钢铁产品制造过程中,在线预测铸坯质量缺陷产生位置,并对存在缺陷的铸坯及时下线清理有助于提升连铸连轧生产稳定性,实现钢铁企业节能减排、绿色化生产。然而,连铸生产过程具有多变量、时变性和多态性等特点,必须结合设备、钢种、缺陷的特性,定制铸坯质量缺陷预测模型,才能够准确预测铸坯质量缺陷。因此,将数据通信技术、人工智能技术、C#与Matlab混合编程技术等相结合,建立了智能化铸坯质量在线判定系统,研究了铸坯质量预测模型智能化定制方法,并以板坯纵裂纹预测模型为例,介绍模型定制过程。研究结果表明,该方法能够辅助工艺工程师针对钢种及缺陷智能化定制预测模型,降低模型开发及优化难度,提高铸坯质量预测模型的可靠性。
Abstract
In the process of steel products manufacturing, online prediction of the location of casting billet quality defects and timely offline cleaning of the defective billet are helpful to improve the production stability of continuous casting and rolling, and realize energy saving, emission reduction and green production of iron and steel enterprises.However, the continuous casting process has the characteristics of multivariable, time-varying and polymorphism. Only by combining the characteristics of equipment, steel grade and defects, and customizing the slab quality defect prediction model, can the slab quality defects be accurately predicted.Therefore, combining data communication technology, artificial intelligence technology, C# and Matlab hybrid programming technology, the intelligent online judgment system for continuous casting slab is established, the intelligent customization method of slab quality prediction model is studied, and the customization process is introduced by taking slab longitudinal crack prediction model as an example. The results show that this method can assist process engineers to intelligently customize the prediction model for steel grades and defects, reduce the difficulty of model development and optimization, and improve the reliability of slab quality prediction model.
关键词
连铸坯 /
预测 /
纵裂纹 /
定制 /
混合编程
{{custom_keyword}} /
Key words
continuous casting slab /
prediction /
longitudinal crack /
customize /
mixed programming
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 朱苗勇.高拉速连铸过程传输行为特征及关键技术探析[J].钢铁,2021,56(7):1.
[2] Ryu S G, Choi D C, Jeon Y J, et al. Detection of scarfing faults on the edges of slab[J]. ISIJ International, 2014, 54(1): 112.
[3] Ai Y H, Xu K. Surface detection of continuous casting slabs based on curvelet transform and kernel locality preserving projections[J]. Journal of Iron and Steel Research International, 2013, 20(5): 80.
[4] Tirian G O, Filip I, ProsTean G. Adaptive control system for continuous steel casting based on neural networks and fuzzy logic[J]. Neurocomputing, 2014, 125: 236.
[5] 段海洋, 王旭东, 姚曼. 基于动态时间弯曲和k近邻分类预测铸坯纵裂纹[J]. 连铸, 2021(4): 66.
[6] 从俊强, 仇圣桃, 米进周, 等. 基于LMBP-MIV的板坯表面纵裂纹成因在线判定[J].炼钢, 2019, 35(4): 59.
[7] 从俊强, 仇圣桃, 徐学华,等. 连铸坯质量判定系统研究现状与发展趋势[J]. 铸造技术, 2017(10): 2329.
[8] 郭贤利, 彭世恒, 仇圣桃. 基于神经网络的连铸板坯质量在线诊断系统[J]. 冶金自动化, 2013, 37(3): 16.
[9] DU F, WANG X, LIU Y, et al. Prediction of longitudinal cracks based on a full-scale finite-element model coupled inverse algorithm for a continuously cast slab[J]. Steel Research International, 2017: 1700013.
[10] 邹雷雷, 黄俊雄, 李权辉, 等. 连铸坯裂纹与偏析预测研究进展[J]. 连铸, 2022(2): 2.
[11] 从俊强, 徐学华, 米进周, 等. 智能化铸坯质量在线判定系统开发[J]. 热加工工艺, 2018(1): 113.
[12] 姚光强, 陈立平. 基于COM技术的C#与Matlab混合编程[J]. 计算机工程, 2008(14): 87.
[13] 孙丽钢, 贾生, 李鑫. 宽厚板连铸坯表面裂纹的产生原因及控制[J]. 连铸, 2020(4):70.
[14] 陈明昕, 汪洪峰, 杨晓江. 连铸板坯表面纵裂纹的形成原因及控制[J]. 连铸, 2020(2): 41.
[15] Thomas B, Jenkins M, Mahapatra R. Investigation of strand surface defects using mould instrumentation and modelling[J]. Ironmaking and Steelmaking, 2013, 31(6): 485.
[16] 赵磊, 任海, 刘国贵. 连铸坯表面纵裂纹浅析[J]. 连铸, 2015 (4): 69.
[17] 刘鸿. 常用人工智能技术在钢铁领域中的应用概述[J]. 冶金自动化, 2019, 43(4): 24.
[18] 于加学, 孙杰, 张殿华. 基于深度学习的热轧带钢头部厚度的命中预测[J]. 钢铁, 2021, 56(9): 19.
[19] ZHAO J, DING H, ZHAO W, et al. Modelling of the hot deformation behaviour of a titanium alloy using constitutive equations and artificial neural network[J]. Computational Materials Science, 2014, 92:47.
[20] 黄华贵, 刘迎港, 黄海林, 等. 热轧板坯头部翘曲机器视觉检测与BP神经网络预测控制[J]. 中国冶金, 2022, 32(9): 85.
[21] 王东城, 徐扬欢, 段伯伟, 等. 数据驱动的热轧带钢边部线状缺陷智能预报模型[J]. 钢铁, 2020, 55(11): 82.
[22] 方洪鹰. 数据挖掘中数据预处理的方法研究[D]. 重庆:西南大学, 2009.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
安徽省自然科学基金项目(2008085QE225)
{{custom_fund}}
PDF(2133 KB)
