摘要
钢铁工业智能故障诊断系统在当前大数据时代背景下面临着新的机遇与挑战;针对工业大数据的特征,分别从数据的采集与实时监控技术,基于机器学习的故障诊断方法,以及迁移学习在工业故障诊断中的应用三个角度对近年来国内外工业故障诊断方法的研究进展进行了总结与回顾;并在此基础上,结合钢铁企业的实际需求与现存问题,提出了将高炉炼铁过程划分为“系统—模块—功能—属性”四层次结构的面向整体的分层故障诊断新思想及未来可能的研究方向,阐明研究多技术融合的智能故障诊断系统对推进钢铁工业在大数据时代的绿色数字化发展具有十分重要的意义。
Big data has brought new opportunities and challenges to intelligent fault diagnosis in iron and steel industry.According to the characteristics of big data,this paper completes an overview on recent industrial fault diagnosis approaches from the perspectives of data collection and real-time monitoring,machine learning-based diagnosis methods,and the applications of transfer learning.Analysing actual needs and current issues of iron and steel enterprises,an overlook on the future development of fault diagnosis for blast furnace ironmaking based on a novel holistic“System-Module-Function-Attribute”hierarchical structure is presented,with the aim of promoting the green and digital transformation in manufacturing industries.
作者
傅筱
韩俊毅
曹阔
Fu Xiao;Han Junyi;Cao Kuo(School of Information Management,Shanghai Lixin University of Accounting and Finance,Shanghai 201209,China;Golden Data Limited,Chongqing 401147,China)
出处
《计算机测量与控制》
2020年第11期1-6,26,共7页
Computer Measurement &Control
关键词
工业大数据
故障诊断
机器学习
迁移学习
钢铁智能制造
industrial big data
fault diagnosis
machine learning
transfer learning
intelligent iron and steel manufacturing
