AI-Enabled Predictive Maintenance of Wellhead and Pressure Control Equipment in Upstream Oil and Gas Operations

References

[1] Wanasinghe TR, Gosine RG, James LA, Mann GKI, de Silva O, Warrian PJ. The Internet of Things in the Oil and Gas Industry: A Systematic Review. IEEE Internet Things J. 2020;7(9):8654-8673. IEEE

[2] Wanasinghe TR, Wroblewski L, Petersen BK, Gosine RG, James LA, de Silva O, et al. Digital Twin for the Oil and Gas Industry: Overview, Research Trends, Opportunities, and Challenges. IEEE Access. 2020;8:104175-104197. IEEE

[3] Choubey S, Karmakar GP. Artificial intelligence techniques and their application in oil and gas industry. Artif Intell Rev. 2021;54:3665-3683. Springer

[4] Sircar A, Yadav K, Rayavarapu K, Bist N, Oza H. Application of machine learning and artificial intelligence in oil and gas industry. Petroleum Research. 2021;6(4):379-391. ScienceDirect

[5] Tariq Z, Aljawad MS, Hasan A, Murtaza M, Mohammed E, El-Husseiny A, et al. A systematic review of data science and machine learning applications to the oil and gas industry. J Pet Explor Prod Technol. 2021;11:4339-4374. Springer

[6] Ji D, Wang C, Li J, Dong H. A review: data driven-based fault diagnosis and RUL prediction of petroleum machinery and equipment. Syst Sci Control Eng. 2021;9(1):724-747. Taylor & Francis

[7] Jahren JH, Matias J, Jaschke J. Data-driven modelling of choke valve erosion using data simulated from a first principles model. Comput Aided Chem Eng. 2021;50:773-778. ScienceDirect

[8] Alarifi SA. Workflow to predict wellhead choke performance during multiphase flow using machine learning. J Pet Sci Eng. 2022;214:110563. ScienceDirect

[9] Ibrahim NM, Alharbi AA, Alzahrani TA, Abdulkarim AM, Alessa IA, Hameed AM, et al. Well Performance Classification and Prediction: Deep Learning and Machine Learning Long Term Regression Experiments on Oil, Gas, and Water Production. Sensors. 2022;22(14):5326. MDPI

[10] Ikegwu AC, Nweke HF, Anikwe CV, Alo UR, Okonkwo OR. Big data analytics for data-driven industry: a review of data sources, tools, challenges, solutions, and research directions. Cluster Comput. 2022;25(5):3343-3387. Springer

[11] Syed FI, AlShamsi A, Dahaghi AK, Neghabhan S. Application of ML & AI to model petrophysical and geomechanical properties of shale reservoirs - A systematic literature review. Petroleum. 2022;8(2):158-166. ScienceDirect

[12] Sprunger C, Muther T, Syed FI, Dahaghi AK, Neghabhan S. State of the art progress in hydraulic fracture modeling using AI/ML techniques. Model Earth Syst Environ. 2022;8:1-13. Springer

[13] Mohamed Almazrouei S, Dweiri F, Aydin R, et al. A review on the advancements and challenges of artificial intelligence based models for predictive maintenance of water injection pumps in the oil and gas industry. SN Appl Sci. 2023;5:391. Springer

[14] Shao X, Cai B, Liu Y, Zhang J, Sui Z, Feng Q. Remaining useful life prediction via a hybrid DBN-KF-based method: A case of subsea Christmas tree valves. Reliab Eng Syst Saf. 2023;238:109405. ScienceDirect

[15] Guo J, Wan JL, Yang Y, Dai L, Tang A, Huang B, et al. A deep feature learning method for remaining useful life prediction of drilling pumps. Energy. 2023;282:128442. ScienceDirect

[16] Szpytko J, Salgado-Duarte Y. Maintenance Management Practices Using Digital Twins Framework: Oil Extraction Pumping System Case Study. IFAC-PapersOnLine. 2023;56(2):6612-6617. ScienceDirect

[17] Dabiri MS, Hadavimoghaddam F, Ashoorian S, Schaffie M, Hemmati-Sarapardeh A, et al. Modeling liquid rate through wellhead chokes using machine learning techniques. Sci Rep. 2024;14:6945. Nature

[18] Kaleem W, Tewari S, Fogat M, Martyushev DA. A hybrid machine learning approach based study of production forecasting and factors influencing the multiphase flow through surface chokes. Petroleum. 2024;10(2):354-371. ScienceDirect

[19] Shao X, Cai B, Gao L, Zhang Y, Yang C, Gao C. Data-model-linked remaining useful life prediction method with small sample data: A case of subsea valve. Reliab Eng Syst Saf. 2024;250:110323. ScienceDirect

[20] Li H, Zhang Z, Li T, Si X. A review on physics-informed data-driven remaining useful life prediction: Challenges and opportunities. Mech Syst Signal Process. 2024;209:111120. ScienceDirect

[21] Wadinger M, Kvasnica M. Adaptable and Interpretable Framework for Anomaly Detection in SCADA-based industrial systems. Expert Syst Appl. 2024;246:123200. ScienceDirect

[22] Wang M, Su X, Song H, Wang Y, Yang X. Enhancing predictive maintenance strategies for oil and gas equipment through ensemble learning modeling. J Pet Explor Prod Technol. 2025;15:46. Springer

[23] Khalili Y, Ahmadi M, Moraveji MK. A Comprehensive Review of Failure Modes in Electrical Submersible Pumps: Diagnosis, Predictive Maintenance, and Engineer's Guide. Arab J Sci Eng. 2025;50(24):20445-20466. Springer

[24] Khalili Y, Ahmadi M, Moraveji MK. Time-aware predictive maintenance of electrical submersible pumps using CatBoost ensemble learning and trend-based labeling. J Pet Explor Prod Technol. 2025;15:147. Springer

[25] Almuraia A, He F, Khan M. AI-Driven Maintenance Optimisation for Natural Gas Liquid Pumps in the Oil and Gas Industry: A Digital Tool Approach. Processes. 2025;13(5):1611. MDPI

[26] Povoas MDS, Moreira JF, Martins Neto SV, Carvalho CAS, Cezario BS, Guedes ALA, et al. Artificial Intelligence in the Oil and Gas Industry: Applications, Challenges, and Future Directions. Appl Sci. 2025;15(14):7918. MDPI

[27] Belo RC, Petry RH, Salvador TT, Pimenta M, et al. Fundamental requirements of Digital Twins for production system in Oil and Gas Industry: A systematic literature review. Inf Softw Technol. 2025;184:107742. ScienceDirect

[28] Jones JF, Raj V, Abas PE, Petra MI, Sivan D, Kumar KS, Jose R. Application of artificial intelligence for asset integrity management of offshore oil and gas pipelines. Life Cycle Reliab Saf Eng. 2025;14:503-530.

[29] Alfrayan DW, Almulla NA, Isawi WA, Alali MB. Predicting Choke Valve Erosion in HPHT Gas Wells Using Machine Learning. In: ADIPEC 2025. Society of Petroleum Engineers; 2025. Paper SPE-229201-MS. OnePetro

[30] Li W, Chen J, Chen S, Li P, Zhang B, Wang M, et al. A Comprehensive Review of Artificial Intelligence-Based Algorithms for Predicting the Remaining Useful Life of Equipment. Sensors. 2025;25(14):4481.

@article{1722968,
      author = {Thulasiram Yendrapalli},
      title = {AI-Enabled Predictive Maintenance of Wellhead and Pressure Control Equipment in Upstream Oil and Gas Operations},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {10},
      number = {3},
      pages = {2359-2371},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1722968.pdf},
      abstract = {The oil and gas wells which are operated in the upstream sector depend on wellhead and pressure-control equipment in order to maintain pressure inside acceptable limits, to regulate the production of multiple phases, and to isolate the well should the operating limits be exceeded. Because failures may take place in the production chokes, in the master and wing valves, in the surface safety valves, in the actuators, in the seals, and in the pressure monitoring equipment, such failures can result in a slowdown of production, a loss of containment, or an emergency shutdown. This review analyses in detail how artificial intelligence is able to transform the maintenance of such assets from one that is based on scheduled inspections and corrective repairs to one that is based on the current condition and is capable of making predictions. In order to answer four closely related questions the body of literature published between 2020 and 2025 was compiled: what failure mechanisms give rise toward observable data signatures; which machine-learning, deep-learning, probabilistic, and hybrid methods are suitable for carrying out anomaly detection, diagnosis, health-index estimation, and prediction of remaining useful life; what limits the application of these methods in practice; and how the predictions should be incorporated into operational and maintenance decisions. Evidence from studies on wellhead choke modelling, subsea valve prognostics, petroleum machinery, SCADA anomaly detection, pump predictive maintenance, and digital-twin research shows that the greatest real benefits are not due to any single algorithm but result from combining reliable sensing, knowledge of the operating mode, time-aware features, uncertainty estimates, and engineering constraints. Ensemble models work well with structured and noisy field data, whereas temporal deep-learning methods are useful when high-frequency historical data and representative failure cases are available. Investigations into valves with small sample sizes have shown the advantages of using probabilistic and data-model linked prognostics, and digital twins provide a way of continuously aligning asset physics with data-based intelligence. The review proposes an asset-centred deployment architecture in which detection, diagnosis, prognosis, risk ranking, and maintenance planning form a closed loop. The research priorities are the development of standardised failure taxonomies, cross-field validation, physics-informed learning, uncertainty-aware remaining life, explainable alarms, and prospective field trials that focus on safety outcomes, the avoidance of production deferment, and improvements in maintenance efficiency instead of on accuracy alone.},
      month = {September},
  }