工程科学学报，第41卷，第4期：521-527,2019年4月 Chinese Journal of Engineering,Vol.41,No.4:521-527,April 2019 DOI:10.13374/j.issn2095-9389.2019.04.013:http://journals.ustb.edu.cn 基于函数型数字孪生模型的转炉炼钢终点碳控制技术 徐 钢12》，黎敏)区，徐金梧2》，贾春挥》，陈兆富) 1)北京科技大学计算机与通讯工程学院，北京1000832)钢铁共性技术协同创新中心，北京100083 3)鞍钢股份鲅鱼圈钢铁分公司，鲅鱼图115007 ☒通信作者，E-mail:limin@usth.cdu.cn 摘要由于转炉治炼过程中的热力学和动力学反应复杂，副枪控制模型和传统的烟气分析模型存在很大的局限性，导致了 转炉治炼终点碳含量的预测精度偏低，是实现智能炼钢的主要技术瓶颈.针对上述问题，提出了基于烟气分析的炼钢过程函 数型数字孪生模型.首先，利用烟气分析得到连续监测的实时数据，以此来实时监控转炉熔池内钢水的碳氧反应状态：然后， 根据熔池反应所处的不同阶段，利用函数型数据分析方法建立吹炼前期和吹炼后期的函数型预测模型：在此基础上，按照吹 炼前期和吹炼后期这两个阶段来分别自动修正模型中的系数函数，从而能在复杂的实际工况条件下完成对熔池碳含量的准 确预测.通过260t氧气转炉的工业应用实例，证实函数型数字孪生模型具有良好的自学习和自适应能力，对异常冶炼状态具 有良好的鲁棒性，可以实现全过程的熔池碳含量动态预测，终点碳质量分数在±0.02%范围内的命中率为95%·利用函数型 数字孪生模型在拉碳阶段对钢水中碳含量的预测值来控制终吹点.更为重要的是，在保证入炉原料成分、温度、质量等参数稳 定的前提下，采用该模型可以有望取消基于副枪的停吹取样步骤，从而降低生产成本，提高产品质量和生产效率，具有广泛的 工业应用前景 关键词转炉炼钢：数字孪生模型：烟气分析：函数型数据分析：终点碳控制 分类号TP277 Control technology of end-point carbon in converter steelmaking based on functional digital twin model XU Gang'2),LI Min2,XU Jin-wu2,JIA Chun-hui),CHEN Zhao-fu) 1)School of Computer and Communication Engineering,University of Science and Technology Beijing,Beijing 100083,China 2)Collaborative Innovation Center of Steel Technology,University of Science and Technology Beijing,Beijing 100083,China 3)Ansteel Co.,Lid.Bayuquan Iron and Steel Branch,Bayuquan 115007,China Corresponding author,E-mail:limin@ustb.edu.cn ABSTRACT An important part of the iron-and-steel production process,converter steelmaking is the most widely used and efficient method of steelmaking in the world.Under the requirements of "China Manufacturing 2025,"ensuring intelligent steelmaking,impro- ving smelting production efficiency,and reducing production cost are major concerns that should be addressed urgently in converter steelmaking.Owing to the complex thermodynamic and dynamic reactions in the converter smelting process,sublance control and tradi- tional flue-gas analysis models have limitations that result in low prediction accuracy of the end-point carbon in converter smelting, thereby causing the main technical bottleneck in intelligent steelmaking.Therefore,a functional digital twin model of the steelmaking process based on flue-gas analysis was proposed.First,continuously monitored real-time data were obtained by flue gas analysis to ob- serve the carbon and oxygen reaction state of molten steel in the converter.Then,according to various stages of the converter reaction, the functional data analysis method was used to establish the functional prediction models for the early and late stages of blowing.The 收稿日期：201807-23 基金项目：国家高技术研究发展计划(863计划)资助项目(2014AA041801-2)工程科学学报，第 41 卷，第 4 期: 521--527，2019 年 4 月 Chinese Journal of Engineering，Vol． 41，No． 4: 521--527，April 2019 DOI: 10． 13374 /j． issn2095--9389． 2019． 04． 013; http: / /journals． ustb． edu． cn 基于函数型数字孪生模型的转炉炼钢终点碳控制技术 徐 钢1，2) ，黎 敏2) ，徐金梧2) ，贾春辉3) ，陈兆富3) 1) 北京科技大学计算机与通讯工程学院，北京 100083 2) 钢铁共性技术协同创新中心，北京 100083 3) 鞍钢股份鲅鱼圈钢铁分公司，鲅鱼圈 115007  通信作者，E-mail: limin@ ustb． edu． cn 摘 要 由于转炉冶炼过程中的热力学和动力学反应复杂，副枪控制模型和传统的烟气分析模型存在很大的局限性，导致了 转炉冶炼终点碳含量的预测精度偏低，是实现智能炼钢的主要技术瓶颈． 针对上述问题，提出了基于烟气分析的炼钢过程函 数型数字孪生模型． 首先，利用烟气分析得到连续监测的实时数据，以此来实时监控转炉熔池内钢水的碳氧反应状态; 然后， 根据熔池反应所处的不同阶段，利用函数型数据分析方法建立吹炼前期和吹炼后期的函数型预测模型; 在此基础上，按照吹 炼前期和吹炼后期这两个阶段来分别自动修正模型中的系数函数，从而能在复杂的实际工况条件下完成对熔池碳含量的准 确预测． 通过 260 t 氧气转炉的工业应用实例，证实函数型数字孪生模型具有良好的自学习和自适应能力，对异常冶炼状态具 有良好的鲁棒性，可以实现全过程的熔池碳含量动态预测，终点碳质量分数在 ± 0. 02% 范围内的命中率为 95% ． 利用函数型 数字孪生模型在拉碳阶段对钢水中碳含量的预测值来控制终吹点． 更为重要的是，在保证入炉原料成分、温度、质量等参数稳 定的前提下，采用该模型可以有望取消基于副枪的停吹取样步骤，从而降低生产成本，提高产品质量和生产效率，具有广泛的 工业应用前景． 关键词 转炉炼钢; 数字孪生模型; 烟气分析; 函数型数据分析; 终点碳控制 分类号 TP277 收稿日期: 2018--07--23 基金项目: 国家高技术研究发展计划( 863 计划) 资助项目( 2014AA041801--2) Control technology of end-point carbon in converter steelmaking based on functional digital twin model XU Gang1，2) ，LI Min2)  ，XU Jin-wu2) ，JIA Chun-hui3) ，CHEN Zhao-fu3) 1) School of Computer and Communication Engineering，University of Science and Technology Beijing，Beijing 100083，China 2) Collaborative Innovation Center of Steel Technology，University of Science and Technology Beijing，Beijing 100083，China 3) Ansteel Co． ，Ltd． Bayuquan Iron and Steel Branch，Bayuquan 115007，China  Corresponding author，E-mail: limin@ ustb． edu． cn ABSTＲACT An important part of the iron-and-steel production process，converter steelmaking is the most widely used and efficient method of steelmaking in the world． Under the requirements of“China Manufacturing 2025，”ensuring intelligent steelmaking，impro￾ving smelting production efficiency，and reducing production cost are major concerns that should be addressed urgently in converter steelmaking． Owing to the complex thermodynamic and dynamic reactions in the converter smelting process，sublance control and tradi￾tional flue-gas analysis models have limitations that result in low prediction accuracy of the end-point carbon in converter smelting， thereby causing the main technical bottleneck in intelligent steelmaking． Therefore，a functional digital twin model of the steelmaking process based on flue-gas analysis was proposed． First，continuously monitored real-time data were obtained by flue gas analysis to ob￾serve the carbon and oxygen reaction state of molten steel in the converter． Then，according to various stages of the converter reaction， the functional data analysis method was used to establish the functional prediction models for the early and late stages of blowing． The