Register      Login
Australian Energy Producers Journal Australian Energy Producers Journal Society
Journal of Australian Energy Producers
RESEARCH ARTICLE (Non peer reviewed)

Using unsupervised machine learning to identify risk of failure at ageing oil and gas assets

Uday Manchanda A * and Ammar Pervez B
+ Author Affiliations
- Author Affiliations

A Kent, Perth, WA, Australia.

B Deloitte, Perth, WA, Australia.

* Correspondence to: uday.manchanda@kentplc.com

The APPEA Journal 62 S149-S152 https://doi.org/10.1071/AJ21173
Accepted: 15 March 2022   Published: 13 May 2022

© 2022 The Author(s) (or their employer(s)). Published by CSIRO Publishing on behalf of APPEA.

Abstract

Asset integrity management of ageing oil and gas assets is an ongoing challenge. This paper uses unsupervised algorithms (i.e. clustering technique) to identify carbon steel piping with increased probability of failure due to various internal corrosion mechanisms. The application used over 20 variables including wellhead planktonic bacterial counts, Fe2+ levels, oil and water production rates, historical Non-Destructive Testing (NDT) records, remaining life of downstream equipment, previous remediation data and geographical location data. An unsupervised machine learning clustering algorithm was written grounded in mathematical techniques of Principal Component Analysis (PCA) and k-means clustering. The probabilistic algorithm identified implicit patterns, which were then used to identify critical and non-critical piping clusters. Outputs from the clustering model were used to prioritise field measurements, and while these are ongoing there appears to be a good agreement with model predictions. The paper further discusses the measures that have a higher impact on the classification accuracy of the algorithm.

Keywords: asset integrity, carbon steel piping, clustering, corrosion, internal corrosion mechanism, k-means, machine learning, oil and gas, principal component analysis, probability of failure, unsupervised machine learning.

Uday Manchanda is a technical safety engineering professional who holds a Bachelor of Science, Master of Professional Engineering and a Master of Commerce degrees from The University of Western Australia. His experience includes risk assessments, asset integrity, technical safety, flow-induced and acoustic vibrations, and continuous improvement. He was awarded a technical design prize by Yara for designing the best demineralisation system at an ammonia plant and the 2020 Perron Prize. He has a keen interest in advanced probabilistic risk assessments techniques.

Ammar Pervez is an Artificial Intelligence consultant. He holds a Masters degree in Data Science and a Bachelors in Mechanical Engineering, has extensive experience in development and implementation of asset integrity management processes around oil and gas hubs. Recently, he has been involved in promoting the application of more advanced statistical techniques within the process industry and an advocate for explainable AI. His interests include predictive analytics, time-series predictions, natural-language processing, asset integrity management and other facets of artificial intelligence.


References

Ayustyana E, Wibisonso SA, Sihombing FMH (2021) Coal characterization of South Sumatera Basin using the unsupervised machine learning method. In ‘IOP Conference Series: Earth and Environmental Science, Indonesia, 23–25 October 2020’. (IOP Publishing).

Baby NV, Paricha B, Naik SJ (2016) Determination of Corrosion rates and remaining life of piping using API and ASME standards in oil and gas industries. International Research Journal of Engineering and Technology 3, 772–777.

El Naqa I, Murphy MJ (2015) What is machine learning? In ‘Machine Learning in Radiation Oncology’. (Eds I El Naqa, R Li, MJ Murphy) pp. 3–11. (Springer: Cham)

El-Shamy AM, Zakaria K, Abbas MA, Zein El Abedin S (2015) Anti-bacterial and anti-corrosion effects of the ionic liquid 1-butyl-1-methylpyrrolidinium trifluoromethylsulfonate. Journal of Molecular Liquids 211, 363–369.
Anti-bacterial and anti-corrosion effects of the ionic liquid 1-butyl-1-methylpyrrolidinium trifluoromethylsulfonate.Crossref | GoogleScholarGoogle Scholar |

Geladi, P, Linderholm, J (2020) Principal component analysis. In ‘Molecular Sciences and Chemical Engineering’. (Eds S Brown, R Tauler, B Walczak) pp. 17–37. (Elsevier)

Gupta P, Venkatesan M (2020) Mineral identification using unsupervised classification from hyperspectral data. In ‘Emerging Research in Data Engineering Systems and Computer Communications’. (Eds PV Krishna, M Obaidat) pp. 259–268. (Springer)

Jin X, Han J (2011) K-means clustering. In ‘Encyclopaedia of Machine Learning’. (Eds C Sammut, GI Webb) pp. 17–37. (Springer)

Ossai CI (2019) A data-driven machine learning approach for corrosion risk assessment – a comparative study. Big Data and Cognitive Computing 3, 28
A data-driven machine learning approach for corrosion risk assessment – a comparative study.Crossref | GoogleScholarGoogle Scholar |

Popoola LT, Grema AS, Latinwo GK, et al. (2013) Corrosion problems during oil and gas production and its mitigation. International Journal of Industrial Chemistry 4, 35
Corrosion problems during oil and gas production and its mitigation.Crossref | GoogleScholarGoogle Scholar |

Sattari F, Lefsrud L, Kurian D, Macciotta R (2022) A theoretical framework for data-driven artificial intelligence decision making for enhancing the asset integrity management system in the oil & gas sector. Journal of Loss Prevention in the Process Industries 74, 104648
A theoretical framework for data-driven artificial intelligence decision making for enhancing the asset integrity management system in the oil & gas sector.Crossref | GoogleScholarGoogle Scholar |

Sircar A, Yadav K, Rayavarapu K, Bist N, Oza H (2021) Application of machine learning and artificial intelligence in the oil and gas industry. Petroleum Research 6, 379–391.
Application of machine learning and artificial intelligence in the oil and gas industry.Crossref | GoogleScholarGoogle Scholar |

Skovhus TL, Eckert RB (2017) Management of MIC in the oil and gas industry. In ‘Microbiologically influenced corrosion in the upstream oil and gas industry’. (Eds TL Skovhus, D Enning, JS Lee) pp. 167–181. (CRC Press)

Syakur M, Khotimah B, Rohman E, Satoto BD (2018) Integration K-means clustering method and elbow method for identification of the best customer profile cluster In ‘IOP Conference Series: Materials Science and Engineering, Indonesia, April 2018’. (IOP Publishing)

Temizel C, Canbaz CH, Palabiyik Y, Aydin H, Tran M, Ozyurtikan MH, Yurukcu M, Johnson P (2021) ‘A Thorough Review of Machine Learning Applications in the Oil and Gas Industry.’ (Society of Petroleum Engineers)
| Crossref |

Wang H, Yajima A, Liang RY, Castaneda H (2015) A clustering approach for assessing external corrosion in a buried pipeline based on hidden Markov random field model. Structural Safety 56, 18–29.
A clustering approach for assessing external corrosion in a buried pipeline based on hidden Markov random field model.Crossref | GoogleScholarGoogle Scholar |

Xiao K, Li Z, Song J, Bai Z, Xue W, Wu J, Dong C (2021) Effects of concentrations of Fe2+ and Fe3+ on the corrosion behaviour of carbon steel in Cl− and SO42− aqueous environments. Metals and Materials International 27, 2623–2633.
Effects of concentrations of Fe2+ and Fe3+ on the corrosion behaviour of carbon steel in Cl and SO42− aqueous environments.Crossref | GoogleScholarGoogle Scholar |