Fault diagnosis of strip breaking in hot strip rolling based on kernel principal component analysis
WU Kai1, SUN Yan-guang2, ZHANG Lin3
1. Steel Rolling and Driving Dept., Beijing Aritime Intelligent Control Co.,Ltd., Beijing 100070, China; 2. Automation Research and Design Institute of Metallurgical Industry, Beijing 100071, China; 3. State Key Laboratory of Hybrid Process Industry Automation Systems and Equipment Technology, Automation Research and Design Institute of Metallurgical Industry, Beijing 100071, China
Abstract:Equipment and quality faults often occur during hot strip rolling process. In order to quickly determine the fault cause and eliminate faults, it is necessary to monitor the production process and diagnose the faults. Based on the data collected in the hot rolling process, the kernel principal component analysis is used to monitor the relevant data of the finishing mill and diagnose the broken strip fault. Using SPE statistics to monitor the production process, based on the kernel principal component analysis, the contribution plot of each variable is drawn, and the main influencing variables causing the fault are found out according to the contribution rate. Compared with principal component analysis, kernel principal component analysis is more efficient and accurate. The fault diagnosis of hot strip breaking based on kernel principal component analysis can save the fault analysis time and provide the basis for the adjustment and troubleshooting of hot rolling production process, which has important theoretical significance and practical application value.
武凯, 孙彦广, 张琳. 基于核主成分分析的热连轧断带故障诊断[J]. 中国冶金, 2020, 30(11): 60-65.
WU Kai, SUN Yan-guang, ZHANG Lin. Fault diagnosis of strip breaking in hot strip rolling based on kernel principal component analysis[J]. China Metallurgy, 2020, 30(11): 60-65.
Imen Hamrouni,Hajer Lahdhiri,Khaoula ben Abdellafou,et al. Fault detection of uncertain nonlinear process using reduced interval kernel principal component analysis (RIKPCA) [J]. The International Journal of Advanced Manufacturing Technology,2020,106 (9/10):4567.
[9]
Jia F,Martin E B,Morris A J. Nonlinear principal components analysis with application to process fault detection[J]. International Journal of Systems Science,2000,31(11):1473.