Journal of Textile Research ›› 2019, Vol. 40 ›› Issue (01): 52-56.doi: 10.13475/j.fzxb.20180305606

• Textile Engineering • Previous Articles     Next Articles

Prediction of cotton yarn quality based on four-layer BP neural network

ZHA Liugen, XIE Chunping()   

  1. Key Laboratory of Eco-Textiles (Jiangnan University), Ministry of Education, Wuxi, Jiangsu 214122, China
  • Received:2018-03-23 Revised:2018-09-07 Online:2019-01-15 Published:2019-01-18
  • Contact: XIE Chunping E-mail:wxxchp@vip.163.com

Abstract:

In order to further improve the accuracy and training speed of the BP neural network in yarn quality prediction, a four-layer BP neural network with double hidden layers was proposed for cotton yarn quality prediction on the basis of the conventional three-layer BP neural network model of single hidden layer. By constructing the model of the breaking strength of pure cotton yarn and the CV model of yarn levelness, a three-layer BP neural network and a four-layer BP neural network were designed under each model, and the final training and simulation were performed using MatLab. In order to ensure the comparability of the final results, the training parameters of the two network models and the data used are consistent. The experimental results show that under the fracture strength model, the maximum number of training steps in the four-layer network compared to the three-layer network is reduced from 740 to 533, and the relative average error decreases from 9.6% to 7.5%. In the yarn levelness CV value model under the four-layer network, compared with the three-layer network, the maximum number of training steps decreases from 929 to 604, and the relative average error decreases from 10.2% to 8.3%.

Key words: yarn quality prediction, cotton yarn, four-layer BP neural network, MatLab simulation

CLC Number: 

  • TS111.9

Fig.1

Basic BP neural network model"

Tab.1

Partial raw data"

马克隆
上半部
平均长度/
mm
整齐
短纤维
指数
强度/
(cN·dtex-1)
断裂
强力/
cN
条干
CV值/%
4.4 29.7 83.3 14.3 29.9 714.6 12.4
4.4 29.8 83.3 14.2 30.1 689.5 12.0
4.5 29.8 83.3 14.3 30.1 699.3 12.3
4.4 29.5 82.6 16.4 28.8 679.6 12.6
4.3 29.4 82.6 16.1 29.0 685.8 12.7
4.3 29.4 82.6 16.1 29.0 683.6 12.8
4.4 29.4 82.8 15.8 29.0 671.6 13.0
4.3 29.3 83.0 15.9 28.6 673.7 13.2
4.4 29.3 83.0 15.9 28.6 652.4 12.7
4.3 29.1 82.7 16.4 28.5 676.3 12.9

Fig.2

Three layer network structure"

Fig.3

Four-layer network structure"

Fig.4

Three-layer network error curve of yarn breaking strength model"

Fig.5

Four-layer network error curve of yarn breaking strength model"

Fig.6

Three-layer network error curve of yarn levelness"

Fig.7

Four-layer network error curve of yarn levelness"

Tab.2

Simulation results"

网络模型 结构 精度
目标
最大训练
步数
相对平均
误差/%
断裂强力
模型
三层 5-4-1 0.001 740 9.6
四层 5-3-2-1 0.001 533 7.5
条干
CV值模型
三层 5-4-1 0.001 929 10.2
四层 5-3-2-1 0.001 604 8.3

Fig.8

Correlation analysis. (a) Three-layer BP neural network yarn breaking strength model; (b) Four-layer BP neural network yarn breaking strength model; (c) Three-layer BP neural network yarn levelness CV value model; (d) Four-layer BP neural network yarn levelness CV Value model"

[1] 郝海涛, 谢春萍. 利用神经网络与AFIS纤维测试系统预测纱线质量[J]. 棉纺织技术, 2003(8):8-11.
HAO Haitao, XIE Chunping. Predicting yarn quality using neural network and AFIS fiber testing system[J]. Cotton Textile Technology, 2003(8):8-11.
[2] 陈俊杰, 谢春萍, 郝海涛. 人工神经网络纱线质量预报模型的实用化[J]. 天津纺织科技, 2005(4):17-19.
CHEN Junjie, XIE Chunping, HAO Haitao. Practicalization of artificial neural network yarn quality forecasting model[J]. Tianjin Textile Science and Technology, 2005(4):17-19.
[3] GUO Daqi. Study on the linear primary function forward three layers neural network's architecture[J]. Journal of Computers, 2005,13(9):429-443.
[4] 史峰. MatLab智能算法30个案例分析[M]. 北京: 北京航空航天大学出版社, 2011: 23-25.
SHI Feng. 30 Cases Analysis of MatLab Intelligent Algorithm[M]. Beijing: Beijing University of Aeronautics and Astronautics Press, 2011: 23-25.
[5] 焦李成. 神经网络系统理论[M]. 西安: 西安电子科技大学出版社, 1996: 47-55.
JIAO Licheng. Theory of Neural Network Systems[M]. Xi'an: Xidian University Press, 1996: 47-55.
[6] 张思源, 包燕平, 张超杰, 等. BP神经网络IF钢铝耗的预测模型[J]. 工程科学学报, 2017(4):6-8.
ZHANG Siyuan, BAO Yanping, ZHANG Chaojie, et al. Prediction model for aluminum consumption of IF steel in BP neural network[J]. Journal of Engineering Sciences, 2017(4):6-8.
[7] 马发民, 吴红乐, 张林, 等. 基于BP神经网络的疲劳判定[J]. 计算机与数字工程, 2017(6):24-30
MA Famin, WU Hongle, ZHANG Lin, et al. Fatigue determination based on BP neural network[J]. Computer and Digital Engineering, 2017(6):24-30.
[8] ABBASIAA, VOSSOUGHI G R, AHMADIAN M T. Deformation prediction by a feed forward artificial neural network during mouse embryo micromanipulation[J]. Animal Cells and Systems, 2012 (2):114-123.
doi: 10.1080/19768354.2020.1737225 pmid: 32489691
[9] 李翔, 彭志勤, 金凤英, 等. 基于神经网络的精纺毛纱性能预测模型比较[J]. 纺织学报, 2011,32(3):27-31.
LI Xiang, PENG Zhiqin, JIN Fengying, et al. Comparison of performance prediction models of worsted wool yarns based on neural network[J]. Journal of Textile Research, 2011,32(3):27-31.
[10] 沈花玉, 王兆霞, 高成耀, 等. BP 神经网络隐含层单元数的确定[J]. 天津理工大学学报, 2008,24(5):13-15.
SHEN Huayu, WANG Zhaoxia, GAO Chengyao, et al. Determination of hidden layer units in BP neural network[J]. Tianjin University of Technology, 2008,24(5):13-15.
[1] LI Ruiqing, WANG Wei, WEI Bingju, ZHOU Changwen, ZHANG Shutao. Sulfur black dyeing process with environment friendly reducing agent [J]. Journal of Textile Research, 2020, 41(08): 50-54.
[2] DING Yongsheng, DAI Yamin, ZHONG Yi, XU Hong, MAO Zhiping, ZHANG Linping, CHEN Zhize. Dyeing kinetics of reactive dye on cotton yarn in Pickering emulsion system [J]. Journal of Textile Research, 2020, 41(07): 101-108.
[3] WANG Jiandong, XIA Fenglin, LI Yalin, ZHAO Yuning. Optimal sliding mode control of electronic transverse servo for comb bar of warp knitting machine [J]. Journal of Textile Research, 2020, 41(02): 143-148.
[4] . Microwave assisted preparation and sizing properties of carboxymethyl corn starch [J]. Journal of Textile Research, 2018, 39(12): 59-66.
[5] . Application and process optimization of false twist in ring spinning machine [J]. JOURNAL OF TEXTILE RESEARCH, 2018, 39(07): 27-31.
[6] . Algorithm of processing code generation for pattern spline curves [J]. JOURNAL OF TEXTILE RESEARCH, 2018, 39(07): 153-158.
[7] . Influence of humidity and temperature on properties of sized polyester/cotton 65/35 yarn [J]. JOURNAL OF TEXTILE RESEARCH, 2018, 39(06): 70-74.
[8] . Influence of sizing pretreatment agent in properties of warp knitting cotton yarn  [J]. JOURNAL OF TEXTILE RESEARCH, 2018, 39(04): 82-86.
[9] . Design and production method of cotton warp outerwear knitted fabrics [J]. JOURNAL OF TEXTILE RESEARCH, 2017, 38(05): 53-57.
[10] . Construction of radial basis function neural network models for typical cross section curve of shorts [J]. JOURNAL OF TEXTILE RESEARCH, 2015, 36(05): 83-88.
[11] . Dynamic analysis od shogging motion mechanism of guide bar on warp knitting machine [J]. JOURNAL OF TEXTILE RESEARCH, 2012, 33(11): 121-126.
[12] LIU Deju;WANG Huayin. Stripping technology for vortex spun cotton yarn packages [J]. JOURNAL OF TEXTILE RESEARCH, 2011, 32(2): 84-87.
[13] LIU Deju;WU Huanling;WANG Yiqun. Influence of dyeing and finishing process on breaking strength of cotton MVS yarn [J]. JOURNAL OF TEXTILE RESEARCH, 2010, 31(5): 86-90.
[14] Hao Xinmin;LI Hongwei;LI Xinxin. Research on moisture absorption and permeability properties of hemp, cotton and their blended yarns [J]. JOURNAL OF TEXTILE RESEARCH, 2010, 31(2): 33-38.
[15] XU Fuguan;XU Liqin. Unification of cotton yarn count conversion factor between English count and tex system [J]. JOURNAL OF TEXTILE RESEARCH, 2008, 29(11): 35-38.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!