Predictive Analytics for Sucker Rod Pump Failures in Kazakhstani Oil Wells Using Machine Learning
Source abstract
Abstract
In the process of developing mature deposits, a number of geological and technological complications arise. In order to increase the smooth operation of downhole pumping equipment in oil and gas wells, companies use various methods and techniques. This article presents a novel methodology for predicting downhole pumping equipment failures. A detailed analysis was conducted on historical data regarding downhole pumping equipment failures, which were then incorporated into algorithms to calculate the operation of downhole equipment. As a result, it was discovered that in order to predict failures of downhole equipment, it is crucial to consider the historical data of the field and perform an assessment of the well’s potential. In the process of building a failure prediction model, the authors encountered the quality and completeness of historical data from the pilot field. They concluded that the data classes needed to be more balanced. The authors applied machine learning approaches to an imbalanced dataset. The significance of our approach lies in its ability to forecast equipment failures, thereby ensuring the smooth operation of wells operated by sucker rod pumps.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Oil and Gas Production Techniques · Mineral Processing and Grinding · Drilling and Well Engineering
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Petroleum Engineering, Satbayev University, Almaty 050013, Kazakhstan
KMG Engineering LLP, Astana Z05H9E8, Kazakhstan
Laboratory of Artificial Intelligence and Robotics, Institute of Information and Computational Technologies, Almaty 050000, Kazakhstan
Department of Mechanics, Al-Farabi Kazakh National University, Almaty 050040, Kazakhstan
National Laboratory Astana, Nazarbayev University, Astana 010000, Kazakhstan
Funding
Science Committee of the Ministry of Science and Higher Education of the Republic of Kazakhstan
References (17)
Show source-provided references
1. Merembayev, T., Kurmangaliyev, D., Bekbauov, B., and Amanbek, Y. (2021). A Comparison of machine learning algorithms in predicting lithofacies: Case studies from Norway and Kazakhstan. Energies, 14.10.3390/en14071896 ↗
2. 70 years of machine learning in geoscience in review10.1016/bs.agph.2020.08.002 ↗
3. Interpretable machine learning models for failure cause prediction in imbalanced oil pipeline data10.1088/1361-6501/ad3570 ↗
4. Marin, A.A., Busaidy, S., Murad, M., Balushi, I., Riyami, A., Jahwari, S., Ghadani, A., Ferdiansyah, E., Shukaili, G., and Amri, F. (2019, January 12). ESP Well and Component Failure Prediction in Advance using Engineered Analytics—A Break-through in Minimizing Unscheduled Subsurface Deferments. Proceedings of the Abu Dhabi International Petroleum Exhibition & Conference, Abu Dhabi, United Arab Emirates.
5. Khabibullin, R.A., Shabonas, A.R., Gurbatov, N.S., and Timonov, A.V. (2020, January 26). Prediction of ESPs Failure Using ML at Western Siberia Oilfields with Large Number of Wells. Proceedings of the SPE Russian Petroleum Technology Conference, Virtual.10.2118/201881-RU ↗
6. Peng, L., Han, G., Pagou, A.L., Zhu, L., Ma, H., Wu, J., and Chai, X. (2021, January 22). A Predictive Model to Detect the Impending Electric Submersible Pump Trips and Failures. Proceedings of the SPE Annual Technical Conference and Exhibition, Dubai, United Arab Emirates.10.2118/206150-MS ↗
7. Andrade Marin, A., Al Balushi, I., Al Ghadani, A., Al Abri, H., Al Zaabi, A.K.S., Dhuhli, K., Hadhrami, I.A., Al Hinai, S.H., Al Aufi, F.M., and Al Bimani, A.A. (2021, January 18). Real Time Implementation of ESP Predictive Analytics—Towards Value Realization from Data Science. Proceedings of the Abu Dhabi International Petroleum Exhibition and Conference, Abu Dhabi, United Arab Emirates.10.2118/207550-MS ↗
8. Li, Q., Li, Q., Wang, F., Wu, J., and Wang, Y. (2024). The carrying behavior of water-based fracturing fluid in shale reservoir fractures and molecular dynamics of sand-carrying mechanism. Processes, 12.10.3390/pr12092051 ↗
9. A hybrid cluster-borderline SMOTE method for imbalanced data of rock groutability classification10.1007/s10064-021-02523-9 ↗
10. Chen, Z., Selere, O.O., and Seng, N.L.C. (2022, January 27–28). Equipment failure analysis for oil and gas industry with an ensemble predictive model. Proceedings of the 9th International Conference on Computational Science and Technology (ICCST 2022), Johor Bahru, Malaysia.
11. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., and Liu, T.-Y. (2017, January 4–9). LightGBM: A highly efficient gradient boosting decision tree. Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA.
12. Boosting methods for multi-class imbalanced data classification: An experimental10.1186/s40537-020-00349-y ↗
13. Grandini, M., Bagli, E., and Visani, G. (2020). Metrics for multi-class classification: An overview. arXiv.
14. Opitz, J., and Burst, S. (2019). Macro f1 and macro f1. arXiv.
15. Merembayev, T., and Amanbek, Y. (2023, January 3–6). Natural fracture network model using machine learning approach. Proceedings of the Computational Science and Its Applications—ICCSA 2023 Workshops, Athens, Greece. Proceedings, Part IV.10.1007/978-3-031-37114-1_26 ↗
16. Merembayev, T., and Amanbek, Y. (2022, January 26–29). Time-series event prediction for the uranium production wells using machine learning algorithms. Proceedings of the 56th U.S. Rock Mechanics/Geomechanics Symposium, Santa Fe, NM, USA.10.56952/ARMA-2022-0415 ↗
17. Lundberg, S. (2017). A unified approach to interpreting model predictions. arXiv.