Machine Learning-Based Failure Prediction in Electric Submersible Pumps
DOI:
https://doi.org/10.15377/2409-5818.2026.13.2Keywords:
Artificial lift, Fault diagnosis, Machine learning, Ensemble models, Predictive maintenance, Artificial intelligence in oilfields, Electric submersible pumps (ESP).Abstract
Electric Submersible Pumps (ESP) are among the most widely used artificial lift methods in oil production wells. Despite their effectiveness, ESP systems are prone to unexpected failures that lead to significant production losses, costly workover operations, and extended downtime. Traditional maintenance strategies, such as corrective or time-based maintenance, are often insufficient in predicting early signs of ESP degradation. This paper presents a machine learning–based approach for predicting potential ESP failures using historical SCADA operational data. The framework integrates statistical feature engineering via a sliding window, SMOTE-based class balancing, and a comparative evaluation of four algorithms: Random Forest, XGBoost (XGB), Extra Trees, and Logistic Regression. The results demonstrate that tree-based ensemble models achieve exceptional classification accuracy up to 99.96%, significantly outperforming the baseline Logistic Regression model (99.83%) and effectively identifying early indicators of ESP performance degradation with minimal misclassifications. Feature importance evaluation revealed that thermal and hydraulic parameters serve as the primary diagnostic anchors. Integrating this predictive framework into automated systems can successfully transition field operations from reactive to proactive maintenance, optimizing the economic lifecycle of ESP systems.
References
[1] Mahmud WM, Jaluta AM. Decade of electrical submersible pumps failures: case study on performance and optimization. J Appl Sci Environ Manag. 2025; 29(3): 747-54. http://doi.org/10.4314/jasem.v29i3.8
[2] Mahmud WM. Integrated evaluation of electric submersible pump failures under diverse field operating conditions. J Energy Power Technol. 2026; 8(1): 006. http://doi.org/10.21926/jept.2601006
[3] Christopher EF, Igbinere SA, Ohenhen I. A machine learning-based predictive maintenance for electric submersible pumps in petroleum production. NIPES Adv Eng Des Technol. 2025; 7(2): 37-42. http://doi.org/10.5281/zenodo.16637356
[4] Abdalla R, Samara H, Perozo N, Paz Carvajal C, Jaeger P. Machine learning approach for predictive maintenance of the electrical submersible pumps (ESPs). ACS Omega. 2022; 7(21): 17641-51. http://doi.org/10.1021/acsomega.1c05881
[5] Peng L, Han G, Sui X, Pagou AL, Zhu L, Shu J. Predictive approach to perform fault detection in electrical submersible pump systems. ACS Omega. 2021; 6(11): 7554-63. http://doi.org/10.1021/acsomega.0c05808
[6] Jani B, Chetan C, Baria D, Barjod U, Chauhan J, Bangadiya J. Performance investigation on double acting reciprocating pump. Int J Innov Res Technol. 2019; 6(4): 14-7.
[7] Brasil J, Maitelli C, Nascimento J, Chiavone-Filho O, Galvão E. Diagnosis of operating conditions of the electrical submersible pump via machine learning. Sensors (Basel). 2023; 23(1): 279. http://doi.org/10.3390/s23010279
[8] Batallas E, Enriquez L, Hill C, Reyes L, Lavin M, Loaiza R, et al. Enhancing ESP reliability: synergizing novel machine learning prediction model with operational practices for failure reduction. In: ADIPEC; 2025 Nov; Abu Dhabi, United Arab Emirates. http://doi.org/10.2118/230049-MS
[9] Jordan MI, Mitchell TM. Machine learning: trends, perspectives, and prospects. Science. 2015; 349(6245): 255-60. http://doi.org/10.1126/science.aaa8415
[10] Saptadi S, Widodo A, Athaillah M, Ayyasyi M. Implementation of machine learning methods in predicting failures in electrical submersible pump machines. Multidiscip Sci J. 2024; 7(3): 2025137. http://doi.org/10.31893/multiscience.2025137
[11] Buabeng A, Simons A, Frempong NK, Ziggah YY. A novel hybrid predictive maintenance model based on clustering, SMOTE and multi-layer perceptron neural network optimised with grey wolf algorithm. SN Appl Sci. 2021; 3: 593. https://doi.org/10.1007/s42452-021-04598-1
[12] Chen T, Guestrin C. XGBoost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '16); 2016. p. 785-94. http://doi.org/10.1145/2939672.2939785
[13] Guedes GS, Silva GC, Rossi RG. Time-aware predictive maintenance of electrical submersible pumps using CatBoost ensemble learning and trend-based labeling. Eng Appl Artif Intell. 2023; 123: 106206.
[14] Jani DB. An overview on causes and prevention of cavitation in centrifugal pumps. Int J Inventory Control Manag. 2019; 9(1): 1-10.
[15] Adhikari S, Majumdar J. Cost-optimised machine learning model comparison for predictive maintenance. In: 2020 IEEE International Conference on Advent Trends in Multidisciplinary Research and Innovation (ICATMRI); 2020. p. 1-6.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Walid Mohamed Mahmud, Masara A. Aldowro
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
All the published articles are licensed under the terms of the Creative Commons Attribution Non-Commercial License (CC BY-NC 4.0) which permits unrestricted, non-commercial use, distribution and reproduction in any medium, provided the work is properly cited.
