Browse > Article
http://dx.doi.org/10.7469/JKSQM.2017.45.1.025

Clustering-based Monitoring and Fault detection in Hot Strip Roughing Mill  

SEO, MYUNG-KYO (POSCO)
YUN, WON YOUNG (Department of Industrial Engineering, Pusan National University)
Publication Information
Journal of Korean Society for Quality Management / v.45, no.1, 2017 , pp. 25-38 More about this Journal
Abstract
Purpose: Hot strip rolling mill consists of a lot of mechanical and electrical units. In condition monitoring and diagnosis phase, various units could be failed with unknown reasons. In this study, we propose an effective method to detect early the units with abnormal status to minimize system downtime. Methods: The early warning problem with various units is defined. K-means and PAM algorithm with Euclidean and Manhattan distances were performed to detect the abnormal status. In addition, an performance of the proposed algorithm is investigated by field data analysis. Results: PAM with Manhattan distance(PAM_ManD) showed better results than K-means algorithm with Euclidean distance(K-means_ED). In addition, we could know from multivariate field data analysis that the system reliability of hot strip rolling mill can be increased by detecting early abnormal status. Conclusion: In this paper, clustering-based monitoring and fault detection algorithm using Manhattan distance is proposed. Experiments are performed to study the benefit of the PAM with Manhattan distance against the K-means with Euclidean distance.
Keywords
Hot strip rolling mill; Fault detection and classification; K-means; PAM;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Andrea Cerioli. 2005. "K-means Cluster Analysis and Mahalanobis Metrics: a problematic match or an overlooked opportunity." Statistica Applicata 17:61-73.
2 Asha Bharambe, Rahul Ravindra, Riya Suchdev, and Yash Tanna. 2014. "A Robust Anomaly Detection System." IEEE International Conference on Advances in Engineering & Technology Research 1-7.
3 Ghislain Verdier, and Ariane Ferreira. 2011. "Adaptive Mahalanobis Distance and k-Nearest Neighbor Rule for Fault Detection in Semiconductor Manufacturing." IEEE Transactions Semiconductor Manufacturing 24(1):59-68.   DOI
4 J. E. Kwak, and C. W. Kim. 2013. "Adaptive Clustering Based k-Nearest Neighbor Algorithm for Process Fault Detection." Proceedings of KORMS/KIIE Spring Joint Conference 1169-1175.
5 J. M. Pena, J. A. Lozano, and P. Larranaga. 1999. "An Empirical Comparison of Four Initialization Methods for the K-means Algorithm.", Pattern Recognition Letters 20(10):1-17.   DOI
6 Ji Hoon Kang, and Seoung Bum Kim. 2013. "A Clustering Algorithm-Based Control Chart for Inhomogeneously Distributed TFT-LCD Process." International Journal of Production Research 51(18):5644-5657.   DOI
7 Malika Charrad, Nadia Ghazzali, Veronique Boiteau, and Azam Niknafs. 2014. "NbClust: An R Package for Determining the Relevant Number of Clusters in a Dataset." Journal of Statistical Software 61(6):1-36.
8 R. Gnanadesikan, J. W. Harvey, and J. R. Kettenring. 1993. "Mahalanobis Metrics for Cluster Analysis." The Indian Journal of Statistics Series A(1961-2002) 55(3):494-505.
9 S. Bersimis, S. Psarakis, and J. Panaretos. 2007. "Multivariate Statistical Process Control Charts: an Overview." Quality and Reliability Engineering International 23(5):517-543.   DOI
10 Sachin Kumar, W. S. Chow, and Michael Pecht. 2010. "Approach to Fault Identification for Electronic Products Using Mahalanobis Distance." IEEE Transactions on Instrumentation and Measurement 59(8):2055-2064.   DOI
11 T. H. Lee, and C. W. Kim. 2013. "Statistical Comparison of Data Mining Models for Fault Diagnosis in an Etching process." Proceedings of KORMS/KIIE Spring Joint Conference 1887-1895.