正在加载图片...
工程科学学报,第37卷,增刊1:6-11,2015年5月 Chinese Journal of Engineering,Vol.37,Suppl.1:6-11,May 2015 DOI:10.13374/j.issn2095-9389.2015.s1.002:http://journals.ustb.edu.cn 超宽带钢冷连轧机板形模式识别 包仁人”,张杰),李洪波”,贾生晖”,刘海军”,肖晟) 1)北京科技大学机械工程学院,北京1000832)武汉钢铁(集团)公司,武汉430083 ☒通信作者,E-mail:izhang@usth.cdu.cm 摘要引入以三次勒让德多项式表达的基本板形模式,并采用基于最小二乘原理的回归分析方法建立了板形模式识别模 型.对某厂2180 mm CVC(continuously variable crown)超宽带钢冷连轧机和I550 mm UCM(universal crown mill)普通宽带钢冷 连轧机的板形进行了识别与对比,结果表明对超宽带钢轧机板形而言,引入三次基本板形模式识别出的主要板形缺陷为“OS (operating side)单侧肋浪加Ds(driving side)单边浪”,未引入为“四分之一浪”;对普通宽带钢轧板形,识别结果均为“中浪”, 验证了在超宽带钢冷连轧机板形模式识别中引入三次基本板形模式的必要性,并讨论了此模型在超宽带钢冷连轧机板形缺 陷分析及非对称板形控制方面的应用价值 关键词超宽轧机:三次板形缺陷:模式识别:最小二乘回归:非对称板形控制 分类号TC333.7 Flatness pattern recognition of ultra-wide tandem cold rolling mill BAO Ren-ren”,ZHANG Jie',LI Hong-bo”,JIA Sheng-hui?,LIU Hai-jun》,XA0 Sheng》 1)School of Mechanical Engineering,University of Science and Technology Beijing,Beijing 100083,China 2)Wuhan Iron and Steel (Group)Corporation,Wuhan 430083,China Corresponding author,E-mail:jzhang@ustb.edu.en ABSTRACT A cubic flatness patter expressed by Legendre orthogonal polynomial was introduced to build a new flatness pattern recognition model,in which least squares regression was used.Using the new model,flatness pattern recognitions of 2180 mm CVC ultra-wide cold rolling mill and 1550 mm UCM common wide tandem cold rolling mill were carried out.The results show that the main flatness error of the strip produced by the ultra-wide mill is "OS cubic waves"considering the cubic flatness pattern,and "quarter waves"without the cubic pattern:the main flatness error of the strip produced by the common wide mill is "central waves"whether considering the cubic flatness pattern or not,therefore the necessity of introducing cubic flatness pattern for the new recognition model is verified.Finally,potential application of the new model in the asymmetric flatness control of ultra-wide tandem cold rolling mill is discussed. KEY WORDS ultra-wide mill;cubic flatness error;pattern recognition:least squares regression:asymmetric flatness control 板形模式识别作为板形控制和分析过程中的重要 近年来国内新建了一批可轧带钢宽度超过 一环,其本质就是采用一定的数学方法识别实测张应1800mm的超宽带钢冷连轧机,最大可轧宽度达 力信号中的板形缺陷类型.识别结果既可应用于板形 2080mm,所轧带钢宽厚比达到了4000因,相比普通宽 闭环控制过程,为确定板形控制策略提供依据0,也 带钢冷连轧机,超宽带钢冷连轧机轧辊具有更大的长 可应用于带钢板形缺陷分析,为提高冷轧设备板形控 径比,所轧带钢的板形应力分布更加复杂,如图1所 制能力提供指导四. 示.某厂的2I80mm全六辊CVC(continuously variable 收稿日期:201501-20 基金项目:中央高校基本科研业务费专项资金资助项目(FRF-TP-14019A2)工程科学学报,第 37 卷,增刊 1: 6--11,2015 年 5 月 Chinese Journal of Engineering,Vol. 37,Suppl. 1: 6--11,May 2015 DOI: 10. 13374 /j. issn2095--9389. 2015. s1. 002; http: / /journals. ustb. edu. cn 超宽带钢冷连轧机板形模式识别 包仁人1) ,张 杰1) ,李洪波1) ,贾生晖2) ,刘海军2) ,肖 晟2) 1) 北京科技大学机械工程学院,北京 100083 2) 武汉钢铁( 集团) 公司,武汉 430083  通信作者,E-mail: jzhang@ ustb. edu. cn 摘 要 引入以三次勒让德多项式表达的基本板形模式,并采用基于最小二乘原理的回归分析方法建立了板形模式识别模 型. 对某厂 2180 mm CVC( continuously variable crown) 超宽带钢冷连轧机和 1550 mm UCM( universal crown mill) 普通宽带钢冷 连轧机的板形进行了识别与对比,结果表明对超宽带钢轧机板形而言,引入三次基本板形模式识别出的主要板形缺陷为“OS ( operating side) 单侧肋浪加 DS( driving side) 单边浪”,未引入为“四分之一浪”; 对普通宽带钢轧板形,识别结果均为“中浪”, 验证了在超宽带钢冷连轧机板形模式识别中引入三次基本板形模式的必要性,并讨论了此模型在超宽带钢冷连轧机板形缺 陷分析及非对称板形控制方面的应用价值. 关键词 超宽轧机; 三次板形缺陷; 模式识别; 最小二乘回归; 非对称板形控制 分类号 TG333. 7 Flatness pattern recognition of ultra-wide tandem cold rolling mill BAO Ren-ren1) ,ZHANG Jie 1)  ,LI Hong-bo 1) ,JIA Sheng-hui 2) ,LIU Hai-jun2) ,XIAO Sheng2) 1) School of Mechanical Engineering,University of Science and Technology Beijing,Beijing 100083,China 2) Wuhan Iron and Steel ( Group) Corporation,Wuhan 430083,China  Corresponding author,E-mail: jzhang@ ustb. edu. cn ABSTRACT A cubic flatness pattern expressed by Legendre orthogonal polynomial was introduced to build a new flatness pattern recognition model,in which least squares regression was used. Using the new model,flatness pattern recognitions of 2180 mm CVC ultra-wide cold rolling mill and 1550 mm UCM common wide tandem cold rolling mill were carried out. The results show that the main flatness error of the strip produced by the ultra-wide mill is“OS cubic waves”considering the cubic flatness pattern,and“quarter waves”without the cubic pattern; the main flatness error of the strip produced by the common wide mill is“central waves”whether considering the cubic flatness pattern or not,therefore the necessity of introducing cubic flatness pattern for the new recognition model is verified. Finally,potential application of the new model in the asymmetric flatness control of ultra-wide tandem cold rolling mill is discussed. KEY WORDS ultra-wide mill; cubic flatness error; pattern recognition; least squares regression; asymmetric flatness control 收稿日期: 2015--01--20 基金项目: 中央高校基本科研业务费专项资金资助项目( FRF--TP--14--019A2) 板形模式识别作为板形控制和分析过程中的重要 一环,其本质就是采用一定的数学方法识别实测张应 力信号中的板形缺陷类型. 识别结果既可应用于板形 闭环控制过程,为确定板形控制策略提供依据[1],也 可应用于带钢板形缺陷分析,为提高冷轧设备板形控 制能力提供指导[2]. 近年 来 国 内 新 建 了 一 批 可 轧 带 钢 宽度超过 1800 mm 的 超 宽 带 钢 冷 连 轧 机,最 大 可 轧 宽 度 达 2080 mm,所轧带钢宽厚比达到了 4000 [3],相比普通宽 带钢冷连轧机,超宽带钢冷连轧机轧辊具有更大的长 径比,所轧带钢的板形应力分布更加复杂,如图 1 所 示. 某厂的 2180 mm 全六辊 CVC( continuously variable
向下翻页>>
©2008-现在 cucdc.com 高等教育资讯网 版权所有