Taking the continuous rolling process of seamless steel pipes as research object, a monitoring model for the continuous rolling process of seamless steel pipes based on the kernel entropy component analysis method is established, and abnormal production situations and quality problems are diagnosed based on the kernel space fault contribution rate algorithm. Through process monitoring and fault diagnosis experiments on simulation data and seamless steel pipe onsite production data, the experimental results demonstrated the effectiveness of the proposed method in the field. Compared with common process monitoring methods, the actual production data results show that the kernel entropy component analysis–dissimilarity method has better monitoring performance, with an FDR of 97.0% and a FAR of 1.0%. Finally, a process monitoring and fault diagnosis system for the production site is developed for producing abnormal alarms.
BilanITRomantsevBABelonozhkoSS, et al.Modernization of cooling system for mandrel of PQF continuous rolling mill at JSC tagmet. Metallurgist2022; 66: 863–867.
2.
ChenCChenJXShuangYH, et al.Short-flow rolling process and heat treatment of seamless titanium alloy tube. Metals (Basel)2023; 13: 527.
3.
LiuHLiQGuiHL, et al.Technology optimization analysis of three-roll rotary piercing process for seamless steel pipe. JOM2024; 76: 3465–3475.
4.
LiJDWangXCYangQ, et al.Rolling force prediction in cold rolling process based on combined method of T-S fuzzy neural network and analytical model. Int J Adv Manuf Technol2022; 121: 4087–4098.
5.
SongKCFengHCaoTL, et al.MFANet: multifeature aggregation network for cross-granularity few-shot seamless steel tubes surface defect segmentation. IEEE Trans Ind Inform2024; 20: 9725–9735.
6.
ChenXZhuMJWuM, et al.Rolling force modeling for seamless steel pipe combining mechanism model and neural network prediction. Metall Ind Autom2015; 39: 32–37.
7.
YangDMCuiYRYuZY, et al.Deep learning based steel pipe weld defect detection. Appl Artif Intell2021; 35: 1237–1249.
8.
LangbauerRNunnerGZmekT, et al.Development of an artificial neural network (ANN) model to predict the temperature of hot-rolled steel pipes. Adv Ind Manuf Eng2022; 5: 100090.
9.
Ojeda-LópezABotana-GalvínMGonzález-RoviraL, et al.Numerical simulation as a tool for the study, development, and optimization of rolling processes: a review. Metals (Basel)2024; 14: 737.
10.
WangLLSongHCWanGX, et al.A surface defect detection method for steel pipe based on improved YOLO. Math Biosci Eng2024; 21: 3016–3036.
11.
OrlovGAAshkanovSEOrlovAG. Analysis of the end wall thickness variation of tubes after reducing mill. Ferrous Metall2017; 60: 250–251.
12.
LiJDWangXCYangQ, et al.Rolling force prediction in cold rolling process based on combined method of T-S fuzzy neural network and analytical model. Int J Adv Manuf Technol2022; 121: 4087–4098.
13.
JiangYZHuYFFuZG, et al.Defects analysis of the ultra-thick steel pipes using a mixed explicit–implicit FEM. Metall Res Technol2022; 119: 111.
14.
WangTFengKLingJT, et al.Pipeline condition monitoring towards digital twin system: a case study. J Manuf Syst2024; 73: 256–274.
15.
XuDCHouHSLiuCX, et al.Defect type identification of thin-walled stainless steel seamless pipe based on eddy current testing. Insight Non-Destr Test Cond Monit2021; 63: 697–703.
16.
TianGYSophianATaylorD, et al.Wavelet-based PCA defect classification and quantification for pulsed eddy current NDT. IEE Proc Sci Meas Technol2005; 152: 141–148.
17.
ZhangYCJiJCRenZH, et al.Digital twin-driven partial domain adaptation network for intelligent fault diagnosis of rolling bearing. Reliab Eng Syst Saf2023; 234: 109186.
18.
IsaDRajkumarR. Pipeline defect prediction using support vector machines. Appl Artif Intell2009; 23: 758–771.
19.
PanasenkoOAShkuratovEABelovOI, et al.Improvement of the axial adjustment of the equipment of fine quality mill on the basis of contactless measuring 3D systems. Metallurgist2019; 63: 690–694.
20.
LakizaVARomancevBAAleshchenkoAS, et al.Study of mandrel wear during billet piercing on the MISiS-130D screw rolling mill. Metallurgist2024; 67: 1745–1751.
21.
WeiZHWuCJ. New analytical model for determining the roll pitch diameter in three-roll continuous retained mandrel rolling. Metals (Basel)2023; 13: 304.
22.
ZhangHQiYWangL, et al.Fault detection and diagnosis of chemical process using enhanced KECA. Chemometr Intell Lab Syst2017; 161: 61–69.
23.
QiYSMengXBLuCX, et al.KECA similarity-based monitoring and diagnosis of faults in multi-phase batch processes. Entropy2019; 21: 121.
24.
HeFWangCJFanSK. Fault detection and root cause analysis of a batch process via novel nonlinear dissimilarity and comparative granger causality analysis. Ind Eng Chem Res2019; 58: 21842–21854.
25.
LiTHanYMXuWX, et al.Novel adaptive fault detection method based on kernel entropy component analysis integrating moving window of dissimilarity for nonlinear dynamic processes. J Process Control2023; 125: 1–18.
26.
LiJDZhaoJWWangXC, et al.An industrial IoT-based deformation resistance prediction and thickness control method of cold-rolled strip in steel production systems. Inf Sci2024; 674: 120735.
27.
LiJDWangXCZhaoJW, et al.Predicting mechanical properties lower upper bound for cold-rolling strip by machine learning-based artificial intelligence. ISA Trans2024; 147: 328–336.
28.
XiaYDDingQJingNJ, et al.An enhanced fault detection method for centrifugal chillers using kernel density estimation-based kernel entropy component analysis. Int J Refrig2021; 129: 290–300.
29.
HeFDuXFWangCJ. Root cause diagnosis of head narrowing in hot rolled strip based on comparative granger causality. China Mech Eng2020; 31: 2340–2346.
30.
PilarioKECaoY. Canonical variate dissimilarity analysis for process incipient fault detection. IEEE Trans Ind Inform2018; 14: 5308–5315.
