|
|
Research and application of optimal control model for slab width of stainless steel |
LI Wei-gang, CHUN Li-liang, LI Yang, YI Cheng-xin |
Engineering Research Center for Metallurgical Automation and Measurement Technology of Ministry ofEducation, Wuhan University of Science and Technology, Wuhan 430081, Hubei, China |
|
|
Abstract The accurate control of slab width is one of the important factors to ensure slab quality in the continuous casting process. The slab width is too wide or too narrow will bring adverse effects to the rolling process. At present,the slab width is mainly controlled by adjusting parameters of the secondary cooling zone online by manual experience,which has the problems of low qualified rate and large fluctuation. Therefore,it is very important to realize automatic optimization control of slab width for stainless steel. Aiming at the characteristics of high dimension and nonlinear and multivariable coupling in the process control of slab width for stainless steel,an optimal control model of slab width based on the random forest and differential evolution(RF-DE) algorithm is proposed. Firstly,the factors affecting the slab width are selected from a large number of composition and process parameters based on expert experience,and then a data-driven prediction model of slab width is established by the random forest algorithm. Secondly,taking the secondary cooling process parameters of continuous casting as the decision variable,taking the absolute error between the predicted width of the prediction model and the target width of the slab as the objective function,the optimal control model of slab width with process constraints is constructed,and the differential evolution algorithm is used to optimize the above model,obtain the secondary cooling process parameters of continuous casting. Finally,the model is verified by the actual production data of a steel plant,and the experimental results show that compared with gradient boosting decision tree,support vector regression,multilayer perceptron and other models,random forest model has strong generalization ability and high accuracy, and is more suitable for the prediction of slab width,the mean absolute error is 0.047 2 mm. Moreover,the qualified rate of slab width for stainless steel is increased,and the proposed model has high control precision.
|
Received: 22 April 2021
|
|
|
|
[1] 温冬辉,吕阳,李震,等. 强碳化物形成元素与碳的配比关系和稳定化处理对310S奥氏体不锈钢析出相行为的影响[J]. 材料导报,2019,33(18):3101.(WEN Dong-hui,LÜ Yang,LI Zhen,et al. Effects of ratios of strong carbide-forming elements to C and stabilization treatment on precipitation behavior of 310S austenitic stainless steel[J]. Materials Reports,2019,33(18):3101.) [2] 邵毅,李彦默,刘晨曦,等. 低碳铁素体不锈钢高频直缝电阻焊管退火工艺优化[J]. 金属学报,2019,55(11):1367.(SHAO Yi,LI Yan-mo,LIU Chen-xi,et al. Annealing process optimization of high frequency longitudinal resistance welded low-carbon ferritic stainless steel pipe[J]. Acta Metallurgica Sinica,2019,55(11):1367.) [3] 莫金强,罗刚. 配分时间对低铬铁素体不锈钢组织和性能的影响[J]. 上海金属,2018,40(5):62.(MO Jin-qiang, LUO Gang. Effect of partitioning time on microstructure and properties of a low-chrome ferrite stainless steel[J]. Shanghai Metals,2018,40(5):62.) [4] 李海明,安婧婷. 利用激光快速测量热态铸坯宽度的方法[J]. 连铸,2020,45(5):73.(LI Hai-ming,AN Jing-ting. A method to measure width of hot casting blank by laser[J]. Continuous Casting,2020,45(5):73.) [5] 苏锦,许维胜. 铁素体不锈钢板坯宽度控制[C]//全国冶金自动化信息网2008年会论文集. 银川:全国冶金自动化信息网,2008:3.(SU Jin,XU Wei-sheng. Width control of ferrite stainless steel slab[C]//Proceedings of 2008 Annual Meeting of National Metallurgical Automation Information Network. Yinchuan:National Metallurgical Automation Information Network,2008:3.) [6] 田伟光,徐佳林,陆海飞,等. 410S不锈钢连铸坯宽度控制研究[J]. 连铸,2018,43(4):33.(TIAN Wei-guang,XU Jia-lin,LU Hai-fei,et al. Study on the width control of continuous casting slabs of 410S stainless steel[J]. Continuous Casting,2018,43(4):33.) [7] 王伟,孙仁宝,陈法涛. 钢水温度对铁素体不锈钢铸坯宽度的影响[C]//第十一届中国钢铁年会论文集. 北京:中国金属学会,2017:6.(WANG Wei,SUN Ren-bao,CHEN Fa-tao. Effect of molten steel temperature on width of ferrite stainless steel slab[C]// Proceedings of the 11th CSM Steel Congress. Beijing:The Chinese Society for Metals,2017:6.) [8] 苏锦. 基于宽度变化机理的铁素体不锈钢板坯宽度控制[J]. 冶金自动化,2020,44(4):52.(SU Jin. Width control of ferrite stainless steel slab based on width change mechanism[J]. Metallurgical Industry Automation,2020,44(4):52.) [9] 李江江,殷齐敏,赵刚,等. 400系不锈钢板坯宽度控制优化实践[J]. 山东冶金,2017,39(3):20.(LI Jiang-jiang,YIN Qi-min,ZHAO Gang,et al. Process optimization and improvements on the slab width of the 400 series stainless steel[J]. Shandong Metallurgy,2017,39(3):20.) [10] 翟莹莹,厉英,敖志广. 基于改进多目标遗传算法的连铸二冷过程优化[J]. 东北大学学报(自然科学版),2019,40(5):658.(ZHAI Ying-ying,LI Ying,AO Zhi-guang. Optimization of continuous casting secondary cooling based on an enhanced multi objective genetic algorithm[J]. Journal of Northeastern University(Natural Science),2019,40(5):658.) [11] YU Yang,LUO Xiao-chuan,ZHANG Hua-xi,et al. Dynamic optimization method of secondary cooling water quantity in continuous casting based on three-dimensional transient nonlinear convective heat transfer equation[J]. Applied Thermal Engineering,2019,160:113988. [12] LI Y R,ZHANG C N. Intelligent optimization strategy analysis of secondary cooling water distribution for billet[J]. Metalurgija,2019,58(3/4):239. [13] Breiman L. Random forests[J]. Machine Learning,2001,45(1):5. [14] Breiman L. Bagging predictors[J]. Machine Learning,1996,24(2):123. [15] 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.(YANG Wei,LI Wei-gang,ZHAO Yun-tao,et al. Mechanical property prediction of steel and influence factors selection based on random forests[J]. Iron and Steel,2018,53(3):44.) [16] 何伦茜. 基于随机森林的热轧带钢产品缺陷预测方法及系统开发[D]. 沈阳:东北大学,2017.(HE Lun-xi. Random Forest Based Defect Prediction Method for Hot Rolled Strip and System Development[D]. Shenyang:Northeastern University,2017.) [17] 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 Ying-jun,YONG Xiao-yue,LIU Ying-lin,et al. Random forest based quality analysis and prediction method for hot-rolled strip[J]. Journal of Northeastern University(Natural Science),2019,40(1):11.) [18] 刘小杰,邓勇,李欣,等. 基于大数据技术的高炉铁水硅含量预测[J]. 中国冶金,2021,31(2):10.(LIU Xiao-jie,DENG Yong,LI Xin,et al. Prediction of silicon content in hot molten of blast furnace based on big data technology[J]. China Metallurgy,2021,31(2):10.) [19] XIAO Xiong,DENG Dao-ming,XIAO Yu-xiong,et al. Research on prediction method of finish rolling power consumption of multi-specific strip steel based on random forest optimization model[C]//2020 39th Chinese Control Conference(CCC). Shenyang:IEEE,2020:5977. [20] Storn R,Price K. Differential evolution-A simple and efficient heuristic for global optimization over continuous spaces[J]. Journal of Global Optimization,1997,11(4):341. |
[1] |
CHE Zhi-chao, LIU Wei, YANG Shu-feng, LI Jing-she, LIU Chao, YUAN Hao. Modification of sulfide inclusions in 15-5PH steel by tellurium treatment[J]. Iron and Steel, 2022, 57(3): 36-43. |
[2] |
ZHANG Huayu1, LI Guodong2, QIN Fengming2, YANG Weimin2, ZHANG Wei3, CHEN Huiqin2. Creep behavior and life prediction of 309S heatresistant steel[J]. JOURNAL OF IRON AND STEEL RESEARCH , 2022, 34(3): 264-271. |
[3] |
WU Lei, KE Rui, HU Chengyang, WAN Xiangliang. Effect of deformation strengthening and phase reversion fine grain strengthening on microstructure and properties of Fe-18Cr-8Ni steel[J]. JOURNAL OF IRON AND STEEL RESEARCH , 2022, 34(2): 162-168. |
[4] |
WANG Yu, PENG Xiang-fei, LI Jun, YANG Yang, LI Guo-ping, LIU Yan-lin. Research progress on strengthening mechanism and ballistic performance of high nitrogen austenitic stainless steels[J]. Iron and Steel, 2022, 57(1): 28-38. |
[5] |
LI Xue-tong, SUN Ye-xuan, LIU Ya-xing, REN Ming-hua, TAN Chao-chao, JIA Wen-qiang. Optimal setting technology of process lubrication system for 3+1 emulsion system[J]. Iron and Steel, 2022, 57(1): 110-118. |
[6] |
ZHAO Li-jie, ZHANG Fang, PENG Jun, CHANG Hong-tao, MING Shou-lu, ZHU Rui-zong. Chemical composition and pore control of porous stainless steel prepared by rolling steel skin[J]. Iron and Steel, 2022, 57(1): 133-142. |
|
|
|
|