数据驱动的热轧过程故障诊断研究现状与展望  

Status and prospects in data-driven fault diagnosis for hot rolling process

作　　者：马亮[1,2] 施富中[1,2] 彭开香 MA Liang;SHI Fuzhong;PENG Kaixiang(Key Laboratory of Knowledge Automation for Industrial Processes of Ministry of Education,Beijing 100083,China;School of Automation and Electrical Engineering,University of Science and Technology Beijing,Beijing 100083,China)

机构地区：[1]工业过程知识自动化教育部重点实验室,北京100083 [2]北京科技大学自动化学院,北京100083

出　　处：《中国冶金》2023年第12期12-17,共6页China Metallurgy

基　　金：国家自然科学基金资助项目(62003030,62373041,U21A20483,72171171)。

摘　　要：热轧过程是钢铁制造全流程承上启下的重要环节,其安全生产与高效运行对钢铁工业的高质量发展具有重要意义。由于热轧过程工况复杂多变、层级分工明确、回路互联耦合,多个工序、设备、子系统或控制回路可能同时发生或先后级联出现故障,使得热轧过程故障诊断理论与方法成为过程控制领域的研究热点。从分析热轧过程特点与主要故障类型出发,重点总结了数据驱动的热轧过程故障诊断方法的研究现状,定性分析了主流方法的优势和劣势,梳理了热轧过程故障诊断中存在的问题,并进一步展望了热轧过程故障诊断方法未来的发展方向。Hot rolling process is an important cohering link in the whole production process of iron and steel,whose safe production and efficient operation are of great significance to the high-quality developments of iron and steel industry.Due to the complex and variable working conditions,clear system hierarchy of hot rolling process as well as the interconnection and coupling of loops,faults of multiple processes,equipment,subsystems or control loops may occur at the same time or cascade successively,which make the theories and methods of fault diagnosis for hot rolling process become research hotspots in the field of process control.Based on the analysis of characteristics and main fault types of hot rolling process,the research status of data-driven fault diagnosis methods for hot rolling process is emphatically summarized,the advantages and disadvantages of mainstream methods are qualitatively analyzed.Moreover,the existing problems of fault diagnosis for hot rolling process are sorted out,and the development directions in the future are further prospected.

关 键 词：热轧过程 故障诊断 数据驱动 智能诊断 机器学习 深度学习 

分 类 号：TG335.11[金属学及工艺—金属压力加工]