Approaches to generation of fluid model for an oil field at the late stage of development

UDK: 622.276.1/.4.001.57

DOI: 10.24887/0028-2448-2026-5-102-107

Key words: PVT model, integrated modeling, statistical data analysis, anomaly detection, Black Oil, compositional model, history match, fluid flow simulation

Authors: Z.A. Loshcheva (TatNIPIneft, RF, Almetyevsk); R.R. Kashapov (TatNIPIneft, RF, Almetyevsk); I.R. Mavlyavov (TatNIPIneft, RF, Almetyevsk); G.G. Faizrakhmanov (TatNIPIneft, RF, Almetyevsk); I.I. Gadelshin (TatNIPIneft, RF, Almetyevsk); N.O. Nekrasov (TatNIPIneft, RF, Almetyevsk); A.A. Kildyushov (TatNIPIneft, RF, Almetyevsk); I.I. Khairullin (TatNIPIneft, RF, Almetyevsk)

Integrated oil field modeling necessitates development of an accurate thermodynamic (PVT) reservoir fluid model. The main challenge is associated with handling mixed laboratory data, which requires careful data organization and processing. To address this challenge, a step-wise approach is proposed, that consists of fluid sampling based on quality criteria, multivariate statistical analysis, history match of compositional model based on representative composition, and its further conversion to «Black Oil» model. Special focus is on application of machine learning methods to reveal undetected anomalies in multivariate data to substantially improve model accuracy and minimize the risk of errors. The developed method comprises conventional approaches to data verification (material balance) and advanced methods for analysis and correction of experimental data. Application of the proposed method enabled the creation of PVT model that showed excellent convergence with experimental data obtained using real downhole samples. Implementation of the proposed approach ensures required continuity and consistency of data in the geology – fluid mechanics – borehole – network chain. The proposed method minimizes personal errors during data selection and evaluation, thus significantly improving the quality of input information and ensuring its representativeness. This facilitates more accurate and efficient computing to improve forecast quality at various stages of field development. Hence, development of a reliable PVT model is an important step to improve the efficiency of oil field management process, especially for fields at late stages of development.

References

1. Coats K.H., Coats K.H., Thomas L.K., Pierson R.G., Compositional and Black Oil Reservoir Simulation, SPE-29111-MS, 2013, DOI: https://doi.org/10.2118/29111-MS

2. Watanasiri S., Brule M.R., Starling K.E., Correlation of Phase-Separation Data for Coal-Conversion Systems, AlChE Journal, 1982, V. 28, No. 4, pp. 626–637,

DOI: https://doi.org/10.1002/aic.690280415

3. Ishmuratov T.A., Islamov R.R., Khisamov A.A. et al., Compositional modeling of gas condensate systems of RN-Purneftegaz LLC field (In Russ.), Vestnik Akademii nauk Respubliki Bashkortostan, 2022, V. 45, No. 4(108), pp. 68–82, DOI: https://doi.org/10.24412/1728-5283_2022_4_68_82

4. Ahmed T., Reservoir Engineering Handbook, Houston: Gulf Professional Publishing, 2018, 1650 p.

5. Danesh A., PVT and Phase Behaviour of Petroleum Reservoir Fluids, Amsterdam: Elsevier Science & Technology Books, 1998, 400 p.

6. Whitson C.H., Brule M.R., Phase Behavior, Richardson: Henry L. Doherty Memorial Fund of AIME; Society of Petroleum Engineers, 2000, V. 20, 320 p.

7. Petroleum Resources Management System (PRMS), SPE, 2018, URL: https://millerandlents.com/wp-content/uploads/2020/03/2018-Petroleum-Resources-Management-System-V1.01.pdf