Anomaly detection using multi-scale dynamic principal component analysis for Tenneesse Eastman Process

K. Ramakrishna Kini, Muddu Madakyaru

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Effective and accurate detection of anomalies is crucial to ensure safe and reliable operation in complex chemical processes. Principal Component Analysis (PCA) technique has been a popular choice for anomaly detection vowing to its capability in handling complex multiple variables. However, data extracted from chemical process plants is correlated in time due to the process dynamics, hereby making the standard PCA approach ineffective in detecting of anomalies. Also, with most industrial process data embedded with heavy noise content, the detection ability of the standard PCA strategy is greatly effected. The dynamics present in the industrial data can be incorporated using dynamic PCA modeling technique where existing PCA model is embedded with lagged variales which would aid in capturing plant dynamics. Further, the effect of measurement noise can be reduced by using wavelet based multi-scale representation of data which extracts useful information from the data in time and frequency domain simultaneously. In the current paper, a multi-scale dynamic PCA (MSDPCA)is used as the modeling frame-work while T2 and Q statistics are used as fault detection indicators. The efficacy of the developed MSDPCA strategy is demonstrated on a bench-mark Tenneesse Eastman(TE) industrial process. The simulation results clearly shows that the MSDPCA strategy is able to detect anomalies effectively in the multi-variate dynamic TE process.

Original languageEnglish
Title of host publication2019 5th Indian Control Conference, ICC 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages219-224
Number of pages6
ISBN (Electronic)9781538662465
DOIs
Publication statusPublished - 14-05-2019
Event5th Indian Control Conference, ICC 2019 - Delhi, India
Duration: 09-01-201911-01-2019

Publication series

Name2019 5th Indian Control Conference, ICC 2019 - Proceedings

Conference

Conference5th Indian Control Conference, ICC 2019
CountryIndia
CityDelhi
Period09-01-1911-01-19

Fingerprint

Anomaly Detection
Dynamic Analysis
Principal component analysis
Principal Component Analysis
Anomaly
Chemical Processes
Representation of data
Dynamic Process
Fault Detection
Modeling
Frequency Domain
Efficacy
Time Domain
Wavelets
Fault detection
Benchmark
Statistics
Strategy
Simulation

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Control and Optimization

Cite this

Kini, K. R., & Madakyaru, M. (2019). Anomaly detection using multi-scale dynamic principal component analysis for Tenneesse Eastman Process. In 2019 5th Indian Control Conference, ICC 2019 - Proceedings (pp. 219-224). [8715552] (2019 5th Indian Control Conference, ICC 2019 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/INDIANCC.2019.8715552
Kini, K. Ramakrishna ; Madakyaru, Muddu. / Anomaly detection using multi-scale dynamic principal component analysis for Tenneesse Eastman Process. 2019 5th Indian Control Conference, ICC 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 219-224 (2019 5th Indian Control Conference, ICC 2019 - Proceedings).
@inproceedings{25a171f7a7cb45b0ba1f790855dd77b8,
title = "Anomaly detection using multi-scale dynamic principal component analysis for Tenneesse Eastman Process",
abstract = "Effective and accurate detection of anomalies is crucial to ensure safe and reliable operation in complex chemical processes. Principal Component Analysis (PCA) technique has been a popular choice for anomaly detection vowing to its capability in handling complex multiple variables. However, data extracted from chemical process plants is correlated in time due to the process dynamics, hereby making the standard PCA approach ineffective in detecting of anomalies. Also, with most industrial process data embedded with heavy noise content, the detection ability of the standard PCA strategy is greatly effected. The dynamics present in the industrial data can be incorporated using dynamic PCA modeling technique where existing PCA model is embedded with lagged variales which would aid in capturing plant dynamics. Further, the effect of measurement noise can be reduced by using wavelet based multi-scale representation of data which extracts useful information from the data in time and frequency domain simultaneously. In the current paper, a multi-scale dynamic PCA (MSDPCA)is used as the modeling frame-work while T2 and Q statistics are used as fault detection indicators. The efficacy of the developed MSDPCA strategy is demonstrated on a bench-mark Tenneesse Eastman(TE) industrial process. The simulation results clearly shows that the MSDPCA strategy is able to detect anomalies effectively in the multi-variate dynamic TE process.",
author = "Kini, {K. Ramakrishna} and Muddu Madakyaru",
year = "2019",
month = "5",
day = "14",
doi = "10.1109/INDIANCC.2019.8715552",
language = "English",
series = "2019 5th Indian Control Conference, ICC 2019 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "219--224",
booktitle = "2019 5th Indian Control Conference, ICC 2019 - Proceedings",
address = "United States",

}

Kini, KR & Madakyaru, M 2019, Anomaly detection using multi-scale dynamic principal component analysis for Tenneesse Eastman Process. in 2019 5th Indian Control Conference, ICC 2019 - Proceedings., 8715552, 2019 5th Indian Control Conference, ICC 2019 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 219-224, 5th Indian Control Conference, ICC 2019, Delhi, India, 09-01-19. https://doi.org/10.1109/INDIANCC.2019.8715552

Anomaly detection using multi-scale dynamic principal component analysis for Tenneesse Eastman Process. / Kini, K. Ramakrishna; Madakyaru, Muddu.

2019 5th Indian Control Conference, ICC 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 219-224 8715552 (2019 5th Indian Control Conference, ICC 2019 - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Anomaly detection using multi-scale dynamic principal component analysis for Tenneesse Eastman Process

AU - Kini, K. Ramakrishna

AU - Madakyaru, Muddu

PY - 2019/5/14

Y1 - 2019/5/14

N2 - Effective and accurate detection of anomalies is crucial to ensure safe and reliable operation in complex chemical processes. Principal Component Analysis (PCA) technique has been a popular choice for anomaly detection vowing to its capability in handling complex multiple variables. However, data extracted from chemical process plants is correlated in time due to the process dynamics, hereby making the standard PCA approach ineffective in detecting of anomalies. Also, with most industrial process data embedded with heavy noise content, the detection ability of the standard PCA strategy is greatly effected. The dynamics present in the industrial data can be incorporated using dynamic PCA modeling technique where existing PCA model is embedded with lagged variales which would aid in capturing plant dynamics. Further, the effect of measurement noise can be reduced by using wavelet based multi-scale representation of data which extracts useful information from the data in time and frequency domain simultaneously. In the current paper, a multi-scale dynamic PCA (MSDPCA)is used as the modeling frame-work while T2 and Q statistics are used as fault detection indicators. The efficacy of the developed MSDPCA strategy is demonstrated on a bench-mark Tenneesse Eastman(TE) industrial process. The simulation results clearly shows that the MSDPCA strategy is able to detect anomalies effectively in the multi-variate dynamic TE process.

AB - Effective and accurate detection of anomalies is crucial to ensure safe and reliable operation in complex chemical processes. Principal Component Analysis (PCA) technique has been a popular choice for anomaly detection vowing to its capability in handling complex multiple variables. However, data extracted from chemical process plants is correlated in time due to the process dynamics, hereby making the standard PCA approach ineffective in detecting of anomalies. Also, with most industrial process data embedded with heavy noise content, the detection ability of the standard PCA strategy is greatly effected. The dynamics present in the industrial data can be incorporated using dynamic PCA modeling technique where existing PCA model is embedded with lagged variales which would aid in capturing plant dynamics. Further, the effect of measurement noise can be reduced by using wavelet based multi-scale representation of data which extracts useful information from the data in time and frequency domain simultaneously. In the current paper, a multi-scale dynamic PCA (MSDPCA)is used as the modeling frame-work while T2 and Q statistics are used as fault detection indicators. The efficacy of the developed MSDPCA strategy is demonstrated on a bench-mark Tenneesse Eastman(TE) industrial process. The simulation results clearly shows that the MSDPCA strategy is able to detect anomalies effectively in the multi-variate dynamic TE process.

UR - http://www.scopus.com/inward/record.url?scp=85066636871&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066636871&partnerID=8YFLogxK

U2 - 10.1109/INDIANCC.2019.8715552

DO - 10.1109/INDIANCC.2019.8715552

M3 - Conference contribution

AN - SCOPUS:85066636871

T3 - 2019 5th Indian Control Conference, ICC 2019 - Proceedings

SP - 219

EP - 224

BT - 2019 5th Indian Control Conference, ICC 2019 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Kini KR, Madakyaru M. Anomaly detection using multi-scale dynamic principal component analysis for Tenneesse Eastman Process. In 2019 5th Indian Control Conference, ICC 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 219-224. 8715552. (2019 5th Indian Control Conference, ICC 2019 - Proceedings). https://doi.org/10.1109/INDIANCC.2019.8715552