染色机助剂智能配送系统的构建及实践

doi:10.13475/j.fzxb.20210911709

摘要/Abstract

摘要：

针对传统的染色机助剂配送系统配送误差大的问题,提出了一种基于推荐预停值和预计用时预测的多层全连接神经网络模型。首先,使用配送过程记录的数据训练网络模型;然后,将需要配送的数据代入训练好的网络模型进行计算,得到推荐预停值和预计用时,推荐预停值与经验预停值按照可变比例算法计算最终预停值,系统根据最终预停值决定配送阀的关闭时机,利用预计用时评估配送过程是否超时。最后,使用4种预停模式各进行1 000次的助剂配送验证实验。结果表明,采用神经网络模型预测模式的配送误差的标准差为23.8 g,平均绝对误差为16.1 g,优于其他3种预停模式的配送误差,取得了较好的助剂配送精度。

关键词: 助剂配送, 预停值, 神经网络, 可变比例, 配送精度, 染色机

Abstract:

Aiming at the large distribution errors in traditional chemicals distribution systems for dyeing machines, a multi-layer fully connected neural network model was proposed based on the prediction of recommended pre-stop value and predicted time. The network model was firstly trained using data recorded in the distribution process, and the data to be distributed was fed into the trained network model for calculation so as to obtain the recommended pre-stop value and predicted time. The recommended pre-stop value and empirical pre-stop value were used to obtain the final pre-stop value according to the variable ratio algorithm. The system worked to determine the closing time of the distribution valve according to the final pre-stop value. The predicted time was used to evaluate whether the distribution process was timed out. Four pre-stop modes were used for chemicals distribution experiments over 1 000 times, and the results show that the standard deviation of distribution error predicted by network model is 23.8 g, the mean absolute error is 16.1 g. It is superior to the other three pre-stop modes with better chemical distribution accuracy.

Key words: chemicals distribution, pre-stop value, neural networks, variable radio, distribution accuracy, dyeing machine

中图分类号:

TP23

张福沐, 刘端武, 胡跃明. 染色机助剂智能配送系统的构建及实践[J]. 纺织学报, 2022, 43(11): 179-187.

ZHANG Fumu, LIU Duanwu, HU Yueming. Construction and experiment of intelligent chemicals distribution system for dyeing machine[J]. Journal of Textile Research, 2022, 43(11): 179-187.

图/表 14

图1

图2

图3

图4

图5

图6

图7

表1

图8

图9

图10

表2

图11

表3

参考文献 13

[1]	何颖婷, 李敏, 王瑞丰, 等. 涤纶织物的连续式轧染工艺[J]. 纺织学报, 2022, 43(3): 110-115.
	HE Yingting, LI Min, WANG Ruifeng, et al. Dyeing process for polyester fabrics with continuous pad dyeing method[J]. Journal of Textile Research, 2022, 43(3): 110-115. doi: 10.1177/004051757304300208
[2]	刘江坚, 万捷. 第17届上海国际纺织工业展览会针织染色机械述评[J]. 针织工业, 2015, 318(7): 28-37.
	LIU Jiangjian, WANG Jie. Review of knitting dyeing machines on the 17th shanghai international textile industry exhibition[J]. Knitting Industries, 2015, 318(7): 28-37.
[3]	李亮. 基于流量计量的助剂自动配重系统[D]. 杭州: 浙江理工大学, 2013:16-63.
	LI Liang. Automatic counterweight system for chemicals based on flow metering[D]. Hangzhou: Zhejiang Sci-Tech University, 2013:16-63.
[4]	付贺檀. 基于PLC的助剂自动配送控制系统[D]. 杭州: 浙江理工大学, 2012:35-36.
	FU Hetan. Automatic distribution control system of chemicals based on PLC[D]. Hangzhou: Zhejiang Sci-Tech University, 2012:35-36.
[5]	LI Z, HUANG Q, DUAN Y, et al. Research on electromagnetic flowmeter based on double-frequency trapezoidal wave excitation[J]. Journal of Physics: Conference Series, 2020. DOI: 10.1088/1742-6596/1549/5/052086. doi: 10.1088/1742-6596/1549/5/052086
[6]	胡梦蝶. 高抗干扰性能的电磁流量计研究与设计[D]. 武汉: 湖北工业大学, 2018:10-15.
	HU Mengdie. Research and design of electromagnetic flowmeter with high anti-interference performance[D]. Wuhan: Hubei University of Technology, 2018:10-15.
[7]	LAPEDES A, FARBER R. Nonlinear signal processing using neural networks: prediction and system model-ing[J]. Los Alamos National Laboratory Report, 1987, 87: 2662.
[8]	LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553): 436-444. doi: 10.1038/nature14539
[9]	AL-MAQALEH B M, AL-MANSOUB A A, AL-BADANI F N. Forecasting using artificial neural network and statistics models[J]. International Journal of Education and Management Engineering, 2016, 6(3):20-32.
[10]	金守峰, 林强强, 马秋瑞, 等. 基于BP神经网络的织物表面绒毛质量的检测方法[J]. 纺织学报, 2020, 41(2): 69-76.
	JING Shoufeng, LIN Qiangqiang, MA Qiurui, et al. Detection method of fabric surface fluff quality based on BP neural network[J]. Journal of Textile Research, 2020, 41(2):69-76.
[11]	夏明, 李沛儒, 丁蓉. 基于要素匹配的连衣裙成本快速核算方法[J]. 纺织学报, 2022, 43(3): 146-152.
	XIA Ming, LI Peiru, DING Rong. Cost rapid accounting technology of dress based on element matching[J]. Journal of Textile Research, 2022, 43(3): 146-152. doi: 10.1177/004051757304300305
[12]	杨恒, 岳建平, 邢尹, 等. 基于深度全连接神经网络的大坝变形预测研究[J]. 大地测量与地球动力学, 2021, 41(2):162-166.
	YANG Heng, YUE Jianping, XING Yin, et al. Research on dam deformation prediction based on deep fully connected neural network[J]. Journal of Geodesy and Geodymics, 2021, 41(2): 162-166.
[13]	AMARI S I. Backpropagation and stochastic gradient descent method[J]. Neurocomputing, 1993, 5(4-5): 185-196. doi: 10.1016/0925-2312(93)90006-O

相关文章 15

[1]	刘锋, 徐杰, 柯文博. 基于深度强化学习的服装缝制过程实时动态调度[J]. 纺织学报, 2022, 43(09): 41-48.
[2]	金守峰, 侯一泽, 焦航, 张鹏, 李宇涛. 基于改进AlexNet模型的抓毛织物质量检测方法[J]. 纺织学报, 2022, 43(06): 133-139.
[3]	夏明, 李沛儒, 丁蓉. 基于要素匹配的连衣裙成本快速核算方法[J]. 纺织学报, 2022, 43(03): 146-152.
[4]	江学为, 田润雨, 卢方骁, 张艺. 基于模拟评分的服装推荐改进算法[J]. 纺织学报, 2021, 42(12): 138-144.
[5]	江慧, 马彪. 基于服装风格的款式相似度算法[J]. 纺织学报, 2021, 42(11): 129-136.
[6]	庹武, 王哓玉, 高雅昆, 于媛媛, 郝潇潇, 刘永亮, 郭鑫. 基于改进边缘检测算法的服装款式识别[J]. 纺织学报, 2021, 42(10): 157-162.
[7]	许雪梅. 基于模拟退火算法改进遗传算法的织物智能配色[J]. 纺织学报, 2021, 42(07): 123-128.
[8]	孟朔, 夏旭文, 潘如如, 周建, 王蕾, 高卫东. 基于卷积神经网络的机织物密度均匀性检测[J]. 纺织学报, 2021, 42(02): 101-106.
[9]	王奕文, 罗戎蕾, 康宇哲. 基于卷积神经网络的汉服关键尺寸自动测量[J]. 纺织学报, 2020, 41(12): 124-129.
[10]	徐守东, 冷奕锦, 吴国新. 基于神经网络的籽棉颜色分级检测[J]. 纺织学报, 2020, 41(10): 34-40.
[11]	张晓侠, 刘凤坤, 买巍, 马崇启. 基于BP神经网络及其改进算法的织机效率预测[J]. 纺织学报, 2020, 41(08): 121-127.
[12]	邵金鑫, 张宝昌, 曹继鹏. 基于图像处理与深度学习方法的棉纤维梳理过程纤维检测识别技术[J]. 纺织学报, 2020, 41(07): 40-46.
[13]	莫帅, 冯战勇, 唐文杰, 党合玉, 邹振兴. 基于神经网络和遗传算法的锭子弹性管性能优化[J]. 纺织学报, 2020, 41(04): 161-166.
[14]	王泽霞, 陈革, 陈振中. 基于改进卷积神经网络的化纤丝饼表面缺陷识别[J]. 纺织学报, 2020, 41(04): 39-44.
[15]	王晓华, 姚炜铭, 王文杰, 张蕾, 李鹏飞. 基于改进YOLO深度卷积神经网络的缝纫手势检测[J]. 纺织学报, 2020, 41(04): 142-148.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

数据集	预停值损失值/g	预停值平均绝对误差/g	预计用时损失值/s	预计用时平均绝对误差/s
训练集	564.10	15.23	2.24	1.02
校验集	588.86	15.40	2.11	1.00

模式	方差/g	标准差/g	平均绝对误差/g
模式1	1 022.0	32.0	158.4
模式2	907.4	30.1	61.4
模式3	746.1	27.3	32.8
模式4	564.1	23.8	16.1

统计方法	方差/s	标准差/s	平均绝对误差/s
模式4(包含1次配送异常)	551.3	23.5	1.8
模式4(剔除1次配送异常)	2.5	1.6	1.0