Font Size: a A A

Study On The Strength Prediction Model Of Filament Yarn With Different Yarn Structures

Posted on:2018-06-18Degree:DoctorType:Dissertation
Country:ChinaCandidate:W HuangFull Text:PDF
GTID:1311330566452287Subject:Textile Engineering
Abstract/Summary:PDF Full Text Request
The performance of textiles is largely determined by the performance of the fibers and yarns.The processes of spinning,weaving and finishing provide a versatile end use for the final product.Studying the performance and process parameters of the fiber and yarn has practical significance for the textile design and quality improvement.The use of experiments to find the best yarn structure to produce new yarns exsits the problems such as long process flow,inefficient implementation and has a large effect on the the spinning equipment,environment and test methods.And the experiments can not theoretically and quantitativly explain the strength enhancement effects,the existence of critical twist factor and the relationship between fiber interaction and tensile properties.If a more accurate mathematical model can be used to predict the tensile properties of the yarn without destroying,it is efficient,but also provides a theoretical basis for the yarn production and even the fabric design.The strength of the yarn is one of the most important indicators of yarn quality.The main work of the thesis is to establish a prediction model of yarn strength with different yarn structures.The prediction model can be used to predict the mechanical properties of the yarn without the destructive tests.The parameters of fiber and yarn are used to describe the effects of fiber properties,process parameters and yarn structure parameters on stress-strain behavior of the yarn.The model can also be used to predict the changes of fiber size,modulus,twist level and other parameters.The effect of the changes of fiber parameters on the final yarn products can be determined in quantitative.The basic idea of this thesis is to establish the geometric model of the yarn,to derive the mathematical relationship among the variables,to verify the correctness of the model and further improve it.The research object is the twisted multifilament yarn,two-ply yarn and wrapped yarn composed of filaments.The main contents of the thesis are as follows:(1)The basic model of yarn strength is established.The yarn with different yarn structures are considered a continuum.The constitutive equation is used to describe the response of the continuous medium model to the external action by means of the assumption and regularity of continuum mechanics.The geometrical models are established with the hypothesis of the properties and distribution of yarns and filaments.After the mechanical analysis on short segment,the relationships of the external stress-strain behavior and the tensile modulus are deduced.The three basic models are obtained,and the possible shortcomings are given.(2)The modified model is established.In order to solve the shortcomings of the basic model,the factors of the arrangement of filaments,the precision of tensile modulus,the pre-stress caused by twisting,the change of yarn structure parameters with twist and strain and the cohesion between the single yarns are mainly discussed.Then the influences of the main parameters on the final yarn strength are analyzed quantitatively.Finally,the prediction ability of the basic and modified models is compared with the results of the spinning experiments.The effectiveness of the models is verified,and the necessity of a further correction is raised.(3)Aiming at the situation that the yarn strength cannot be well predicted by the parametric model at the higher twist levels,a semi-parametric model of the yarn strength is proposed.The estimation of the variables is transformed from the theoretical relation to the regression estimation by experimental observation.The method is proposed mainly for the determination of yarn modulus and diameter.The actual tensile modulus is obtained by differential operation on the data of tensile curve.The diameter of the yarn is measured by video microscope and image processing method.Their distributions are estimated by a two-stage regression.Finally,the prediction ability of the parameter and the semi-parametric models are compared by spinning experiments.(4)The nonparametric estimation of yarn diameter in semi-parametric model under larger twist levels is studied.A continuous measurement system for the yarn diameter is constructed to obtain the actual changes in of the yarn diameter,and its regression relationship with the strain and twist level.The system is composed of the hardware part of the device design and construction,and the software part of the image processing and interface design.In addition,the GUI is designed,and is packaged as an executable program.The system can dynamically measure the yarns during the stretching procedure,providing reliable data for the yarn’s semi-parametric model.(5)It is proposed that the prediction of yarn strength should adopt the parameter model and the semi-parametric one according to the twist level.And the values of predictions and tests during the whole twist range are compared,verifying the forecasting ability of the strength model.Besides,the applications on the tensile curve and the tenacity utilization ratio are studied.Through the research of this subject,the following conclusions can be obtained:(1)Parametric models(basic and modified models)of twisted multifilament yarn,two-ply yarn and wrapped yarn are established and verified.An untwisted polyester FDY multifilament yarn are adopted as the single yarn for spinning.For twisted multifilament yarn,the deviations of the basic model are 60.37%,21.30%,8.19%,7.76% and 4.05%,while the deviations of the modified model are 5.49%,1.35%,3.64%,0.12% and 1.65% respectively.The results of the modified model are better than those of the basic one at every twist level.For two-ply yarn,the deviations of the basic model are 6.41%,18.41%,41.47%,69.32% and 100.16%,while the deviations of modified model are 6.57%,1.69%,3.27%,7.47% and 11.08% respectively.The results of the modified model are generally better than those of the basic one at every twist level.For the wrapped yarn,the deviations of the basic model are 2.53%,34.18%,60.32%,48.73% and 90.68%,while the deviations of the modified model are 14.18%,5.12%,0.51%,10.38% and 18.07% respectively.The results of the modified model are better than those of the basic one except at the first twist level.In general,the modified model is superior to the basic one in the prediction ability.It is also found that the deviation is quite large when the twist level is high.(2)Aiming at the situation that the deviations at higher twist levels are larger,the semiparametric model of the yarn strength is proposed and verified.Under the experimental conditions,the deviations of parametric model are 6.57%,1.69%,3.27%,7.47% and 11.08%,while the deviations of the semi-parametric one are 18.14%,17.14%,13.53 %,5.83% and 9.03% respectively.The results of the parameter model are better than those of the semiparametric one at low-medium twist levels,and the semi-parametric model is superior to the parametric model at higher twist levels.(3)For the lack of stretch ratio and data volume,a continuous measurement system for yarn diameter integrating yarn measurement and image analysis is designed.The device allows dynamic measurement for the image and the tensile process of the yarn.The results show that the diameter distribution of the low-twist yarn changes smoothly between adjacent points,and the diameter distribution of the medium-high twist is periodic.The relationship estimated by nonparametric regression amng the test results of the measurement system can be used in the semi-parametric model under a higher twist level,which the deviations under two twist levels are 0.50% and 1.60%.(4)Based on the above research,it is proposed that the yarn strength model should adopt the parameter model and the semi-parametric model according to the twist level.And the applications on the tensile curve and the tenacity utilization ratio are studied.Under the experimental conditions,for the multifilament yarn,the parametric model is adopted,the deviation is winthin 6%,and the mean deviation is 3.09%.For the two-ply yarn,the parametric model is adopted when the twist is between 0 and 706 tpm,while the semi-parametric model is adopted when the twist is over 707 tpm.The deviation is within 7% and the mean deviation is 2.95%.For the wrapped yarn,the parametric(basic)model is adopted when the twist is between 0 and 141 tpm,the parametric(modified)model is adopted when the twist is between 142 and 706 tpm,and the semi-parametric model is adopted when the twist is over 707 tpm.The deviation is within 8%,and the mean deviation is 3.82%.The results verifies the correctness of prediction model.The strength model is applied to predict the tensile curve.The mean deviations among the data points is 5.60%,7.01%,3.01%,8.68% and 1.63%,respectively.The theoretical curves are quite close to the tested ones.The parametric model is applied to predict the tenacity utilization ratio.The single yarn’s strength utilization coefficient of the three kinds of structural yarns with the twist factor can be predicted,and the single yarn’s tenacity utilization coefficient can be further calculated.In addition to the higher twist levels,the prediction results are quite close to the experimental results.The mean deviation is 3.92% for multifilament yarn,5.54% for two-ply yarn and 9.82% for wrapped yarn.In general,the parametric model can be used to determine the change of the tenacity utilization coefficient of the structural yarns and the critical twist factor.The prediction model of yarn strength and the related discussion and analysis provide a theoretical basis for studying the relationship among the yarn strength and fiber properties,process parameters and structural parameters.It also provides guidance for the optimization of spinning process,while the method provides reference for the research of the related problems of yarn strength.
Keywords/Search Tags:filament yarn, strength prediction, parametric model, semi-parametric model, yarn geometry, two-ply yarn, wrapped yarn
PDF Full Text Request
Related items