工程科学学报，第41卷，第7期：940-946.2019年7月 Chinese Journal of Engineering,Vol.41,No.7:940-946,July 2019 D0L:10.13374/j.issn2095-9389.2019.07.013;htp:/journals.usth.edu.cm 热轧带钢终轧温度的多模式控制 王海玉)四，郭立伟)，杨荃2)，王晓晨)，王淑志)，董立杰) 1)北京首钢自动化信息技术有限公司自动化研究所，北京100041 2)北京科技大学国家板带生产先进装备工程技术研究中心，北京100083 3)北京首钢股份有限公司，迁安064400 ☒通信作者，E-mail:wanghaiyu813@163.com 摘要结合国内某2250mm热连轧精轧机组，实现速度调节、机架间水调节、速度和机架间水耦合调节三种控制模式，能够 根据热连轧过程中的不同钢种和不同工况采用相适应的控制模式，以获取最佳的控制效果.同时，利用二次规划优化法在线 优化不同控制模式的调节量，以满足带钢全长终轧温度的控制要求.将多模式控制模型在线应用后，带钢终轧温度控制偏差 在±20℃以内，连续三个月命中率为99%以上.结果表明，该控制模型响应速度快，计算精度高，能够满足不同钢种和不同工 况下的终轧温度控制要求，从而提高带钢轧制稳定性和终轧温度控制精度，提升产品竞争力. 关键词带钢：热轧：终轧温度：多模式；二次规划法；加速度 分类号TG335.11 Multi-mode control of strip-finishing-temperature in hot-strip mills WANG Hai-yu,GUO Li-wei,YANG Quan2),WANG Xiao-chen2),WANG Shu-zhi),DONG Li-jie) 1)Automation Research Institute,Beijing Shougang Automation Information Technology Co.,Ltd.,Beijing 100041,China 2)National Engineering Research Centerof Flat Rolling Equipment,University of Science and Technology Beijing,Beijing 100083,China 3)Beijing Shougang Co.,Lid,Qianan 064400.China Corresponding author,E-mail:wanghaiyu813@163.com ABSTRACT At present,hot-rolled strip manufacturing has gradually exhibited more diversity and process complexity.Using the sin- gle control strategy,the traditional strip-finishing temperature-control mode shows some defects and deficiencies,for example,low con- trol precision,slow production rhythm,and great fluctuation in the strip-finishing-temperature curve,which cannot meet the require- ments for high precision and high-performance product control.For use with domestic 2250 mm hot-strip mills,a multi-mode control model was developed on a quadratic programming algorithm for the strip-finishing temperature.The proposed multi-mode control model has three control modes to regulate the speed,inter-stand cooling,and coupled speed and inter-stand cooling.To obtain the best con- trol effect,the appropriate control mode can be adopted depending on the different steels used and different working conditions in the hot-rolling process.At the same time,based on the cooling capacity of the adjustable rack and the calculated strip-finishing tempera- ture,Newton-Raphson iteration and the acceleration calculation model were used to calculate the large acceleration region and the quadratic programming optimization method to optimize the on-line adjustment of different control modes to meet all the strip-finishing temperature-control requirements.The on-line application of the proposed multi-model realized a99%hit rate or better on the strip-fin- ishing temperature for three consecutive months,with a deviation in the strip-finishing-temperature control of 20 C.A 97.2%hit rate or better was realized on the strip-finishing temperature for three consecutive months with a deviation in the strip-finishing-tempera- ture control of 15 C.These results show that the control model has the advantages of a fast response speed and high precision and 收稿日期：2018-05-23 基金项目：国家自然科学基金资助项目(51604024)：中央高校基本科研业务费专项资金资助项目(FRF-TP-17-002A2):北京市自然科学基金 资助项目工程科学学报,第 41 卷,第 7 期:940鄄鄄946,2019 年 7 月 Chinese Journal of Engineering, Vol. 41, No. 7: 940鄄鄄946, July 2019 DOI: 10. 13374 / j. issn2095鄄鄄9389. 2019. 07. 013; http: / / journals. ustb. edu. cn 热轧带钢终轧温度的多模式控制 王海玉1) 苣 , 郭立伟1) , 杨 荃2) , 王晓晨2) , 王淑志3) , 董立杰3) 1) 北京首钢自动化信息技术有限公司自动化研究所, 北京 100041 2) 北京科技大学国家板带生产先进装备工程技术研究中心, 北京 100083 3) 北京首钢股份有限公司, 迁安 064400 苣 通信作者, E鄄mail:wanghaiyu813@ 163. com 摘 要 结合国内某 2250 mm 热连轧精轧机组,实现速度调节、机架间水调节、速度和机架间水耦合调节三种控制模式,能够 根据热连轧过程中的不同钢种和不同工况采用相适应的控制模式,以获取最佳的控制效果. 同时,利用二次规划优化法在线 优化不同控制模式的调节量,以满足带钢全长终轧温度的控制要求. 将多模式控制模型在线应用后,带钢终轧温度控制偏差 在 依 20 益以内,连续三个月命中率为 99% 以上. 结果表明,该控制模型响应速度快,计算精度高,能够满足不同钢种和不同工 况下的终轧温度控制要求,从而提高带钢轧制稳定性和终轧温度控制精度,提升产品竞争力. 关键词 带钢; 热轧; 终轧温度; 多模式; 二次规划法; 加速度 分类号 TG335郾 11 收稿日期: 2018鄄鄄05鄄鄄23 基金项目: 国家自然科学基金资助项目(51604024);中央高校基本科研业务费专项资金资助项目(FRF鄄鄄TP鄄鄄17鄄鄄002A2);北京市自然科学基金 资助项目 Multi鄄mode control of strip鄄finishing鄄temperature in hot鄄strip mills WANG Hai鄄yu 1) 苣 , GUO Li鄄wei 1) , YANG Quan 2) , WANG Xiao鄄chen 2) , WANG Shu鄄zhi 3) , DONG Li鄄jie 3) 1) Automation Research Institute,Beijing Shougang Automation Information Technology Co. , Ltd. , Beijing 100041, China 2) National Engineering Research Centerof Flat Rolling Equipment, University of Science and Technology Beijing, Beijing 100083, China 3) Beijing Shougang Co. ,Ltd, Qianan 064400, China 苣 Corresponding author, E鄄mail: wanghaiyu813@ 163. com ABSTRACT At present, hot鄄rolled strip manufacturing has gradually exhibited more diversity and process complexity. Using the sin鄄 gle control strategy, the traditional strip鄄finishing temperature鄄control mode shows some defects and deficiencies, for example, low con鄄 trol precision, slow production rhythm, and great fluctuation in the strip鄄finishing鄄temperature curve, which cannot meet the require鄄 ments for high precision and high鄄performance product control. For use with domestic 2250 mm hot鄄strip mills, a multi鄄mode control model was developed on a quadratic programming algorithm for the strip鄄finishing temperature. The proposed multi鄄mode control model has three control modes to regulate the speed, inter鄄stand cooling, and coupled speed and inter鄄stand cooling. To obtain the best con鄄 trol effect, the appropriate control mode can be adopted depending on the different steels used and different working conditions in the hot鄄rolling process. At the same time, based on the cooling capacity of the adjustable rack and the calculated strip鄄finishing tempera鄄 ture, Newton鄄鄄Raphson iteration and the acceleration calculation model were used to calculate the large acceleration region and the quadratic programming optimization method to optimize the on鄄line adjustment of different control modes to meet all the strip鄄finishing temperature鄄control requirements. The on鄄line application of the proposed multi鄄model realized a 99% hit rate or better on the strip鄄fin鄄 ishing temperature for three consecutive months, with a deviation in the strip鄄finishing鄄temperature control of 依 20 益 . A 97郾 2% hit rate or better was realized on the strip鄄finishing temperature for three consecutive months with a deviation in the strip鄄finishing鄄tempera鄄 ture control of 依 15 益 . These results show that the control model has the advantages of a fast response speed and high precision and