Application of machine learning methods to assess progressive cavity pumps (PCPs) performance in coal seam gas (CSG) wells
Fahd Saghir A C , M. E. Gonzalez Perdomo A and Peter Behrenbruch BA Australian School of Petroleum and Energy Resources, Santos Petroleum Engineering Building, University of Adelaide, SA 5005, Australia.
B Bear and Brook Consulting, 135 Hilda Street Corinda, Qld 4075, Australia.
C Corresponding author. Email: fahd.saghir@adelaide.edu.au
The APPEA Journal 60(1) 197-214 https://doi.org/10.1071/AJ19044
Submitted: 12 December 2019 Accepted: 23 January 2020 Published: 15 May 2020
Abstract
In Queensland, progressive cavity pumps (PCPs) are the artificial lift method of choice in coal seam gas (CSG) wells, and this choice of artificial lift production stems from the ability of PCPs to better manage the production of liquids with suspended solids. As with any mechanical pumping system, PCPs are prone to natural wear and tear over their operational life, and with the production of coal fines and inter-burden, the run life of PCPs in CSG wells is significantly reduced. Another factor to consider with the use of PCPs is their reliability. As per the CSG production data available through the Queensland Government Data Portal, there are approximately 6400 wells operational in the state as of December 2018. This number is expected to grow significantly over the next decade to meet both international and domestic gas utilisation requirements. Operators supervising these wells rely on a reactive or exception-based approach to manage well performance. In order to efficiently operate thousands of PCP wells, it is pertinent that a benchmark methodology is devised to autonomously monitor PCP performance and allow operators to manage wells by exception. In this study, we will cover the application of machine learning methods to understand anomalous PCP behaviour and overall pump performance based on the analysis of multivariate time-series data. An innovative time-series data approximation and image conversion technique will be discussed in this paper, along with machine learning methods, which will focus on a scalable and autonomous approach to cluster PCP performance and detection of anomalous pump behaviour in near real-time. Results from this study show that clustering real-time data based on converted time-series images helps to pro-actively detect change in PCP performance. Discovery of anomalous multivariate events is also achieved through time-series image conversion. This study also demonstrates that clustering time-series data noticeably improves the real-time monitoring capabilities of PCP performance through improved visual analytics.
Keywords: artificial lift, data analytics, time-series data, visual analytics.
Fahd Saghir is an Automation Engineer with 10+ years of experience in the digital oilfield domain. Since completing his BSc in Electrical Engineering from the University of Houston in 2006, Fahd has been involved in creating digital solutions for the production and operations verticals within the oil and gas sector, based on innovative hardware and software technologies. Fahd is currently pursuing a PhD in Petroleum Engineering from the University of Adelaide. His research work focuses on the use of Machine Learning methods to classify and detect abnormal PCP performance in CSG wells. This research investigates an innovative approach where time-series data is transformed into heatmap images, and the images are then used to classify PCP performance in near real-time. Fahd is also an active SPE volunteer and has participated as a speaker and moderator at multiple SPE conferences and webinars. He is currently a member of the Digital Solutions Committee, which falls under SPE’s Digital Energy Technical Section. |
Mary Gonzalez is a senior lecturer at the Australian School of Petroleum and Energy Resources (ASPER) at the University of Adelaide. Her research and teaching focus is on reservoir and production engineering, particularly production enhancement and optimisation. She joined the ASP in 2009 after several years of experience in the oil and gas industry, where she provided practical petroleum engineering, consultancy services and solutions in the areas of subsurface and production engineering. Mary has published several articles in peer-reviewed journals and presented at international conferences. She has served as a reviewer for different journals and as a mentor for young professionals, and she is the Community Education Chair and the ASPER Faculty Officer for the SPE. |
Professor Peter Behrenbruch is currently the Managing Director of Bear and Brook Consulting Pty Ltd (since 2003). He is also an Adjunct Professor at the Ho Chi Minh University of Technology, Faculty of Geology and Petroleum Engineering, Vietnam. Behrenbruch’s last full-time industry position (2008–2009) was Chief Operating Officer/Managing Director for East Puffin (SINOPEC) for the Puffin offshore development project, Timor Sea. He held a similar position (2007–2008) for AED Oil Ltd on the same project. He was also the inaugural Head of the School of Petroleum Engineering and Management (2001–2003) and full-time Professor at the University of Adelaide (2001–2007), with tenure since 2004. More recently, he taught as a Visiting Professor at the University of Western Australia (2014), Curtin University (2014), Stanford University (2000) and several other institutions. |
References
Ali, M., Jones, M., Xie, X., and Williams, M. (2019). TimeCluster: dimension reduction applied to temporal data for visual analytics. The Visual Computer 35, 1013–1026.| TimeCluster: dimension reduction applied to temporal data for visual analytics.Crossref | GoogleScholarGoogle Scholar |
Bettaiah, V. (2014). The hierarchical piecewise linear approximation of time series data. PhD Thesis, Department of Computer Science, University of Alabama in Hunstville. Available at https://cdm16608.contentdm.oclc.org/digital/collection/p16608coll23/id/2485 [Verified 30 January 2020]
Biniwale, S. S., and Trivedi, R. (2012). Managing LNG Deliverability: An Innovative Approach Using Neural Network and Proxy Modeling for Australian CSG Assets. In ‘Abu Dhabi International Petroleum Conference and Exhibition, 11-14 November, Abu Dhabi, UAE.’ (Society of Petroleum Engineers.)
Campello, R. J. G. B., Moulavi, D., and Sander, J. (2013). Density-Based Clustering Based on Hierarchical Density Estimates. In: ‘Advances in Knowledge Discovery and Data Mining’. (Eds J. Pei, V. S. Tseng, L. Cao, H. Motoda and G. Xu.) pp. 160–172. (Springer: Berlin Heidelberg)
Chaovalit, P., Gangopadhyay, A., Karabatis, G., and Chen, Z. (2011). Discrete wavelet transform-based time series analysis and mining. ACM Computing Surveys 43, 1–37.
| Discrete wavelet transform-based time series analysis and mining.Crossref | GoogleScholarGoogle Scholar |
Dan, J., Shi, W., Dong, F., and Hirota, K. (2013). Piecewise Trend Approximation: A Ratio-Based Time Series Representation. Abstract and Applied Analysis 2013, 603629.
| Piecewise Trend Approximation: A Ratio-Based Time Series Representation.Crossref | GoogleScholarGoogle Scholar |
Duvignau, R., Gulisano, V., Papatriantafilou, M., and Savic, V. (2018). Piecewise Linear Approximation in Data Streaming: Algorithmic Implementations and Experimental Analysis. Available at https://arxiv.org/abs/1808.08877 [Verified 30 January 2020]
Firouzi, M., and Rathnayake, S. (2019). Prediction of the Flowing Bottom-Hole Pressure Using Advanced Data Analytics. In ‘SPE/AAPG/SEG Asia Pacific Unconventional Resources Technology Conference, 18-19 November, Brisbane, Australia.’. (Unconventional Resources Technology Conference).
Gao, Y., and Lin, J. (2018). Efficient Discovery of Variable-length Time Series Motifs with Large Length Range in Million Scale Time Series. Available at https://arxiv.org/abs/1802.04883 [verified 30 January 2020]
Guigou, F., Collet, P., and Parrend, P. (2017). Anomaly detection and motif discovery in symbolic representations of time series. Technical Report No 69427/2. (Complex System Digital Campus, UNITWIN UNESCO). Available at https://hal.archives-ouvertes.fr/hal-01507517/document [Verified 30 January 2020]
Hatami, N., Gavet, Y., and Debayle, J. (2017). Classification of Time-Series Images Using Deep Convolutional Neural Networks. Available at https://arxiv.org/abs/1710.00886 [Verified 30 January 2020]
Hoday, J. P., Knafl, M., Prosper, C., and Braas, M. (2013). Diagnosing PCP Failure Characteristics using Exception Based Surveillance in CSG. In ‘SPE Progressing Cavity Pumps Conference. Calgary, Alberta, Canada.’ (Society of Petroleum Engineers.)
Kumar, N., Lolla, V. N., Keogh, E. J., Lonardi, S., Ratanamahatana, C., and Wei, L. (2005). Time-series Bitmaps: a Practical Visualization Tool for Working with Large Time Series Databases. In ‘Proceedings of the 2005 SIAM International Conference on Data Mining’. Available at https://doi.org/10.1137/1.9781611972757.55 [Verified 30 January 2020]
Lin, J., Keogh, E., and Lonardi, S. (2005). Visualizing and discovering non-trivial patterns in large time series databases. Information Visualization 4, 61–82.
| Visualizing and discovering non-trivial patterns in large time series databases.Crossref | GoogleScholarGoogle Scholar |
Lin, J., Keogh, E., Wei, L., and Lonardi, S. (2007). Experiencing SAX: a novel symbolic representation of time series. Data Mining and Knowledge Discovery 15, 107–144.
| Experiencing SAX: a novel symbolic representation of time series.Crossref | GoogleScholarGoogle Scholar |
Matthews, C. M., Zahacy, T. A., Alhanati, F. J. S., Skoczylas, P., and Dunn, L. J. (2007). ‘Petroleum Engineering Handbook. Production Operations Engineering’. (Society of Petroleum Engineers: Richardson, Texas).
McKinney, W. (2013). ‘Python for Data Analysis’. (O’Reilly Media: Sebastopol, California).
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E. (2011). Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research 12, 2825–2830.
Prosper, C., and West, D. (2018). Case Study Applied Machine Learning to Optimise PCP Completion Design in a CBM Field. In ‘SPE Asia Pacific Oil and Gas Conference and Exhibition, 23–25 October, Brisbane, Australia.’ (Society of Petroleum Engineers.)
Saghir, F., Gonzalez Perdomo, M. E., and Behrenbruch, P. (2019a). Application of Exploratory Data Analytics (EDA) in Coal Seam Gas Wells with Progressive Cavity Pumps (PCPs). In ‘ SPE/IATMI Asia Pacific Oil & Gas Conference and Exhibition, 29–31 October, Bali, Indonesia.’ (Society of Petroleum Engineers.)
Saghir, F., Perdomo, M. E. G., and Behrenbruch, P. (2019b). Converting Time Series Data into Images: An Innovative Approach to Detect Abnormal Behavior of Progressive Cavity Pumps Deployed in Coal Seam Gas Wells. In ‘SPE Annual Technical Conference and Exhibition, 30 September–2 October, Calgary, Canada: (Society of Petroleum Engineers.)
Saghir, F., Gonzalez Perdomo, M. E., and Behrenbruch, P. (2019c). Machine Learning for Progressive Cavity Pump Performance Analysis: A Coal Seam Gas Case Study. In ‘SPE/AAPG/SEG Asia Pacific Unconventional Resources Technology Conference, 18–19 November,. Brisbane, Australia.’ (Unconventional Resources Technology Conference.)
Shaw, P. K., Saha, D., Ghosh, S., Janaki, M. S., and Iyengar, A. N. S. (2015). Investigation of coherent modes in the chaotic time series using empirical mode decomposition and discrete wavelet transform analysis. Chaos, Solitons and Fractals 78, 285–296.
| Investigation of coherent modes in the chaotic time series using empirical mode decomposition and discrete wavelet transform analysis.Crossref | GoogleScholarGoogle Scholar |
Sivaraks, H., and Ratanamahatana, C. A. (2015). Robust and Accurate Anomaly Detection in ECG Artifacts Using Time Series Motif Discovery. Computational and Mathematical Methods in Medicine 2015, 453214.
| Robust and Accurate Anomaly Detection in ECG Artifacts Using Time Series Motif Discovery.Crossref | GoogleScholarGoogle Scholar | 25688284PubMed |
Vidaurre, D., Iead, R., Harrison, S., Smith, S., and Woolrich, M. (2014). Dimensionality reduction for time series data. Available at https://arxiv.org/abs/1406.3711 [Verified 30 January 2020]
Wang, Z., and Oates, T. (2015). Imaging Time-Series to Improve Classification and Imputation. In Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence (IJCAI 2015). Available at https://www.ijcai.org/Proceedings/15/Papers/553.pdf [Verified 30 January 2020]
Zhong, R., Johnson, R. L., and Chen, Z. (2019). Using Machine Learning Methods to Identify Coals from Drilling and Logging-While-Drilling LWD Data. In ‘SPE/AAPG/SEG.) Asia Pacific Unconventional Resources Technology Conference, 18–19 November,. Brisbane, Australia.’ (Unconventional Resources Technology Conference.)
