钢铁冶金数字化高炉研究

Research on Digitization of Blast Furnace in Ferrous Metallurgy

高炉数字孪生系统是智能制造应用布局在炼铁领域的重要推进,是“5G+工业互联网”战略的深度探索。通过结合推理及大数据机器学习建立整套高炉数字孪生系统,其中包括实时诊断预判、高炉状况综合评价体系,实时有效诊断预警失常炉况,并可自动生成炉况分析报告,在促进炼铁技术整合、加强工序协同、提高生产效率、提升本质安全等方面发挥积极作用。系统采用“云—边—端”一体化协同模式,以工业传感器、物联网技术为基础,建立“数字化监测—数据分析—炉况诊断—工艺优化”炼铁管控体系,开辟“专家经验+机理模型+推理机+机器学习”工艺难题解析新路径,形成高炉过程数字驱动仿真及数字化三维模型,建立起高炉“数字孪生”系统;将总体生产状况和关键设备运行状态实时展示、数字化、具象化,实时监测重点设备,及时掌握运行状况,实现快速检测、快速诊断、精准维修和远程在线巡检;统计、分析、学习数据,用于生产工艺及设备工况优化、预测性维护,有效减少设备故障频次、缩短故障时间。

Digital twin system of blast furnace is a great progress of intelligent manufacturing applying in the ferrous metallurgy,and a deep exploration in 5G+Industrial Internet.Combining inference and big data machine learning,digital twin system of blast furnace is established,including real-time diagnostic prediction and comprehensive assessment of blast furnace condition,which could provide early warning and effectively diagnose the abnormal furnace practice in time,automatically generate analysis report of furnace practice.This system plays a positive role in technology integration of ferrous metallurgy,enhancement of progress synergy,enhancement of production efficiency and promotion of essential safety.Based on industrial sensor and Internet of Things,"Cloud-border-end"integrated collaborative model is adopted,"digital monitoring+data analysis+furnace practice diagnose+process optimization"is built,a new path"expertise+mechanism model+inference engine+machine learning"for solving process problems is open up.The digital driving simulation and digital three-dimensional model of blast furnace process are formed,and the"digital twin"system of blast furnace is established.In this system,general production condition and operating status of key equipment could be demonstrated,digitalized and visualized in time,with important equipment monitored real timely,running state gained in time,rapid detection,rapid diagnose,accurate maintenance and remote online inspection realized.By applying statistic,analysis and learning of data in process/operation condition promotion and predictive maintenance,equipment failure frequency and time could be cut down effectively.