数据驱动的热连轧板平直度分类预测

李广焘; 龚殿尧; 张殿华; 梁勋国; 陈驰

doi:10.13228/j.boyuan.issn0449-749x.20230690

PDF(1732 KB)

钢铁 ›› 2024, Vol. 59 ›› Issue (7) : 83-93. DOI: 10.13228/j.boyuan.issn0449-749x.20230690

压力加工

数据驱动的热连轧板平直度分类预测

李广焘¹, 龚殿尧², 张殿华², 梁勋国¹, 陈驰¹

作者信息 +

Data-driven classification prediction of hot strip flatness

李广焘¹, 龚殿尧², 张殿华², 梁勋国¹, 陈驰¹

Author information +

文章历史 +

摘要

在热连轧带钢生产过程中,平直度是重要的板形质量指标。为提高热连轧带钢板形的控制水平,针对热连轧带钢平直度检测滞后与预测精度不高的问题,建立基于数据驱动的热连轧板平直度分类预测模型。将工艺参数、设备参数以及轧件参数中对带钢平直度影响显著的66个关键参数作为输入变量。按照平直度仪在轧制方向上的实际测量间距以及在带钢宽度方向上的测量通道宽度,将钢卷离散化为一系列条元,并生成不同条元带钢对应的平直度分类标签,研究实际以带钢宽度方向上中部条元的平直度分类标签作为输出变量。基于提出的数据预处理方法,获取用于构建板平直度分类预测模型的可靠且高质量的建模数据。采用深度置信网络（deep belief networks,DBN）构建基于数据驱动的热连轧板平直度分类预测模型。以交叉熵为评价指标,分析超参数对DBN模型性能的影响,确定模型的最佳超参数。在测试集样本类别分布不均衡的情况下,构建的DBN模型对大中浪平直度类别、无中浪平直度类别以及微中浪平直度类别的预测准确度分别达到了100%、96.87%与77.78%,平直度的错误预测类别主要为与其平直度值更接近的相邻平直度类别。该方法能对热连轧带钢平直度进行准确的分类预测,对提高板形预设定模型的计算精度与改善带钢板形具有重要作用。

Abstract

The flatness is an important quality indicator of the strip shape in hot strip rolling. In order to further improve strip shape control level, and in view of the problems of hot-rolled strip flatness on detection lag and low prediction accuracy, a data-driven classification model of the hot-rolled strip flatness is developed. The 66 key parameters that have a significant impact on the strip flatness among the process parameters, equipment parameters and strip parameters are used as input variables. According to the actual measured spacing of the flatness meter in the rolling direction and the measurement channel width along the strip width direction, the strip coils are dispersed into a series of strip units, and generating flatness classification labels corresponding to different strip units, the flatness classification labels of the middle strip units along the strip width direction are actually used as output variables in this study. The method of data preprocessing is proposed to obtain reliable and high-quality modeling data for building the flatness classification model. DBN algorithm is used to construct the data-driven classification model of the hot-rolled strip flatness.Using cross entropy as the evaluation index, the influence of hyperparameters on the performance of DBN model is analyzed to determine the best one. Although the sample categories in testing set are unbalanced, the accuracy rates of the three flatness categories, including the large-medium wave, the no-medium wave and micro-medium wave, are 100%, 96.87% and 77.78% respectively, the wrong classification categories of flatness are mainly adjacent flatness categories, and their actual flatness values are very close. This method can accurately classify the finished hot-rolled strip flatness, and plays an important role in improving the calculation accuracy of the shape preset model and improving strip shape.

关键词

热连轧 / 带钢 / 板形预测 / 数据驱动建模 / 数据预处理 / 深度置信网络

Key words

hot rolling / strip / strip shape prediction / data-driven modeling / data preprocessing / deep belief networks

图表

引用本文

EndNote

Ris (Procite)

Bibtex

导出引用

李广焘, 龚殿尧, 张殿华, 等. 数据驱动的热连轧板平直度分类预测[J]. 钢铁, 2024, 59(7): 83-93 https://doi.org/10.13228/j.boyuan.issn0449-749x.20230690

LI Guangtao, GONG Dianyao, ZHANG Dianhua, et al. Data-driven classification prediction of hot strip flatness[J]. Iron and Steel, 2024, 59(7): 83-93 https://doi.org/10.13228/j.boyuan.issn0449-749x.20230690

参考文献

[1] 张殿华,彭文,丁敬国,等. 板带热连轧自动化系统的现状与展望[J]. 轧钢,2015,32(2):6.(ZHANG D H, PENG W, DING J G, et al. Automation system and its development for hot strip mill[J]. Steel Rolling, 2015, 32(2): 6.)
[2] 张进之,孙威,刘军,等. 板带连轧过程控制理论与模型的发展简述[J]. 冶金设备,2013(4):40.(ZHANG J Z, SUN W, LIU J, et al. Tandem rolling process dynamic theories and develop introduction of profile[J]. Metallurgical Equipment, 2013(4): 40.)
[3] 王国栋,张殿华,孙杰. 建设数据驱动的钢铁材料创新基础设施加速钢铁行业的数字化转型[J]. 冶金自动化,2023,47(1):2.(WANG G D, ZHANG D H, SUN J. Creating data-driven innovation infrastructure for iron and steel material to accelerate digital transformation of iron and steel industry[J]. Metallurgical Industry Automation, 2023, 47(1): 2.)
[4] 张殿华,彭文,孙杰,等. 板带轧制过程中的智能化关键技术[J]. 钢铁研究学报,2019,31(2):174.(ZHANG D H, PENG W, SUN J, et al. Key intelligent technology of steel strip rolling process[J]. Journal of Iron and Steel Research, 2019, 31(2): 174.)
[5] 王国栋,刘振宇,张殿华. RAL关于钢材热轧信息物理系统的研究进展[J]. 轧钢,2021,38(1):1.(WANG G D, LIU Z Y, ZHANG D H. Research progress on the cyber-physical system of steel hot rolling in RAL[J]. Steel Rolling, 2021, 38(1): 1.)
[6] 王国栋,刘振宇,张殿华,等. 钢铁企业创新基础设施及研究进展[J].钢铁,2023,58(9):2.(WANG G D, LIU Z Y, ZHANG D H, et al. Steel Enterprise Innovation Infrastructure(SEII) and its research development[J]. Iron and Steel, 2023, 58(9): 2.)
[7] 王国栋,刘振宇,张殿华,等. 材料科学技术转型发展与钢铁创新基础设施的建设[J]. 钢铁研究学报,2021,33(10):1003.(WANG G D, LIU Z Y, ZHANG D H, et al.Transformation and development of materials science and technology and construction of iron and steel innovation infrastructure[J]. Journal of Iron and Steel Research, 2021, 33(10): 1003.)
[8] 王国栋. 创建钢铁企业数字化创新基础设施加速钢铁行业数字化转型[J]. 轧钢,2022,39(6):2.(WANG G D. Creating a digital innovation infrastructure for iron and steel enterprises to accelerate the digital transformation of the iron and steel industry[J].Steel Rolling, 2022, 39(6): 2.)
[9] WANG Z H, GONG D Y, LI X, et al.Prediction of bending force in the hot strip rolling process using artificial neural network and genetic algorithm(ANN-GA)[J]. The International Journal of Advanced Manufacturing Technology, 2017, 93(9): 3325.
[10] LI X, LUAN F, WU Y.A comparative assessment of six machine learning models for prediction of bending force in hot strip rolling process[J]. Metals, 2020, 10(5): 1.
[11] WU Y, NI H C, LI X, et al.Prediction of bending force in the hot strip rolling process using multilayer extreme learning machine[J]. Mathematical Problems in Engineering, 2021, 2021: 1.
[12] ZHANG X L, ZHANG S Y, TAN G Z, et al.A novel method for flatness pattern recognition via least squares support vector regression[J]. Journal of Iron and Steel Research International, 2012, 19(3): 25.
[13] ZHANG X L, ZHANG S Y, ZHAO W B, et al.Flatness intelligent control via improved least squares support vector regression algorithm[J]. Journal of Central South University, 2013, 20(3): 688.
[14] ZHANG X L, ZHAO L, ZANG J Y, et al.Flatness intelligent control based on TS cloud inference neural network[J]. ISIJ International, 2014, 54(11): 2608.
[15] ZHANG X L, ZHAO L, ZANG J Y, et al.Visualization of flatness pattern recognition based on TS cloud inference network[J]. Journal of Central South University, 2015, 22(2): 560.
[16] ZHANG X L, XU T, ZHAO L, et al.Research on flatness intelligent control via GA-PIDNN[J]. Journal of Intelligent Manufacturing, 2015, 26(2): 359.
[17] ZHANG X L, CHENG L, HAO S, et al.The new method of flatness pattern recognition based on GA-RBF-ARX and comparative research[J]. Nonlinear Dynamics, 2016, 83(3): 1535.
[18] ZHANG X L, CHENG L, HAO S, et al.Optimization design of RBF‐ARX model and application research on flatness control system[J]. Optimal Control Applications and Methods, 2017, 38(1): 19.
[19] VANNOCCI M, RITACCO A, CASTELLANO A, et al.Flatness defect detection and classification in hot rolled steel strips using convolutional neural networks[C]//Advances in Computational Intelligence: 15th International Work-Conference on Artificial Neural Networks. Gran Canaria:[s.n.],2019: 220.
[20] 王青龙. 带材轧制过程数值模拟与板形控制理论模型研究[D]. 沈阳:东北大学,2020.(WANG Q L.Research on Numerical Simulation of Strip Rolling Process and Flatness Control Theoretical Model[D]. Shenyang: Northeastern University, 2020.)
[21] SIKDAR S, KUMARI S.Neural network model of the profile of hot-rolled strip[J]. The International Journal of Advanced Manufacturing Technology, 2009, 42(5/6): 450.
[22] WANG Z H, LIU Y M, GONG D Y, et al.A new predictive model for strip crown in hot rolling by using the hybrid AMPSO‐SVR‐based approach[J]. Steel Research International, 2018, 89(7): 1800003.
[23] WANG Z H, MA G S, GONG D Y, et al.Application of mind evolutionary algorithm and artificial neural networks for prediction of profile and flatness in hot strip rolling process[J]. Neural Processing Letters, 2019, 50(3): 2455.
[24] DENG J F, SUN J, PENG W, et al.Application of neural networks for predicting hot-rolled strip crown[J]. Applied Soft Computing, 2019, 78: 119.
[25] SUN J, DENG J F, PENG W, et al.Strip crown prediction in hot rolling process using random forest[J]. International Journal of Precision Engineering and Manufacturing, 2021, 22(2): 301.
[26] WU Y, LI X, LUAN F, et al.A novel GPR-based prediction model for strip crown in hot rolling by using the improved local outlier factor[J]. IEEE Access, 2020, 9: 458.
[27] HINTON G E.Deep belief networks[J]. Scholarpedia, 2009, 4(5): 5947.
[28] HINTON G E, OSINDERO S, TEH Y W.A fast learning algorithm for deep belief nets[J]. Neural Computation, 2006, 18(7): 1527.
[29] LI S, WANG J, CHEN S.Quality prediction of strip in finishing rolling process based on GBDBN-ELM[J]. Journal of Sensors, 2021, 2021: 1.
[30] 孙建亮,孙孟乾,郭贺松,等. 热轧带材多缺陷和单缺陷表面质量综合预报[J]. 钢铁,2021,56(1):51.(SUN J L, SUN M Q, GUO H S, et al. Comprehensive prediction of surface quality of hot-rolled strip with multiple and single defects[J]. Iron and Steel, 2021, 56(1): 51.)
[31] 张笑雄. 基于深度置信网络的带钢厚度预测[D]. 北京:冶金自动化研究设计院,2020.(ZHANG X X.Hot Strip Thickness Prediction Based on Deep Belief Network[D]. Beijing: Automation Research and Design Institute of Metallurgical Industry, 2020.)
[32] HU B H, WANG Y L, HAN Y H, et al.A flatness predict model based on deep belief network for steel rolling process[C]//2019 IEEE 5th International Conference on Computer and Communications (ICCC). Chengdu: IEEE, 2019: 235.
[33] HUANG S Y, TIAN T.Prediction of mechanical properties of hot rolled strip based on DBN and composite quantile regression[C]//Proceedings of the 2021 4th International Conference on Algorithms, Computing and Artificial Intelligence. Sanya:[s.n.], 2021: 1.
[34] ZHANG B S, HUO H, ZHI Y F.Prediction model of rolling force based on DBN[J]. Scientific Journal of Intelligent Systems Research Volume, 2021, 3(4): 347.
[35] PENG R R, ZHANG X Z, SHI P M.Bearing fault diagnosis of hot-rolling mill utilizing intelligent optimized self-adaptive deep belief network with limited samples[J]. Sensors, 2022, 22(20): 7815.
[36] 夏治国. 热连带钢产线多测温装置的故障诊断[D]. 沈阳:东北大学,2018.(XIA Z G.Fault Diagnosis of Multi-Temperature Measuring Device for Hot Strip Steel Production Line[D]. Shenyang: Northeastern University, 2018.)
[37] DENG J F, SUN J, PENG W, et al.Imbalanced multiclass classification with active learning in strip rolling process[J]. Knowledge-Based Systems, 2022, 255: 109754.
[38] FISCHER A, IGEL C.An introduction to restricted Boltzmann machines[C]//Iberoamerican Congress on Pattern Recognition. Buenos Aires: The Argentina Society for Pattern Recognition, 2012: 14.
[39] HINTON G E.Training products of experts by minimizing contrastive divergence[J]. Neural Computation, 2002, 14(8): 1771.
[40] GAO F, LI B, CHEN L, et al.A softmax classifier for high-precision classification of ultrasonic similar signals[J]. Ultrasonics, 2021, 112: 106344.
[41] GINZBURG V B.High-quality steel rolling: Theory and Practice[M]. New York: CRC Press, 1993.
[42] WANG L, HAN M, LI X J, et al.Review of classification methods on unbalanced data sets[J]. IEEE Access, 2021(9): 64606.
[43] KARTHIGA R, USHA G, RAJU N, et al.Transfer learning based breast cancer classification using one-hot encoding technique[C]//2021 International Conference on Artificial Intelligence and Smart Systems(ICAIS). Coimbatore: JCT College of Engineering and Technology, 2021: 115.