Abstract:The equipment and automation technology of the wide and heavy plate production line have reached a high level,but further improving the product yield has encountered a bottleneck. A steel plate image processing and contour feature extraction algorithm based on the fusion of deep learning algorithms was constructed,then a high-precision wide-width steel plate contour online detection device based on machine vision was developed,and high-precision online detection of steel plate contour was realized,which has a width perception accuracy of ±2 mm,a length perception error of less than 0.5%,and a lateral bending measurement accuracy of ±5 mm. The shear accuracy of irregular deformation area of head and tail is ±5 mm. Based on the monitoring data of machine vision,the rolling size,rolling process parameters and PVPC parameters were taken as input variables,and the metal volume in the deformation area of the plate head was taken as output variables,then a digital twin model of plan view pattern control(PVPC) based on random configuration network was established,and the PVPC curve of controllable points was set according to the steel plate under different broadening ratio and extension ratio. Based on the PVPC setting model and the feedback data of the plan view of machine vision,the volume change corresponding to the head controllable point setting model was calculated,and the corresponding adjustment values of three Gaussian curve functions were calculated from the change amount. Finally,a rolling optimization method of the PVPC model based on machine vision was established,which realizes intelligent prediction,dynamic setting and feedback optimization of controllable point PVPC. The practical application results show that the comprehensive yield based on the traditional plane shape control method is 92.28%,and the comprehensive yield based on the machine vision feedback-based PVPC CPS optimization system is increased to 93.36%,which is more than 1% higher than before. This method has created remarkable economic benefits for enterprises and greatly enhanced the market competitiveness of enterprises.
张殿华, 李旭, 丁敬国, 董子硕. 中厚板平面形状数字孪生模型与CPS优化系统[J]. 钢铁, 2023, 58(9): 137-147.
ZHANG Dianhua, LI Xu, DING Jingguo, DONG Zishuo. Digital twin model and CPS optimization system for plan view pattern control of wide and heavy plates[J]. Iron and Steel, 2023, 58(9): 137-147.
[1] 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.) [2] 王国栋. 高质量中厚板生产关键共性技术研发现状和前景[J]. 轧钢,2019,36(1):1. (WANG G D. Status and prospects of research and development of key common technologies for high-quality heavy and medium plate production[J]. Steel Rolling,2019,36(1):1.) [3] 张殿华,孙杰,丁敬国,等. 基于CPS架构的板带热轧智能化控制[J]. 轧钢,2021,38(2):1. (ZHANG D H,SUN J,DING J G,et al. Intelligent control of hot strip rolling based on CPS architecture[J].Steel Rolling,2021,38(2):1.) [4] 刘天纬,朱建军. 3 800 mm 中厚板轧机平面形状控制功能的应用[J]. 金属世界,2010(2):28. (LIU T W,ZHU J J. Application of plan view pattern control function for 3 800 mm plate mill[J]. Metal World,2010(2):28.) [5] DING J G,WANF G Q,HE Y,et al. Controllable points setting method for plan view pattern control in plate rolling process[J].Steel Research International,2020,91(1):2070011. [6] 李宏伟. 基于CPS的钢铁企业电网智能管控[J]. 冶金能源,2023,42(3):53. (LI H W. Intelligent management and control of power grid in iron and steel enterprises based on CPS[J]. Energy for Metallurgical Industry,2023,42(3):53.) [7] 王超,鲁效平,孙明,等. 数字孪生技术在智能轧钢车间的应用与探索[J]. 轧钢,2022,39(3):85. (WANG C,LU X P,SUN M,et al. Application of digital twin technology in intelligent steel rolling workshop[J]. Steel Rolling,2022,39(3):85.) [8] DING J G,HE Y H C,KONG L P,et al. Camber prediction based on fusion method with mechanism model and machine learning in plate rolling[J]. ISIJ International,2021,61(10):2540. [9] WU D,MA X,OLSON D. Financial distress prediction using integrated Z-score and multilayer perceptron neural networks[J]. Decision Support Systems,2022,159:113814. [10] WANG D,LI M. Stochastic configuration networks:Fundamentals and algorithms[J]. IEEE Transactions on Cybernetics,2017,47:3466. [11] WANG D,LI M. Robust stochastic configuration networks with kernel density estimation for uncertain data regression[J]. Information Science,2017,412/413:210. [12] WANG Z H,ZHANG D H,GONG D Y,et al. A new data-driven roll force and roll torque model based on FEM and hybrid PSO-ELM for hot strip rolling[J]. ISIJ International 2019,59:1604. [13] 菅志刚,金旭. 数据挖掘中数据预处理的研究与实现[J]. 计算机应用研究,2004,20(7):117. (JIAN Z G,JIN X. Research on data preprocess in data mining and its application[J]. Computer Application Research,2004,20(7):117.) [14] 计智伟,胡珉,尹建新. 特征选择算法综述[J]. 电子设计工程,2011,19(9):46. (JI Z W,HU M,YIN J X. A survey of feature selection algorithm[J]. Electronic Design Engineering,2011,19(9):46.) [15] CHEN T,GUESTRIN C. XGboost:A scalable tree boosting system[C]//Proceedings of the 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining.San Francisco,CA:ASSOC Comp Machinery,2016:785. [16] BUUREN S V,B0SHUIZEN H C,KNOOK D L. Multiple imputation of missing blood pressure covariates in survival analysis[J]. American Journal of Epidemiology,2003,157(1):78. [17] ARNOLD A M,KRONMAL R A. Multiple imputation of baseline data in the cardiovascular health study[J]. American Journal of Epidemiology,2003,157(1):74. [18] WILLMOTT C J,ROBESON S M,MATSUURA K. Geographic box plots[J]. Physical Geography,2013,28(4):331. [19] WILLIAMS J,HILL R R,PIGNATIELLO J J,et al. Wavelet analysis of variance box plot[J]. Journal of Applied Statistics,2021,28(1):1. [20] SONNTAG D,ZILLNER S,SMAGT P V D,et al. Overview of the CPS for smart factories project:Deep learning,knowledge acquisition,anomaly detection and intelligent user interfaces[J]. Springer International Publishing,2017,12(13):487. [21] ZHANG J J,ZONG C Q. Deep neural networks in machine translation:An overview[J]. IEEE Intelligent Systems,2015,30(5):16. [22] ZHANG K,ZUO W M,CHEN Y J,et al. Beyond a gaussian denoiser:Residual learning of deep CNN for image denoising[J]. IEEE Transactions on Image Processing,2017,26(7):3142. [23] HINTON G E. A practical guide to training restricted Boltzmann machines[J]. Momentum,2012,9(1):599. [24] MELCHIOR J,WANG N,WISKOTT L. Gaussian-binary restricted boltzmann machines for modeling natural image statistics[J]. Plos One,2017,12(2):246. [25] 纪英俊,勇晓玥,刘英林,等. 基于随机森林的热轧带钢质量分析与预测方法[J]. 东北大学学报(自然科学版),2019,40(1):11. (JI Y J,YONG X Y,LIU Y L,et al. Random forest based quality analysis and prediction method for hot-rolled strip[J]. Journal of Northeast University(Natural Science),2019,40(1):11.) [26] GUO J,WANG J,LI Q,et al. Construction of prediction model of neural network railway bulk cargo floating price based on random forest regression algorithm[J]. Neural Computing and Applications,2019,31(12):8139. [27] 吕何,孔政敏,张成刚. 基于混合优化随机森林回归的短期电力负荷预测[J]. 武汉大学学报(工学版),2020,53(8):704. (LÜ H,KONG Z M,ZHANG C G. Short-term power load forecasting based on hybrid optimization stochastic forest regression[J]. Journal of Wuhan University(Engineering),2020,53(8):704.) [28] JI A B,PANG J H,QIU H J. Support vector machine for classification based on fuzzy training data[J]. Expert Systems with Applications,2010,37(4):3495. [29] XI Z,YONG S,LEE J,et al. Customer churn prediction based on feature clustering and nonparallel support vector machine[J]. International Journal of Information Technology and Decision Making,2014,13(5):1013. [30] 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.