzzy logic method. fabric type. The results were also analysed for The effects of the chenille yarn count, twist level and significance in differences using three-way repeated pile length on the fabric abrasion behaviour have been measure analysis of variance, and the means were studied at constant weave construction and yarn compared by Student-Newman-Keuls (SNK) test at 5 material using fuzzy logic. The chenille yarn % significance level in the COST A T statistical parameters that affect fabric abrasion have been package. Correlation analysis was performed to definedand an experimental study program designed. observe the relationship between actual fabric mass 2 Materials and Methods loss values and predicted fabric mass loss values 2.1 Preparation of Yarn and Fabric Samples obtained from fuzzy logic. The experiments involved chenille yarn samples 2.3 Determination of Chenille Fabric Abrasion Behaviour by manufactured from viscose pile yarns and acrylic core Fuzzy Logic and Construction of Basic Rules yarns. Since the viscose chenille yarns are highly The aim of this study was to check whether the vulnerable to abrasive forces, these are preferred to fuzzy logic system can be used for predicting the use, expecting that the assessment of the results will chenille fabric abrasion behaviour. To achieve this be more explicit. aim, it was necessary to compare the results obtained Chenille yarns were produced with a final count of by fuzzy logic with those obtained by real 4 Nm and 6 Nm incorporating two different pile measurements. lengths (0.7-1.0 mm) and two different twists (700- In fuzzy logic, uncertainty conditions are 850 turns/m in S direction) on a Gigliotti & determined by pre-defined membership functions. If Gualchieri chenille fancy yarn machine. Pile and core the most approximate element of the results is defined yarn materials were spun into chenille yarn under as 1, it can be understood that the other elements are identical conditions on this machine. 4 Nm count between 0 and 1 and they are changing continuously. chenille yarns were produced using two core yarns of The values of every element with the variation the count 2011 Ne (yarn twist 385 turns/m in Z between 0 and 1 are called membership degrees and direction, staple acrylic fibre) and one pile yarn of the the variation in a subset is called membership (:EVEN el PREDICTING ABRASION BEHAVIOUR OF CHENILLE FABRIC BY FUZZY LOGIC 01. : 503 function. After that, the definition of the fuzzy logic The results of the analysis of variance test rule base is implemented by using either Suggeno or (ANOY A) for mass loss values are summarized in Mamdani's method. After a solution of the Table 1. The P-values show that there are statistically implemented fuzzy logic system has been obtained by significant differences between mass loss values for using the selected method, the solution is defuzzified. different twist levels, pi Ie lengths and yarn counts. The defuzzification is called as the inverse operation Interactions between twist level and pi Ie length. of fuzzification.s between twist level and yarn count and also between Yarn count. pi Ie length and twist level were pi Ie length and yarn count have significant effects on selected as input variables, whereas fabric mass loss mass loss values. was the output variable. The fuzzy model of this Effect of Yarn Count on Abrasion Behaviour 3.1 problem is given in Fig. 2. Real values of fabric mass A comparison of the chenille fabrics in terms of loss have been obtained from the experimental study abrasion properties according to SNK test results for comparison with those obtained by fuzzy logic. (Table I) reveals that the abrasion resistance of the The Fuzzy Toolbox in MATLAB 6.5 was used for mathematical calculations. The numbers of the input membership fu nctions to ) and base widths were obtained using the experimental results and expert knowledge (Fig. 3). As eight output variable specifications that are called fabric mass loss were investigated, the output membership function 4.4 4.6 4.0 5.2 5 4 5.6 5.0 6 was divided into eight intervals, as shown in Fig. 4. 4.2 .
Because of using the experimental results and since Yor1l 3 Results and Discussion The calculated 3-D output-input dependency results by fuzzy logic system are presented in Fig 6. The figure shows the relationship among twist level, yarn count and fabric mass loss. Pile length and yarn count which influence the mass loss of fabric are 700 750 800 850 plotted in Fig. 6b. The relationships among pile TWIst level, tums!m Fig. - Membership functions for (a) yarn count (b) pile Icngth length, twist level and fabric mass loss are given in 3 and twist level Fig. 6c. Yarn parameters, like yarn count, pile length (e) and twist level, are the main effective factors which 3.2 5.1 7 influence chenille fabric abrasion behaviour. .t,.t 5.2 7.1 13.2 23.3 0.5 4 6 8 10 12 14 16 18 20 22 24 InplIt .. twist le-'\'�r' Output variable "Mass loss", 8/0 Fig. 2 -Fuzzy logic model for chenille fabric abrasion Fig. 4 - Membership functions of fabric mass loss 504 INDIAN J. FIBRE TEXT. RES., DECEMBER 2006 2. II (twi.Uevel 850) and (pile. length 0.7) and i, 4) then [mass.loss7. is 5.2) [1) J II (twi,tlevel 700) and [pile.lenglh is and [yarn[yarn.counl.counl 4) lhen (mass.lo,,7. 7. 1) [1) 4. II [Iwi,t.level 850) and Ipile.length i,i, 1) and [yarn.count isi, 4) then (mass.loss7. i,is 3.71 (1) 5. II [twisUevel 700)and [pile. length 0.7)1) and [yarn.count i, 6) Ihen loss::' 13.2) (1 ) 6. II [t"';,Uevel 850) and [pile.length i., 0.7) and [yarn. count is 6) then (mass. loss::'i, [1 ) II llwisUevel 700) and (pile lenglh isIS 1) and [yarn counl is then [mass(mass. l . s " 5.1) 7. If [t"';,Uevel 850) and (pile length i, 1)and [yarn.counl is 6) then [mass.loss7.os ::'is is 47)4) (1 ) B 6) . (1 ) Fig. 5 - Some of the defined rules Table I - Results of variance analysis and Student-New man Keuls tests (SNK) for mass loss values Variance analysis Parameter P value = Main Effects Twist level, tpm 0.0000*** Pile length, mm 0.0000*** Yarncount, Nm 0.0005*** Interaction Twist level pile length 0.0000*** x - Twist level yarn count 0.0002*** rn x � Pile length yarn count 0.0002*** 1.'$ x Twist level pile length yarncount 0.0007*** � 0.7 x x 0.8 Student-Newman-Keuls Test Plf tightly so the fabric will have a more compact decreased at higher pile lengths. This difference IS structure. Consequently, yarn count affects the fabric still significant at 5% significance level. abrasion.4 Effect of Pile Length on Abrasion Behaviour As can be observed in Fig. 6, when the yarn 3.2 becomes coarser (Nm count decreases) the fabric One of the results of abrasion is the gradual mass loss increases. This situation is especially visible removal of cut pile yarns from the twists of lock for lower yarn twist values. But when the twist value yarns. Therefore, factors affecting the cohesion of (EVEN el PREDICTING ABRASION BEHAVIOUR OF CHENILLE FABRIC BY FUZZY LOGIC at. : 505 yarns will influence their abrasion resistance. Pile 2:1 length affects the cohesion -of yarnsand the abrasion 10 (a) properties of chenille fabrics. Chenille yarn with 18 higher pile length can provide a more abrasive 16 -- --- resistant yarn structure by preventing the movement 14 of fibres from the core yarns. According to the SNK 12 . test results, fabrics woven with chenille yarns of 0.7 10 mm pi Ie length are abraded more than that of I mm S pile length. The reason is that the longer fibres 6 -;;c:, incorporated into core yarns confer better than the C;:, '" :I short fi bres because they are difficultto remove from ..., ..:: 16 '" the twists of lock yarn. '" (b) As it is observed in Fig. 6b, the increment in pile <': 14 � ...... -- . .... length leads to decrease in mass loss values at all yarn 11 . counts. Figure 6c shows that when the pile length increases there is a significant decrease in mass loss 10 values. This difference decreases at high twist levels S but it is still significantat 5 % significance level. 6 3.3 Effect of Twist Level on Abrasion Bebaviour 4 The abrasion resistance of a yarn is a function of 700 750 800 both the intrinsic abrasion resistance of the fibre composing the yarn and the geometry of the yarn structure. The geometric arrangements of fibres in Fig. 7 - Mass loss values of fabrics woven with (a) 4 Nm and (b) yarns and yarns in fabrics are equally important when 6 Nm chenille yarns with different pile lengths determined by fuzzy logic [- 0.7 mm fuzzy, -- 0.8 mm fuzzy, - - 0.9 mm e designing fabrics with high abrasion resistance. It is fuzzy and 1.0 mm fuzzy] -.- known that yarn twist is directly related to yarn abrasion resistance. The SNK test results given in Table 1 show that 25 ,------, o� y = 0.8594x + 1.2274 mass loss has a tendency to decrease with increasing � 20 +------�� .. twist level. As shown in Fig. 6a, when the twist value oS � 15 increases, the fabric mass loss decreases for all yarn g +------.r-��------� � 10 +------��------� counts. Figure 6c shows that when the twist value t increases the fabric mass loss decreases for all pile � 5 lengths. It can be concluded that the individual pile c5: fibres in low twist chenille yarns may not be held O +------.------.------.------r-----� o 5 10 15 20 25 together well and hence can be easily pulled out. A.ctl.lal lnaSs loss. evo Mass loss is encouraged by inadequate fibre Fig. 8 - Relationship between actual fabric mass loss values adherence. Also, higher twist yarns will retain obtained from measurements and predicted fabric mass loss cohesiveness, presenting a smaller total surface to be values obtained from fuzzy logic abraded.2.4.6 The mass loss values calculated by using the fuzzy The relationship between actual and predicted fabric logic model are plotted in Fig. 7. The effect of twist mass loss values is shown in Fig. 8. In this analysis, it level on mass loss is higher at short pile lengths while is observed that the relationship between measured it is lower at longer pile lengths. Increasing twist level and predicted fabric mass loss values is positive. The beyond 825 turns/m leads to a reduction in mass loss equation for the correlation between the results of values at all pile lengths. This situation is valid both measured and predicted mass loss values is: for 4 Nm and 6 Nm yarn counts. 0.8594 + 1.2274 Moreover, correlation analysis was performed to y = x observe the relationship between actual and predicted where is the predicted fabric mass loss; and the y x, fabric mass loss values obtained from fuzzy logic. actual fabric mass loss (measured). 2006 506 INDIAN 1. FIBRE TEXT. RES., DECEMBER The linear correlation coefficient for the Usage properties and hand properties must be borne relationship between measured and predicted fabric in mind here. Also, it will be useful to investigate the mass loss values is = 0.978. The border value of the effect of chenille yarn parameters (twist level, pile r correlation coefficient at a random degree n-2=6 and length, pile fibre fineness, pile yarn type and fibre the significance level u=O.05, above which the material) on the dimensional and physical properties correlation exists, is 0.707. of chenille yarns (yarn shrinkage, dye absorption. etc.) by fuzzy logic. 4 Conclusions 4.1 The abrasion resistance of the fabrics with finer Acknowledgement chenille yarns is more than that of the fabrics with The authors wish to thank the authorities of Erol coarser chenille yarns. This difference between the TUrkUn Textile, Industry and Trade Co. for the mass loss values is high at lower yarn twist values. opportunity of the experimental study and to Prof. But when the twist value increases the difference Recep EREN for very useful discussion. between the mass loss values for different yarn counts becomes less. This difference is still significant at 5% References significance level. I Clerck D D, Puissiant P & Langenhove V L, Weavability of chenille yarns on air- jet weaving machines, Int Text Bul/, 4.2 It is found that when yarn count increases (yarn (4) (2004) 980-984. becomes finer) , despite the changes in mass loss 2 & Kalaoglu F Demir E, Chenille yarn propel1ies and values at lower pile lengths, the difference between performance of chenille upholstery fabrics, Text Asia, 3 the mass loss values is decreased at higher pile (200 I) 37-40. 3 & D D lengths. This difference is still significant at 5% Kalaoglu F zdemir , A study of wool chenille yarn Proceedings. I" International Textile, Clothing significance. propertics, & Design Conference (University of Zagreb, Croatia), 2002. 4.3 Since higher pile lengths affect the cohesion of 195- 198. yarns, there is a significant decrease in mass loss 4 Dzdemir D & <;even E K, Influence of chenille yarn values with the increase in pile length at low twist manufacturing parameters on yarn and upholstery fahric levels. The increment in pile length leads to decrease abrasion resistance, Text Res J, 74(6) (2004) 515-520. in mass loss values at all yarn counts. 5 Gong R H & Wright R M, Fancy yams, their manufa cture and application (Woodhead Publishing Limited, U.K), 4.4 When the twist value increases, fabric mass loss 2002, 55-56. 81-84. 6 D D & decreases. A high twist level makes the yarn more zdemir Kalaoglu F, The effect of material and compact which retains cohesiveness, presenting a machine parameters on chenille yarn properties, smaller total surface to be abraded and increases the Proceedings, 1st Autex Conference Tecnitex (The University of Minho, Portugal), 200 I, 184- 1 89. pile fibre adherence. 7 Zadeh L A, Fuzzy sets, Information Control, 8 (1965) 338- 4.5 The fabric mass loss can be easily determined in 353. dependence on yarn count, pile length and twist level 8 Zadeh L A. Fuzzy sets as a basis for a theory of possibility. by fuzzy logic. Correlation analysis confirmed strong Fuzzy Sets Systems, I (I) (1978) 3-10. 9 Branco P J & Dente A, Fuzzy systems modelling in linear relationship with high value of correlation 1 practice, Fuzzy Sets Systems, 121 (200 I) 73-93. coefficient (r = 0.978) between measured and (I) 10 Hu B, Dale D S, Huang Y & Watson M D, Cotton color predicted mass loss values. It is practically possible to classification by fuzzy logic, Text Res J, 72(6) (2002). 504- obtain approaches giving positive results with fabric 509 mass loss optimization in a more economical way, II AS7M Standard: D 4158 - 82, Vol. 07.0 1 (American without carrying out additional production. Society of Testing and Materials, USA), 1995. 12 <;even E K, An Investigation About the Effect of Some 4.6 It will be useful to carry out further studies on the Production Parameters on Yarn Properties at Chenille effect of increasing twist level and pile length on Spinning Machines, Masters thesis, University of Uludag. production costs and the comparison of the results. Bursa, 2002.