Articles & Issues

Language: English

Conflict of Interest: In relation to this article, we declare that there is no conflict of interest.

Publication history: Received April 18, 2019
Accepted October 12, 2019

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/bync/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

All issues

Anomaly detection in a hyper-compressor in low-density polyethylene manufacturing processes using WPCA-based principal component control limit

Byeong Eon Park Ji Seon Kim Jeong-Keun Lee In-Beum Lee^†

Department of Chemical Engineering, Pohang University of Science and Technology, 77 Cheongamro, Namgu, Pohang 37673, Korea

iblee@postech.ac.kr

Korean Journal of Chemical Engineering, January 2020, 37(1), 11-18(8), 10.1007/s11814-019-0403-y

Download PDF

Abstract

Low-Density Polyethylene (LDPE) was synthesized from ethylene at high-temperature and pressure condition. Hyper-compressor used to increase pressure up to 3,500 atm should be monitored and controlled delicately or it cannot guarantee stable operation of the process causing process shutdown (SD), which is directly related to product yield and process safety. This paper presents a data-based multivariate statistical monitoring method to detect anomalies in the hyper-compressor of a LDPE manufacturing process with weighted principal component analysis model (WPCA), which can consider both time-varying and time-invariant characteristic of data combining principal component analysis (PCA) and slow feature analysis (SFA). Operation data of the LDPE manufacturing process was gathered hourly for four years. WPCA-based principal component control limit (PCCL) was used as an index to determine anomaly and applied to five emergency shutdown (ESD) cases, respectively. As a result, all the five anomalies were detected by a PCCL, respectively, as a sign of SD. Moreover, it shows a better anomaly detection performance than the monitoring method using T2 and squared prediction error (SPE) based on PCA, SFA, or WPCA.

Keywords

Fault Detection Fault Isolation Slow Feature Analysis Principal Component Analysis

References

Gajendiran A, Krishnamoorthy S, Abraham J, 3 Biotech, 6, 1 (2016)
Sharmin R, Shah SL, Sundararaj U, Macromol. React. Eng., 2, 12 (2008)
Pichler K, Lughofer E, Pichler M, Buchegger T, Klement EP, Huschenbett M, Mech. Syst. Signal Process, 70-71, 104 (2016)
Farzaneh-Gord M, Khoshnazar H, J. Nat. Gas Sci. Eng., 35, 1239 (2016)
Sivalingam G, Soni NJ, Vakil SM, Can. J. Chem. Eng., 93(6), 1063 (2015)
Kumar V, Sundararaj U, Shah SL, Hair D, Vande Griend LJ, Polym. React. Eng., 11(4), 1017 (2003)
Ku W, Storer RH, Georgakis C, Chemom. Intell. Lab. Syst., 30, 179 (1995)
Scholkopf B, Smola A, Muller KR, Kernel principal component analysis, 583 (1997).
Choi SW, Lee C, Lee JM, Park JH, Lee IB, Chemom. Intell. Lab. Syst., 75, 55 (2005)
Jiang Q, Yan X, Korean J. Chem. Eng., 31(11), 1935 (2014)
Jiang Q, Yan X, Chinese J. Chem. Eng., 21, 633 (2013)
Han K, Park KJ, Chae H, Yoon ES, Korean J. Chem. Eng., 25(1), 13 (2008)
Jiang Q, Yan X, Chemom. Intell. Lab. Syst., 119, 11 (2012)
Zeng J, Liu K, Huang W, Liang J, Korean J. Chem. Eng., 34(8), 2135 (2017)
Gu S, Liu Y, Liu L, Zhang N, Du D, Open Mech. Eng. J., 9, 966 (2015)
Gajjar S, Palazoglu A, Chemom. Intell. Lab. Syst., 154, 122 (2016)
Wiskott L, Sejnowski TJ, Neural Comput., 14, 715 (2002)
Shang C, Yang F, Gao XQ, Huang XL, Suykens JAK, Huang DX, AIChE J., 61(11), 3666 (2015)
Wang HQ, Song ZH, Li P, Ind. Eng. Chem. Res., 41(10), 2455 (2002)
Tukey JW, Exploratory Data Analysis, Addison-Wesley (1977).