Application of machine learning to predict the time evolution of condensate to gas ratio for planning and management of gas - condensate fields

  • Huu Hai Ngo Bien Dong Petroleum Operating Company (Bien Dong POC)
  • Xuan Vinh Trinh Bien Dong Petroleum Operating Company (Bien Dong POC)
  • Ngoc Tan Nguyen Bien Dong Petroleum Operating Company (Bien Dong POC)
  • Ky Son Hoang Bien Dong Petroleum Operating Company (Bien Dong POC)
  • Tuan Anh Ngo Bien Dong Petroleum Operating Company (Bien Dong POC)
  • Ngoc Trung Tran Bien Dong Petroleum Operating Company (Bien Dong POC)
  • Vu Tung Tran Bien Dong Petroleum Operating Company (Bien Dong POC)
  • Sy Tuan Nguyen Van Lang University
Keywords: Machine learning, condensate to gas ratio, production forecast


One of the most important parameters for evaluating, forecasting, and managing gas - condensate fields is the evolution of the condensate to gas ratio (CGR) over time. This parameter tends to decrease as reservoir pressure declines. Conventionally, gas and condensate samples are collected initially at the time starting production and periodically later to conduct laboratory analyses of fluid composition, properties and CGR. However, sampling, transporting and analysing samples take time and effort and, therefore, could be very expensive. To predict CGR over time, likewise, dynamic models are also frequently used. However, these models could include many uncertainties due to the assumption of input data, including reservoir structures, fluid phase interaction, and reservoir property distribution. Therefore, application of machine learning to predict the time evolution of CGR in this research is a new and effective approach to supplement conventional methods. 


Curtis H. Whitson, Øivind Fevang, and Tao Yang, “Gas condensate PVT - what’s really important and why?”, IBC Conference “Optimisation of Gas Condensate Fields”, London, 28 - 29 January 1999.

Meisam Karbalaee Akbari, Farhang Jalali Farahani, and Yaser Abdy, “Dewpoint pressure estimation of gas condensate reservoirs, using artificial neural network (ANN)”, SPE Europec/ EAGE Annual Conference and Exhibition, London, United Kingdom, 11 - 14 June 2007. DOI: 10.2118/107032-MS.

Seyed Mohammad Javad Majidi, Amin Shokrollahi, Milad Arabloo, Ramin Mahdikhani- Soleymanloo, and Mohsen Masihi, “Evolving an accurate model based on machine learning approach for prediction of dew-point pressure in gas condensate reservoirs”, Chemical Engineering Research and Design, Volume 92, Issue 5, pp. 891 - 902, 2014. DOI: 10.1016/j.cherd.2013.08.014.

Zhi Zhong, Siyan Liu, Mohammad Kazemi, and Timothy R. Carr, “Dew point pressure prediction based on mixed-kernels-function support vector machine in gas-condensate reservoir”, Fuel, Volume 232, pp. 600 - 609, 2018. DOI: 10.1016/j. fuel.2018.05.168.

Princewill Ikpeka, Johnson Ugwu, Paul Russell, and Gobind Pillai, “Performance evaluation of machine learning algorithms in predicting dew point pressure of gas condensate reservoirs”, SN Applied Sciences, Volume 2, 2020. DOI: 10.1007/s42452-020-03811-x.

Sohrab Zendehboudi, Mohammad Ali Ahmadi, Lesley James, and Ioannis Chatzis, “Prediction of condensate-to-gas ratio for retrograde gas condensate reservoirs using artificial neural network with particle swarm optimization”, Energy & Fuels, Volume 26, Issue 6, pp. 3432 - 3447, 2012. DOI: 10.1021/ef300443j.

Mohammad Ali Ahmadi, Mohammad Ebadi, Payam Soleimani Marghmaleki, and Mohammad Mahboubi Fouladi, “Evolving predictive model to determine condensate-to-gas ratio in retrograded condensate gas reservoirs”, Fuel, Volume 124, pp. 241 - 257, 2014. DOI: 10.1016/j.fuel.2014.01.073.

Hana AlMatouq, Mohammed Alabbad, and Fatai Anifowose, “An artificial intelligence approach to predict molar compositions of reservoir fluid components”, SPE Gas & Oil Technology Showcase and Conference, Dubai, UAE, 21 - 23 October 2019. DOI: 10.2118/198555-MS.

Triệu Hùng Trường, Trần Vũ Tùng và nnk, “Nghiên cứu xây dựng bộ công cụ trí tuệ nhân tạo hỗ trợ đánh giá phân tích, liên kết tài liệu địa chất, địa vật lý giếng khoan và số liệu khai thác để nâng cao hiệu quả quản lý, khai thác mỏ khí condensate Hải Thạch - Mộc Tinh Lô 05-2; 05-3, thuộc Biển Đông Việt Nam”, Đề tài cấp Nhà nước thuộc “Chương trình khoa học và công nghệ trọng điểm cấp quốc gia phục vụ đổi mới, hiện đại hóa công nghệ khai thác và chế biến khoáng sản đến năm 2025”, Mã số 077.2021.CNKK. QG/HĐKHCN.

Kyungbook Lee, Jungtek Lim, Daeung Yoon, and Hyungsik Jung, “Prediction of shale gas production at Duvernay formation using deep-learning algorithm”, SPE Journal, Volume 24, Issue 6, pp. 2423 - 2437, 2019. DOI: 10.2118/195698-PA.

Cheng Zhan, Sathish Sankaran, Vincent LeMoine, Jeremy Graybill, and Didi-Ooi Sher Mey, “Application of machine learning for production forecasting for unconventional resources”, Unconventional Resources Technology Conference, Denver, Colorado, USA, 22 - 24 July 2019. DOI: 10.15530/urtec-2019-47.

Ngo Huu Hai, Pham Hoang Duy, Nguyen Ngoc Tan, Hoang Ky Son, Tran Ngoc Trung và Tran Vu Tung, “Application of machine learning to decline curve analysis (DCA) for gas-condensate production wells with complex production history due to add-on perforation of new reservoirs”, Petrovietnam Science, Techonology & Innovation, Volume 2, pp. 4 - 9, 2023. DOI: 10.47800/PVSI.2023.02-01.

How to Cite
Ngo, H. H., Trinh, X. V., Nguyen, N. T., Hoang, K. S., Ngo, T. A., Tran, N. T., Tran, V. T., & Nguyen, S. T. (2024). Application of machine learning to predict the time evolution of condensate to gas ratio for planning and management of gas - condensate fields. Petrovietnam Journal, (1), 58-66.

Most read articles by the same author(s)