| 1,225 | 42 | 212 |
| 下载次数 | 被引频次 | 阅读次数 |
以所收集380CL和SPA-H两个钢种的热连轧工业生产大数据为样本,采用多元线性回归、BP神经元网络、随机森林与XGBoost这4种机器学习算法建立带钢力学性能预测模型,并对预测结果的偏差进行了比较。结果表明,随机森林算法的预测精度高于其它算法,屈服强度、抗拉强度和伸长率的预测偏差的标准差分别为15 MPa、12 MPa和2.3%。以样本性能均值为基数,满足±3σ标准要求的预测相对误差分别可以达到15%、7%和22%。该算法可以应用于热轧带钢在线性能预测模型中,作为监控、分析带钢力学性能及优化工艺的有效工具。基于随机森林算法开发的热轧带钢力学性能在线预测模型已经在首钢京唐钢铁联合有限责任公司2250热轧生产线上得到实际应用。
Abstract：Taking the collected hot continuous rolling industry production big data of two steel grades of 380CL and SPA-H as the sample,four machine learning algorithms including multiple linear regression,BP neural network,random forest and XGBoost were adopted to establish the prediction models of mechanical properties strip steel,and then the deviations of prediction results were compared. The results show that the prediction accuracy of random forest algorithm is better than the others. The standard deviations of predicted deviations of yield strength,tensile strength and elongation are 15 MPa,12 MPa and 2.3%,respectively. Compared to the mean values of sample,the relative tolerances of prediction satisfying with the ± 3σ standard are 15%,7% and 22%,respectively. The random forest algorithm can be used in online property prediction model for hot rolled strip as an effective tool to monitor and analyze the mechanical properties and optimize the processing. The online mechanical property prediction model based on random forest algorithm is developed and applied on 2250 hot strip mill in Shougang Jingtang United Iron and Steel Co.,Ltd..
[1]YADA H.Prediction of microstructural changes and mechanical properties in hot strip rolling[M].Proceedings of the Metallurgical Society of the Canadian Institute of Mining&Metallurgy,1988.
[2]SUEHIRO M,SATO K,TSUKANO Y,et al.Computer modeling of microstructural change and strength of low carbon steel in hot strip rolling[J].Transactions ISIJ,1987,(6):439-445.
[3]SENUMA T,SUEHIRO M,YADA H.Mathematical models for predicting microstructural evolution and mechanical properties of hot strips[J].ISIJ International,1992,(3):423-432.
[4]HODGSON P D,GIBB R K.A mathematical model to predict the mechanical properties of hot rolled C-Mn and microalloyed steels[J].ISIJ International,1992,(12):1329-1338.
[5]许云波,刘相华,王国栋.热轧带钢组织-性能预测和控制模型及其应用前景[J].钢铁研究学报,2002,14 (1):65-68.XU Yunbo,LIU Xianghua,WANG Guodong.Prediction-control model for microstructure and property of hot-rolled steel strip and their application[J].Journal of Iron and Steel Research,2002,14 (1):65-68.
[6]干勇.薄板坯连铸连轧(TSCR)热轧过程组织性能预报技术的开发[J].钢铁,2003,38 (8):10-15.GAN Yong.The R&D of process modeling in thin slab hot strip rolling[J].Iron and Steel,2003,38 (8):10-15.
[7]沙孝春,兰勇军,李殿中,等.带钢热轧组织性能预测与控制技术应用与发展[J].轧钢,2003,20 (3):23-26.SHA Xiaochun,LAN Yongjun,LI Dianzhong,et al.Application and development of microstructure and performance forecast and control technique of hot strip rolling[J].Steel Rolling,2003,20 (3):23-26.
[8]SHULKOSKY R,ROSBURG D,CHAPMAN J,et al.A microstructure evolution model used for hot strip rolling[C]//Conf.Proc.Materials Science&Technology,2003:509-527.
[9]余万华,韩静涛,佘广夫,等.HSMM软件在攀枝花钢铁公司热轧板厂的应用[J].钢铁研究学报,2006,18 (11):171-173.YU Wanhua,HAN Jingtao,SHE Guangfu,et al.Application of HSMM software in hot strip mill of Panzhihua Steel Company[J].Journal of Iron and Steel Research,2006,18 (11):171-173.
[10]ANDORFER J,HRIBERNIG G,SAMOILOV A,et al.Properties of hot rolled strip obtained by calculation or testing-A critical comparison[J].ECNDT,2006,1:1-16.
[11]PERLADE A,GRANDEMANGE D,IUNG T.Application of microstructural modelling for quality control and process improvement in hot rolled steels[J].Ironmaking and Steelmaking,2005,(4):299-302.
[12]OHARA K,TSUGENO M,TEZUKA T,等.热轧钢材料性能的预测和控制系统[C]//中国国际钢铁大会,北京,2004:1-5.OHARA K,TSUGENO M,TEZUKA T,et al,Prediction and control system of hot rolled steel material properties[C]//China International Iron and Steel Conference.Beijing,2004:1-5.
[13]ANDORFER J,HRIBERNIG G,LUGE A,et al.使用VAI-QStrip系统首次实现热轧带钢力学性能的全面控制[J].钢铁,2001,36 (1):42-46.ANDORFER J,HRIBERNIG G,LUGE A,et al.For the first time ever:Full metallurgical control of the mechanical properties of hotrolled strip with VAI-Q STRIP[J].Iron and Steel,2001,36(1):42-46.
[14]MUKHOPADHYAY A,POLO A,PEROTTI F,et al.System for controlling mechanical properties of hot-rolled coil[J].Materials and Manufacturing Processes,2010,(25):76-84.
[15]干勇,刘正东,王国栋,等.组织性能预报系统在宝钢2050热轧生产线的在线应用[J].钢铁,2006,41 (3):39-43.GAN Yong,LIU Zhengdong,WANG Guodong,et al.On-line application of structure and property prediction system on hot rolling line on 2050 HSM at Baosteel[J].Iron and Steel,2006,41(3):39-43.
[16]郭朝晖,张群亮.宝钢力学性能预报模型的研究与开发[J].宝钢技术,2011,(5):1-6.GUO Chaohui,ZHANG Qunliang.Research and development of the mechanical property prediction model in Baosteel[J].Baosteel Technical Research,2011,(5):1-6.
[17]CHOQUET P,FABREGUE P,GIUSTI J.Modeling of forces,structure&final properties during the hot rolling process on the hot strip mill[C]//Intl.Symp.on Mathematical Modeling of Hot Rolling of Steel,1990:34-44.
[18]KERN A,DEGENKOLBE J,MOSGEN B,et al.Computer modelling for the prediction of microstructure development and mechanical properties of HSLA steel plates[J].ISIJ International,1992,(3):387-394.
[19]KRISHNAMOORTHY P.Regression analysis modelling of flat rolled sheet steel characteristics vs.processing variables[D].Chicago:Illinois Institute of Technology,1995.
[20]JIRATTHANAKUL N,CHANDRA-AMBHORN S.Prediction of the mechanical properties of hot-rolled low carbon steel strips in correlation to chemical compositions and rolling conditions[J].Key Engineering Materials,2011,462-463:401-406.
[21]刘振宇,王昭东,王国栋,等.应用神经网络预测热轧C-Mn钢力学性能[J].钢铁研究学报,1995,30 (4):61-66.LIU Zhenyu,WANG Zhaodong,WANG Guodong,et al.Predicting mechanical properties of hot-rolled steels by using neural networks[J].Journal of Iron and Steel Research,1995,30 (4):61-66.
[22]LIU Z Y,WANG W D,GAO W.Prediction of the mechanical properties of hot-rolled C-Mn using artificial neural networks[J].Journal of Materials Processing Technology,1996,57 (3-4):332-336.
[23]MYLLYKOSKI P.A study on the causes of deviation in mechanical properties of thin steel sheets[J].Journal of Materials Processing Technology,1998,79 (1-3):9-13.
[24]KUSIAK J,KUZIAK R,Modelling of microstructure and mechanical properties of steel using the artificial neural network[J].Journal of Materials Processing Technology,2002,127 (1):115-121.
[25]JONES D M,WATTON J,BROWN K J.Comparison of hot rolled steel mechanical property prediction models using linear multiple regression,non-linear multiple regression and non-linear artificial neural networks[J].Ironmaking and Steelmaking,2005,(5):435-442.
[26]刘相华,赵启林,黄贞益.人工智能在轧制领域中的应用进展[J].轧钢,2017,34 (4):1-5.LIU Xianghua,ZHAO Qilin,HUANG Zhenyi.Prospect of artificial intelligent application in rolling[J].Steel Rolling,2017,34(4):1-5.
[27]莫春立,李强,李殿中,等.应用回归和神经网络方法预测热轧带钢性[J].金属学报,2003,39 (10):1110-1114.MO Chunli,LI Qiang,LI Dianzhong,et al.Prediction of the properties of hot rolling strip by using regression and neural network[J].Acta Metallrugica Sinica,2003,39 (10):1110-1114.
[28]MUKHOPADHYAY A,IQBAL A.Prediction of mechanical properties of hot rolled,low-carbon steel strips using artificial neural network[J].Materials and Manufacturing Processes,2005,(5):793-812.
[29]王丹民,李华德,周建龙,等.热轧带钢力学性能预测模型及其应用[J].北京科技大学学报,2006,28 (7):687-690.WANG Danmin,LI Huade,ZHOU Jianlong,et al.Quality prediction model of the mechanical properties of hot-rolled steel strips and its application[J].Journal of University of Science and Technology Beijing,2006,28 (7):687-690.
[30]SARAVANAKUMAR P,JOTHIMANI V,SURESHBABU L,et al.Prediction of mechanical properties of low carbon steel in hot rolling process using artificial neural network model[J].Procedia Engineering,2012,34:3418-3425.
[31]DAS S K.Neural network modelling of flow stress and mechanical prediction of mechanical properties of low carbon steel in hot rolling process using artificial neural network model properties for hot strip rolling of TRIP steel using efficient learning algorithm[J].Ironmaking and Steelmaking,2013,(4):298-304.
[32]MOHANTY I,SARKAR S,JHA B,et al.Online mechanical property prediction system for hot rolled IF steel[J].Ironmaking and Steelmaking,2014,(8):618-627.
[33]陈超,李忠辉,熊昆鹏.神经网络模型在沙钢热轧性能预测中的应用[J].冶金自动化,2014,38 (S1):631-633.CHEN Chao,LI Zhonghui,XIONG Kunpeng.Application of neural network model in properties prediction of hot rolling steel in Sha Steel[J].Metallurgical Industry Automation,2014,38(S1):631-633.
[34]贾涛,刘振宇,胡恒法,等.基于贝叶斯神经网络的SPA-H热轧板力学性能预测[J].东北大学学报(自然科学版),2008,29 (4):521-524.JIA Tao,LIU Zhenyu,HU Hengfa,et al,Mechanical property prediction for hot rolled SPA-H steel using Bayesian neural network[J].Journal of Northeastern University (Natural Science),2008,29 (4):521-524.
[35]AGARWAL K,SHIVPURI R.An on-line hierarchical decomposition based bayesian model for quality prediction during hot strip rolling[J].ISIJ International,2012,(10):1862-1871.
[36]刘振宇,曹光明,周晓光,等.组织性能预测技术及其在智能热轧中的核心作用[J].轧钢,2019,36 (2):1-7.LIU Zhenyu,CAO Guangming,ZHOU Xiaoguang,et al.The predicting technologies for microstructure and properties and their core roles in smart hot rolling processes[J].Steel Rolling,2019,36(2):1-7.
[37]胡石雄,李维刚,杨威.基于卷积神经网络的热轧带钢力学性能预报[J].武汉科技大学学报(自然科学版),2018,41(5):338-344.HU Shixiong,LI Weigang,YANG Wei.Mechanical property prediction of hot-rolled strip based on convolutional neural network[J].Journal of Wuhan University of Science and Technology,2018,41 (5):338-344.
[38]马湧,王晓鹏,马莎莎.基于Keras深度学习框架下BP神经网络的热轧带钢力学性能预测[J].冶金自动化,2019,43(2):6-10.MA Yong,WANG Xiaopeng,MA Shasha.Prediction of mechanical properties of hot rolled strip based on BP neural network under Keras deep learning framework[J].Metallurgical Industry Automation,2019,43 (2):6-10.
[39]宋勇,李博,刘超,等.基于改进堆叠自编码器的当力学性能预报模型[J].冶金自动化,2020,44 (6):2-10.SONG Yong,LI Bo,LIU Chao,et al.Prediction model of mechanical properties of hot rolled strip based on improved stacked self-encoder[J].Metallurgical Industry Automation,2020,44(6):2-10.
[40]杨威,李维刚,赵云涛,等.基于随机森林的钢材性能预报与影响因素筛选[J].钢铁,2018,53 (3):44-49.YANG Wei,LI Weigang,ZHAO Yuntao,et al.Mechanical property prediction of steel and influence factors selection based on random forests[J].Iron and Steel,2018,53 (3):44-49.
[41]朱永波,张飞,张勇军,等.基于粒子群优化的带钢凸度神经网络预测模型研究[J].冶金自动化,2019,43 (2):11-15,28.ZHU Yongbo,ZHANG Fei,ZHANG Yongjun,et al.Particle swarm optimized neural network for strip crown prediction model research[J].Metallurgical Industry Automation,2019,43 (2):11-15,28.
[42]曹建国,江军,赵秋芳,等.基于数据挖掘的宽厚板板凸度控制[J].中南大学学报(自然科学版),2019,50 (11):2743-2752.CAO Jianguo,JIANG Jun,ZHAO Qiufang,et al.Wide and heavy plate crown control based on data mining[J].Journal of Central South University (Science and Technology),2019,50(11):2743-2752.
[43]李飞飞,宋勇,刘超,等.热轧带钢力学性能预报模型的误差分布建模研究[J].冶金自动化,2019,43 (6):28-33.LI Feifei,SONG Yong,LIU Chao,et al.Research on error distribution modeling of mechanical performance prediction model for hot rolled strip[J].Metallurgical Industry Automation,2019,43(6):28-33.
[44]陈金香,赵峰,孙彦广,等.基于大数据和XGBoost的热轧板带力学性能预测模型[C]//中国金属学会.第十二届中国钢铁年会,北京,2019:1.CHEN Jinxiang,ZHAO Feng,SUN Yanguang,et al,A prediction model based on big data and XGBoost for the mechanical properties of hot-rolled strips[C].//The Chinese Society for Metals,The12nd Chinese Iron and Steel Conference.Beijing,2019:1.
[45]章顺虎,姜兴睿,尤凤翔,等.融合工业大数据的热轧厚板轧制力模型研究[J].精密成形工程,2020,12 (2):8-14.ZHANG Shunhu,JIANG Xingrui,YOU Fengxiang,et al.Investigation on the model of rolling force by integrating industrial big data[J].Journal of Netshape Forming Engineering,2020,12 (2):8-14.
[46]陈丹,邵健,殷实,等.基于大数据平台的冷连轧轧制力自学习模型优化[J].冶金自动化,2020,44 (6):25-29,61CHEN Dan,SHAO Jian,YIN Shi,et al.Optimization of selflearning model of cold rolling force based on big data platform[J].Metallurgical Industry Automation,2020,44 (6):25-29,61
[47]刘元铭,王振华,王涛,等.热轧带钢出口凸度数据驱动建模及智能化预测分析[J].中国机械工程,2020,31 (22):2728-2733.LIU Yuanming,WANG Zhenhua,WANG Tao,et al.Data-driven modeling and intelligent prediction analysis for hot strip outlet crowns[J].China Mechanical Engineering,2020,31 (22):2728-2733.
[48]BREIMAN L,FRIEDMAN J H,OLSHEN R A,et al.Classification and regression trees[M].Boca Raton:Chapman&Hall/CRC,1984.
[49]BREIMAN L.Random forests[J].Machine Learning,2001,(45):5-32.
[50]VěRA K.Kolmogorov's theorem and multilayer neural networks[J].Neural Networks,1992,5:501-506.
[51]CHEN T Q,GUESTRIN C.XGBoost:A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining-KDD’16-XGBoost.San Francisco,2016:785-794.
基本信息:
中图分类号:TG335.56
引用信息:
[1]王晓东,安瑞东.基于机器学习的热轧带钢力学性能预测模型及应用[J].塑性工程学报,2021,28(10):155-165.
基金信息:
国家重点研发专项(2017YFB0304100)
2021-10-25
2021-10-25
2021-10-25