Development of Machine Learning based Model for Anomaly Detection and Fault Cause Diagnosis of Assets in Petrochemical Industries

Hwawon Hwang; Yojin Kim; Seunghye Lee; Heejeong Choi; Pilsung Kang; Yongha In; Wonwoo Ro; Namwook Kang

doi:10.1109/CITDS54976.2022.9914377

Development of Machine Learning based Model for Anomaly Detection and Fault Cause Diagnosis of Assets in Petrochemical Industries

Hwawon Hwang, Yojin Kim, Seunghye Lee, Heejeong Choi, Pilsung Kang, Yongha In, Wonwoo Ro, Namwook Kang

School of Industrial Management Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Petrochemical companies put much effort into maximizing productivity and optimizing TCO(Total Cost of Operation) by reducing the unplanned downtime for stable operation of assets since unplanned downtime of assets leads to colossal production loss and environmental safety accidents. The PdM (Predictive Maintenance) solution is required to predict prognostic abnormal behavior of assets before the time when asset fault occurs, give warning alarm to engineers, and help them take proactive measures by diagnosing the fault cause and guiding suitable measures.In this research, the PdM model has been developed using Variational AutoEncoder and Isolation Forest algorithms to detect the prognostic abnormal behavior of assets before the unplanned shutdown. Moreover, PdM model for diagnosing the possible causes of abnormal behavior of the centrifugal compressor has also been developed to help domain field engineers take the suitable measures before the unplanned shutdown of the asset. By applying the PdM model to actual data of centrifugal compressor in petrochemical process, the PdM model has been successfully validated and shown feasible results.

Original language	English
Title of host publication	2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	118-123
Number of pages	6
ISBN (Electronic)	9781665496537
DOIs	https://doi.org/10.1109/CITDS54976.2022.9914377
Publication status	Published - 2022
Event	2nd IEEE Conference on Information Technology and Data Science, CITDS 2022 - Virtual, Debrecen, Hungary Duration: 2022 May 16 → 2022 May 18

Publication series

Name	2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings

Conference

Conference	2nd IEEE Conference on Information Technology and Data Science, CITDS 2022
Country/Territory	Hungary
City	Virtual, Debrecen
Period	22/5/16 → 22/5/18

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Keywords

Anomaly detection
Fault cause diagnosis
Isolation Forest
Predictive maintenance
Random forest
Variational AutoEncoder

ASJC Scopus subject areas

Artificial Intelligence
Information Systems
Safety, Risk, Reliability and Quality
Water Science and Technology
Control and Optimization

Access to Document

10.1109/CITDS54976.2022.9914377

Cite this

Hwang, H., Kim, Y., Lee, S., Choi, H., Kang, P., In, Y., Ro, W., & Kang, N. (2022). Development of Machine Learning based Model for Anomaly Detection and Fault Cause Diagnosis of Assets in Petrochemical Industries. In 2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings (pp. 118-123). (2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CITDS54976.2022.9914377

Development of Machine Learning based Model for Anomaly Detection and Fault Cause Diagnosis of Assets in Petrochemical Industries. / Hwang, Hwawon; Kim, Yojin; Lee, Seunghye et al.
2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. p. 118-123 (2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Hwang, H, Kim, Y, Lee, S, Choi, H, Kang, P, In, Y, Ro, W & Kang, N 2022, Development of Machine Learning based Model for Anomaly Detection and Fault Cause Diagnosis of Assets in Petrochemical Industries. in 2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings. 2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 118-123, 2nd IEEE Conference on Information Technology and Data Science, CITDS 2022, Virtual, Debrecen, Hungary, 22/5/16. https://doi.org/10.1109/CITDS54976.2022.9914377

Hwang H, Kim Y, Lee S, Choi H, Kang P, In Y et al. Development of Machine Learning based Model for Anomaly Detection and Fault Cause Diagnosis of Assets in Petrochemical Industries. In 2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2022. p. 118-123. (2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings). doi: 10.1109/CITDS54976.2022.9914377

Hwang, Hwawon ; Kim, Yojin ; Lee, Seunghye et al. / Development of Machine Learning based Model for Anomaly Detection and Fault Cause Diagnosis of Assets in Petrochemical Industries. 2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 118-123 (2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings).

@inproceedings{1f8d92639c6945e8847dbf986928d87b,

title = "Development of Machine Learning based Model for Anomaly Detection and Fault Cause Diagnosis of Assets in Petrochemical Industries",

abstract = "Petrochemical companies put much effort into maximizing productivity and optimizing TCO(Total Cost of Operation) by reducing the unplanned downtime for stable operation of assets since unplanned downtime of assets leads to colossal production loss and environmental safety accidents. The PdM (Predictive Maintenance) solution is required to predict prognostic abnormal behavior of assets before the time when asset fault occurs, give warning alarm to engineers, and help them take proactive measures by diagnosing the fault cause and guiding suitable measures.In this research, the PdM model has been developed using Variational AutoEncoder and Isolation Forest algorithms to detect the prognostic abnormal behavior of assets before the unplanned shutdown. Moreover, PdM model for diagnosing the possible causes of abnormal behavior of the centrifugal compressor has also been developed to help domain field engineers take the suitable measures before the unplanned shutdown of the asset. By applying the PdM model to actual data of centrifugal compressor in petrochemical process, the PdM model has been successfully validated and shown feasible results.",

keywords = "Anomaly detection, Fault cause diagnosis, Isolation Forest, Predictive maintenance, Random forest, Variational AutoEncoder",

author = "Hwawon Hwang and Yojin Kim and Seunghye Lee and Heejeong Choi and Pilsung Kang and Yongha In and Wonwoo Ro and Namwook Kang",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2nd IEEE Conference on Information Technology and Data Science, CITDS 2022 ; Conference date: 16-05-2022 Through 18-05-2022",

year = "2022",

doi = "10.1109/CITDS54976.2022.9914377",

language = "English",

series = "2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "118--123",

booktitle = "2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings",

}

TY - GEN

T1 - Development of Machine Learning based Model for Anomaly Detection and Fault Cause Diagnosis of Assets in Petrochemical Industries

AU - Hwang, Hwawon

AU - Kim, Yojin

AU - Lee, Seunghye

AU - Choi, Heejeong

AU - Kang, Pilsung

AU - In, Yongha

AU - Ro, Wonwoo

AU - Kang, Namwook

PY - 2022

Y1 - 2022

N2 - Petrochemical companies put much effort into maximizing productivity and optimizing TCO(Total Cost of Operation) by reducing the unplanned downtime for stable operation of assets since unplanned downtime of assets leads to colossal production loss and environmental safety accidents. The PdM (Predictive Maintenance) solution is required to predict prognostic abnormal behavior of assets before the time when asset fault occurs, give warning alarm to engineers, and help them take proactive measures by diagnosing the fault cause and guiding suitable measures.In this research, the PdM model has been developed using Variational AutoEncoder and Isolation Forest algorithms to detect the prognostic abnormal behavior of assets before the unplanned shutdown. Moreover, PdM model for diagnosing the possible causes of abnormal behavior of the centrifugal compressor has also been developed to help domain field engineers take the suitable measures before the unplanned shutdown of the asset. By applying the PdM model to actual data of centrifugal compressor in petrochemical process, the PdM model has been successfully validated and shown feasible results.

AB - Petrochemical companies put much effort into maximizing productivity and optimizing TCO(Total Cost of Operation) by reducing the unplanned downtime for stable operation of assets since unplanned downtime of assets leads to colossal production loss and environmental safety accidents. The PdM (Predictive Maintenance) solution is required to predict prognostic abnormal behavior of assets before the time when asset fault occurs, give warning alarm to engineers, and help them take proactive measures by diagnosing the fault cause and guiding suitable measures.In this research, the PdM model has been developed using Variational AutoEncoder and Isolation Forest algorithms to detect the prognostic abnormal behavior of assets before the unplanned shutdown. Moreover, PdM model for diagnosing the possible causes of abnormal behavior of the centrifugal compressor has also been developed to help domain field engineers take the suitable measures before the unplanned shutdown of the asset. By applying the PdM model to actual data of centrifugal compressor in petrochemical process, the PdM model has been successfully validated and shown feasible results.

KW - Anomaly detection

KW - Fault cause diagnosis

KW - Isolation Forest

KW - Predictive maintenance

KW - Random forest

KW - Variational AutoEncoder

UR - http://www.scopus.com/inward/record.url?scp=85140974860&partnerID=8YFLogxK

U2 - 10.1109/CITDS54976.2022.9914377

DO - 10.1109/CITDS54976.2022.9914377

M3 - Conference contribution

AN - SCOPUS:85140974860

T3 - 2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings

SP - 118

EP - 123

BT - 2022 IEEE 2nd Conference on Information Technology and Data Science, CITDS 2022 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2nd IEEE Conference on Information Technology and Data Science, CITDS 2022

Y2 - 16 May 2022 through 18 May 2022

ER -

Development of Machine Learning based Model for Anomaly Detection and Fault Cause Diagnosis of Assets in Petrochemical Industries

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this