Research and application of width blockade tracing for hot-rolled strip
XIE Meng1, LI Wei-gang1, WANG You-long1, ZHU Hong-lin2
1. School of Information Science and Engineering, Wuhan University of Science and Technology, Wuhan 430081, Hubei, China; 2. Research Institute, Baoshan Iron and Steel Co., Ltd., Shanghai 201900, China
Abstract:Tracing of width blockade for hot-rolled strip refers to judging the defect type of width blockade strip after the strip rolling is completed,and tracing the causes of width blockade,so as to provide a reference for on-site technicians. Based on the real hot strip production process data, firstly, Kalman filter method was used to denoise the full-length width data of the strip,then Gauss fitting and peak searching methods were used to extract the width characteristics of the strip and width defect classification rules were established. Finally, combined with the actual production process, width blockade traceability model of hot strip was established. The 582 strips with width blockade of Zhanjiang 1 780 mm hot strip mill in three months were used for the model performance test. The results show that the classification accuracy of the model for width defects is 96.72%,and the traceability accuracy for width blockade is 94.16%, which effectively realizes the automatic classification and automatic traceability for width blockade of hot-rolled strip.
谢孟, 李维刚, 王优龙, 朱红林. 热连轧带钢宽度封锁溯源研究及应用[J]. 中国冶金, 2023, 33(2): 121-128.
XIE Meng, LI Wei-gang, WANG You-long, ZHU Hong-lin. Research and application of width blockade tracing for hot-rolled strip[J]. China Metallurgy, 2023, 33(2): 121-128.
Ray B K,Tsay R S. Bayesian methods for change-point detection in long-range dependent processes[J]. Journal of Time Series Analysis,2010,23(6):687.
[15]
Angelov P. Anomaly detection based on eccentricity analysis[C]//2014 IEEE Symposium on Evolving and Autonomous Learning Systems (EALS). Orlando,FL,USA:IEEE,2015:1.
[16]
Chauhan S,Vig L. Anomaly detection in ECG time signals via deep long short-term memory networks[C]//IEEE International Conference on Data Science and Advanced Analytics. Paris,France:IEEE,2015:1.
[17]
Siami-Namini S,Tavakoli N,Namin A S. A comparison of ARIMA and LSTM in forecasting time series[C]//2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA). Orlando,FL,USA:IEEE,2018:1394.