基于DGRU网络的烘丝机筒壁温度动态预测 - 18新利直播

CN 41-1437/TS ISSN 2096-1553

基于DGRU网络的烘丝机筒壁温度动态预测

张雷,李金学,堵劲松,李龙飞,邹严颉,张二强,李善莲

张雷, 李金学, 堵劲松, 等. 基于DGRU网络的烘丝机筒壁温度动态预测[J]. 轻工学报, 2022, 37(6): 85-91,100. doi: 10.12187/2022.06.011

引用本文:张雷, 李金学, 堵劲松, 等. 基于DGRU网络的烘丝机筒壁温度动态预测[J]. 轻工学报, 2022, 37(6): 85-91,100.doi:10.12187/2022.06.011

ZHANG Lei, LI Jinxue, DU Jinsong, et al. Dynamic prediction of cylinder wall temperature for drum dryer based on DGRU network[J]. Journal of Light Industry, 2022, 37(6): 85-91,100. doi: 10.12187/2022.06.011

Citation:ZHANG Lei, LI Jinxue, DU Jinsong, et al. Dynamic prediction of cylinder wall temperature for drum dryer based on DGRU network[J]. Journal of Light Industry, 2022, 37(6): 85-91,100.doi:10.12187/2022.06.011

基于DGRU网络的烘丝机筒壁温度动态预测

1. 18新利直播电气信息工程学院, 河南郑州 450001;
2. 中国烟草总公司郑州烟草研究院, 河南郑州 450001;
3. 河南中烟工业有限责任公司技术中心, 河南郑州 450000;
4. 陕西中烟工业有限责任公司技术中心, 陕西西安 710065

作者简介:张雷(1988—),男,河南省洛阳市人,18新利直播讲师,博士,主要研究方向为复杂工业智能建模。E-mail:lyzl@zzuli.edu.cn;

基金项目:河南中烟工业有限责任公司科技项目(ZW201806)
烟草行业烟草工艺重点实验室引领计划专项拔尖项目(202022AWHZ01)
国家局烟草科研大数据重大专项项目(110202101083(SJ-07))
河南省科技攻关计划项目( 212102210155)
中图分类号:TS43

Dynamic prediction of cylinder wall temperature for drum dryer based on DGRU network

1. Zhengzhou University of Light Industry, College of Electrical and Information Engineering, Zhengzhou 450001, China;
2. Zhengzhou Tobacco Research Institute of CNTC, Zhengzhou 450001, China;
3. Technology Center, China Tobacco Henan Industrial Co., Ltd., Zhengzhou 450000, China;
4. Technology Center, China Tobacco Shanxi Industrial Co., Ltd., Xi'an 710065, China
Received Date:2022-04-11

CLC number:TS43

摘要:针对烘丝机干燥过程中筒壁温度难以准确进行在线检测的问题,提出基于深度门控循环单元(Deep Gated Recurrent Unit,DGRU)网络的筒壁温度动态预测方法。该方法先对现场工业数据进行小波去噪、归一化等预处理;然后采用互信息理论选择与筒壁温度相关性最强的特征作为模型初始输入变量;最后通过堆叠门控循环单元网络提取工业数据中深层非线性动态特征,输入全连接层中用于估计筒壁温度。基于某烟厂烘丝机工业数据的实验结果表明:DGRU算法预测误差箱体图中的误差中值及均值非常接近零刻度线,且造成的异常点较少。该方法的预测精度较高,能够实现筒壁温度精确动态预测。
- 烘丝机/
- 筒壁温度/
- 深度门控循环单元/
- 动态预测/
- 互信息
Abstract:Aiming at the problem of on-line detection of cylinder wall temperature in the drying process of tobacco dryer, this paper proposed a novel method named deep gated recurrent unit(DGRU)network to predict cylinder wall temperature. In order to improve sample quality, data preprocessing technologies including wavelet denoising and normalization were utilized. Then, mutual information was used to determine the optimal features as model input, which had the largest correlation with the cylinder wall temperature. Next, stacked gate recurrent unit network was introduced to extract the deep hidden nonlinear dynamic representation from thermal data, and then sent into a fully connect network to estimate the cylinder wall temperature. Finally, based on the industrial data of dryer drying process in a tobacco factory, the experiment results showed that the median and mean value of prediction error by DGRU algorithm was very close to the zero scale line,and caused few abnormal points. The method exhibited higher prediction accuracy, and could accurately predict the cylinder wall temperature.
- drum dryer/
- cylinder wall temperature/
- deep gated recurrent unit/
- dynamic prediction/
- mutual information
1. [1]
  邹泉,陈冉,赵云川,等.基于干燥脱水量的叶丝滚筒干燥加工强度品质校正技术[J].烟草科技,2020,53 (7):70-79.
2. [2]
  SUN Q Q,GE Z Q.A Survey on deep learning for data-driven soft sensors[J].IEEE Transactions on Industrial Informatics,2021,17 (9):5853-5866.
3. [3]
  ZHANG X G,ZHANG L,CHEN H,et al.Prediction of coal feeding during sintering in a rotary kiln based on statistical learning in the phase space[J].ISA Transactions,2018,83:248-260.
4. [4]
  ZHANG L,ZHANG X G,CHEN H,et al.A robust temperature prediction model of shuttle kiln based on ensemble random vector functional link network[J].Applied Thermal Engineering,2019,150:99-110.
5. [5]
  JIN L,LI S,HU B.RNN models for dynamic matrix inversion:A control-theoretical perspective[J].IEEE Transactions on Industrial Informatics,2018,14 (1):189-199.
6. [6]
  HWANG C L,JAN C.Recurrent-Neural-Network-Based multivariable adaptive control for a class of nonlinear dynamic systems with time-varying delay[J].IEEE Transactions on Neural Networks and Learning Systems,2016,27 (2):388-401.
7. [7]
  WANG J L,ZHANG J,WANG X X.Bilateral LSTM:A two-dimensional long short-term memory model with multiply memory units for short-term cycle time forecasting in re-entrant manufacturing systems[J].IEEE Transactions on Industrial Informatics,2018,14 (2):748-758.
8. [8]
  LI X C,DUAN F,LOUKOPOULOS P,et al.Canonical variable analysis and long short-term memory for fault diagnosis and performance estimation of a centrifugal compressor[J].Control Engineering Practice,2018,72:177-191.
9. [9]
  TAN P,HE B,ZHANG C,et al.Dynamic modeling of NOX emission in a 660 MW coal-fired boiler with long short-term memory[J].Energy,2019,176:429-436.
10. [10]
  LE T H N,QUACH K G,LUU K,et al.Reformulating level sets as deep recurrent neural network approach to semantic segmentation[J].IEEE Transactions on Image Processing,2018,27(5):2393-2407.
11. [11]
  LIU J T,WU C H,WANG J W.Gated recurrent units based neural network for time heterogeneous feedback recommendation[J].Information Sciences,2018,423:50-65.
12. [12]
  ZHAO R,WANG D Z,YAN R Q,et al.Machine health monitoring using local feature-based gated recurrent unit networks[J].IEEE Transactions on Industrial Electronics,2018,65 (2):1539-1548.
13. [13]
  QUE Z J,JIN X H,XU Z G.Remaining useful life prediction for bearings based on a gated recurrent unit[J].IEEE Transactions on Instrumentation and Measurement,2021,70:1-11.
14. [14]
  PENG H C,LONG F H,DING C.Feature selection based on mutual information: Criteria of max-dependency,max-relevance,and min-redundancy[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(8):1226-1238.
15. [15]
  ZHOU P,GUO D W,WANG H,et al.Data-Driven robust M-LS-SVR-based NARX modeling for estimation and control of molten iron quality indices in blast furnace ironmaking[J].IEEE Transactions on Neural Networks and Learning Systems,2018,29(9):4007-4021.
16. [16]
  XIONG W L,LI Y J,ZHAO Y J,et al.Adaptive soft sensor based on time difference Gaussian process regression with local time-delay reconstruction[J].Chemical Engineering Research and Design,2017,117:670-680.
1. [1]
  王晓,吴洲,王宏伟,王榕,陈浩然. 基于深度学习和蛋白质语言模型的抗菌肽预测模型研究. 轻工学报, 2024, 39(2): 12-18.doi: 10.12187/2024.02.002

WeChat

点击查看大图

计量

PDF下载量:23
文章访问数:815
引证文献数:0

通讯作者:陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

张雷, 李金学, 堵劲松, 等. 基于DGRU网络的烘丝机筒壁温度动态预测[J]. 轻工学报, 2022, 37(6): 85-91,100. doi: 10.12187/2022.06.011

引用本文:张雷, 李金学, 堵劲松, 等. 基于DGRU网络的烘丝机筒壁温度动态预测[J]. 轻工学报, 2022, 37(6): 85-91,100.doi:10.12187/2022.06.011

ZHANG Lei, LI Jinxue, DU Jinsong, et al. Dynamic prediction of cylinder wall temperature for drum dryer based on DGRU network[J]. Journal of Light Industry, 2022, 37(6): 85-91,100. doi: 10.12187/2022.06.011

Citation:ZHANG Lei, LI Jinxue, DU Jinsong, et al. Dynamic prediction of cylinder wall temperature for drum dryer based on DGRU network[J]. Journal of Light Industry, 2022, 37(6): 85-91,100.doi:10.12187/2022.06.011

基于DGRU网络的烘丝机筒壁温度动态预测

作者简介:张雷(1988—),男,河南省洛阳市人,18新利直播讲师,博士,主要研究方向为复杂工业智能建模。E-mail:lyzl@zzuli.edu.cn

1. 18新利直播电气信息工程学院, 河南郑州 450001;
2. 中国烟草总公司郑州烟草研究院, 河南郑州 450001;
3. 河南中烟工业有限责任公司技术中心, 河南郑州 450000;
4. 陕西中烟工业有限责任公司技术中心, 陕西西安 710065

收稿日期: 2022-04-11

基金项目:河南中烟工业有限责任公司科技项目(ZW201806)烟草行业烟草工艺重点实验室引领计划专项拔尖项目(202022AWHZ01)国家局烟草科研大数据重大专项项目(110202101083(SJ-07))河南省科技攻关计划项目( 212102210155)

关键词:

摘要:针对烘丝机干燥过程中筒壁温度难以准确进行在线检测的问题,提出基于深度门控循环单元(Deep Gated Recurrent Unit,DGRU)网络的筒壁温度动态预测方法。该方法先对现场工业数据进行小波去噪、归一化等预处理;然后采用互信息理论选择与筒壁温度相关性最强的特征作为模型初始输入变量;最后通过堆叠门控循环单元网络提取工业数据中深层非线性动态特征,输入全连接层中用于估计筒壁温度。基于某烟厂烘丝机工业数据的实验结果表明:DGRU算法预测误差箱体图中的误差中值及均值非常接近零刻度线,且造成的异常点较少。该方法的预测精度较高,能够实现筒壁温度精确动态预测。

English Abstract

参考文献 (16) 相关文章 (1)

/

下载: 全尺寸图片幻灯片

分享

用微信扫码二维码

分享至好友和朋友圈