Forecasting oil production in unconventional reservoirs using long short term memory network coupled support vector regression method: A case study

Abstract

Production prediction is crucial for the recovery of hydrocarbon resources. However, accurate and rapid production forecasting remains challenging for unconventional reservoirs due to the complexity of the percolation process and the scarcity of available data. To address this problem, a novel model combining a long short-term memory network (LSTM) and support vector regression (SVR) was proposed to forecast tight oil production. Three variables, the tubing head pressure, nozzle size, and water rate were utilized as the inputs of the presented machine-learning workflow to account for the influence of operational parameters. The time-series response of tight oil production was the output and was predicted by the optimized LSTM model. An SVR-based residual correction model was constructed and embedded with LSTM to increase the prediction accuracy. Case studies were carried out to verify the feasibility of the proposed method using data from two wells in the Ma-18 block of the Xinjiang oilfield. Decline curve analysis (DCA) methods, LSTM and artificial neural network (ANN) models were also applied in this study and compared with the LSTM-SVR model to prove its superiority. It was demonstrated that introducing residual correction with the newly proposed LSTM-SVR model can effectively improve prediction performance. The LSTM-SVR model of Well A produced the lowest prediction root mean square error (RMSE) of 5.42, while the RMSE of Arps, PLE Duong, ANN, and LSTM were 5.84, 6.65, 5.85, 8.16, and 7.70, respectively. The RMSE of Well B of LSTM-SVR model is 0.94, while the RMSE of ANN, and LSTM were 1.48, and 2.32.

Keywords

Tight oil / Production forecast / LSTM-SVR / Residual correction

Acknowledgments

The authors gratefully acknowledge the financial support of National Natural Science Foundation of China (52274041 and 51974265), Sichuan science fund for distinguished Young Scholars (2023NSFSC1954), the Ministry of Science and Higher Education of the Russian Federation under Agreement No. 075-15-2022-299 within the framework of the development program for a worldclass Research Center “Efficient development of the global liquid hydrocarbon reserves”, Science and Technology Research Program of Chongqing Municipal Education Commission (KJQN202201510), Natural Science Foundation of Chongqing (CSTB2022NSCQMSX0403), Chongqing Municipal Support Program for Overseas Students Returning for Entrepreneurship and Innovation (2205012980950154), Scientific Research Funding Project of Chongqing University of Science and Technology (ckrc2021040). Thanks for the relevant data provided by Xinjiang oilfield.

References

[1] Y. Assef, A. Kantzas, P. Pereira Almao, Numerical modelling of cyclic CO2 injection in unconventional tight oil resources; trivial effects of heterogeneity and hysteresis in Bakken formation, Fuel 236 (2019) 1512-1528, https://doi.org/10.1016/j.fuel.2018.09.046.

[2] S. Li, L. Sun, L. Wang, Z. Li, et al., Hybrid CO2-N2 huff-n-puff strategy in unlocking tight oil reservoirs, Fuel 309 (2022) 122198, https://doi.org/10.1016/j.fuel.2021.122198.

[3] B. Wei, L. Wang, T. Song, M. Zhong, M.A. Varfolomeev, Enhanced oil recovery by low-salinity water spontaneous imbibition (LSW-SI) in a typical tight sandstone formation of mahu sag from core scale to field scale, Petroleum 7 (3) (2021) 272-281, https://doi.org/10.1016/j.petlm.2020.09.005.

[4] C. Zou, et al., Geological concepts, characteristics, resource potential and key techniques of unconventional hydrocarbon: on unconventional petroleum geology, Petrol. Explor. Dev. 40 (4) (2013) 385-399, https://doi.org/10.11698/PED.2013.04.01.

[5] L. Sun, C. Zou, A. Jia, et al., Development characteristics and orientation of tight oil and gas in China, Petrol. Explor. Dev. 46 (6) (2019) 1073-1087, https://doi.org/10.1016/S1876-3804(19)60264-8.

[6] X. Zhou, Q. Yuan, Y. Zhang, et al., Performance evaluation of CO 2 flooding process in tight oil reservoir via experimental and numerical simulation studies, Fuel 236 (2019) 730-746, https://doi.org/10.1016/j.fuel.2018.09.035.

[7] J.J. Arps, Analysis of decline curves, Transactions of the AIME 160 (1) (1945) 228-247, https://doi.org/10.1016/j.petrol.2010.05.007.

[8] R.D. Carter, Type curves for finite radial and linear gas-flow systems: constantterminal- pressure case, Soc. Petrol. Eng. J. 25 (5) (1985) 719-728, https://doi.org/10.2118/12917-PA.

[9] R.D. Hazlett, U. Farooq, D.K. Babu, A complement to decline curve analysis, SPE J. 26 (4) (2021) 2468-2478, https://doi.org/10.2118/205390-PA.

[10] K. Jongkittinarukorn, N. Last, F.H. Escobar, et al., A straight-line DCA for a gas reservoir, J. Petrol. Sci. Eng. 201 (2021) 108452, https://doi.org/10.1016/j.petrol.2021.108452.

[11] L. He, H. Mei, X. Hu, et al., Advanced flowing material balance to determine original gas in place of shale gas considering adsorption hysteresis, SPE Reservoir Eval. Eng. 22 (4) (2019) 1282-1292, https://doi.org/10.2118/195581-PA.

[12] I.S. Afanasyev, A.V. Timonov, I.V. Sudeev, et al., Analysis of multiple fractured horizontal wells application at Priobskoye field, SPE 162031, in:SPE Russian Oil and Gas Exploration and Production Technical Conference and Exhibition, 2012, https://doi.org/10.2118/162031-MS. Moscow, Russia.

[13] R. Velasco, P. Panja, M. Deo, Moving boundary approach to forecast tight oil production, AIChE J. 67 (2) (2021), e17012, https://doi.org/10.1002/aic.17012.

[14] Y. Wu, L. Cheng, L. Ma, S. Huang, et al., A transient two-phase flow model for production prediction of tight gas wells with fracturing fluid-induced formation damage, J. Petrol. Sci. Eng. 199 (2021) 108351, https://doi.org/10.1016/j.petrol.2021.108351.

[15] l Kuang, et al., Application and development trend of artificial intelligence in petroleum exploration and development, Petrol. Explor. Dev. 48 (1) (2021) 1-11, https://doi.org/10.11698/PED.2021.01.01.

[16] P. Gao, C. Jiang, Q. Huang, et al., Fluvial facies reservoir productivity prediction method based on principal component analysis and artificial neural network, Petroleum 2 (1) (2016) 49-53, https://doi.org/10.1016/j.petlm.2015.12.005.

[17] M.A. Ahmadi, Z. Chen, Comparison of machine learning methods for estimating permeability and porosity of oil reservoirs via petro-physical logs, Petroleum 5 (3) (2019) 271-284, https://doi.org/10.1016/j.petlm.2018.06.002.

[18] E. Martin, P. Wills, D. Hohl, et al., Using machine learning to predict production at a Peace River thermal EOR site, SPE-182696-MS, in:SPE Reservoir Simulation Conference, 2017, https://doi.org/10.2118/182696-MS. Montgomery, USA.

[19] C.I. Noshi, M.R. Eissa, R.M. Abdalla,An intelligent data driven approach for production prediction, OTC-29243-MS, in:Offshore Technology Conference, 2019, https://doi.org/10.4043/29243-MS. Houston, Texas, USA.

[20] Q. Cao, R. Banerjee, S. Gupta, et al., Data driven production forecasting using machine learning, SPE-180984-MS, in:SPE Argentina Exploration and Production of Unconventional Resources Symposium, 2016, https://doi.org/10.2118/180984-MS. Buenos Aires, Argentina.

[21] D. Han, S. Kwon, H. Son, J. Lee, Production forecasting for shale gas well in transient flow using machine learning and decline curve analysis, URTEC- 198198-MS, in:SPE Asia Pacific Unconventional Resources Technology Conference, 2019, https://doi.org/10.15530/AP-URTEC-2019-198198. Brisbane, Australia.

[22] D.N. Nnamdi, V.O. Adelaja, Dynamic Production forecasting using artificial neural networks customized to historical well key flow indicators, SPE-198756-MS, in:SPE Nigeria Annual International Conference and Exhibition, Lagos, Nigeria, 2019, https://doi.org/10.2118/198756-MS.

[23] L. Kubota, D. Reinert, Machine learning forecasts oil rate in Mature Onshore Field jointly driven by water and steam injection, SPE-196152-MS, in:SPE Annual Technical Conference and Exhibition, 2019, https://doi.org/10.2118/196152-MS. Calgary, Alberta, Canada.

[24] A.F. Maqui, X. Zhai, A. Suarez Negreira, et al., A comprehensive workflow for near real time waterflood management and production optimization using reduced-physics and data-driven technologies, SPE-185614-MS, in:SPE Latin America and Caribbean Petroleum Engineering Conference, 2017, https://doi.org/10.2118/185614-MS. Buenos Aires, Argentina.

[25] J. Sun, X. Ma, M. Kazi, Comparison of decline curve analysis DCA with recursive neural networks RNN for production forecast of multiple wells, SPE- 190104-MS, in:SPE Western Regional Meeting, 2018, https://doi.org/10.2118/190104-MS. Garden Grove, California, USA.

[26] K. Lee, J. Lim, D. Yoon, et al., Prediction of shale-gas production at Duvernay formation using deep-learning algorithm, SPE J. 24 (6) (2019) 2423-2437, https://doi.org/10.2118/195698-PA.

[27] j Gu, m Zhou, z Li, et al., Oil well production forecast with long-short term memory network model based on data mining, Special Oil Gas Reservoirs 26 (2) (2019) 77-81, https://doi.org/10.3969/j.issn.1006-6535.2019.02.013.

[28] h Wang, et al., Production prediction at ultra-high water cut stage via recurrent neural network, Petrol. Explor. Dev. 47 (5) (2020) 1009-1015, https://doi.org/10.11698/PED.2020.05.15.

[29] W. Liu, W.D. Liu, J. Gu, Forecasting oil production using ensemble empirical model decomposition based Long Short-Term Memory neural network, J. Petrol. Sci. Eng. 189 (2020) 107013, https://doi.org/10.1016/j.petrol.2020.107013.

[30] W. Jiang, A. Imin, X. Wang, et al., Geochemical characterization and quantitative identification of mixed-source oils from the baikouquan and lower wuerhe formations in the eastern slope of the mahu sag, junggar basin, NW China, J. Petrol. Sci. Eng. 191 (2020) 107175, https://doi.org/10.1016/j.petrol.2020.107175.

[31] S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Comput. 9 (8) (1997) 1735-1780, https://doi.org/10.1162/neco.1997.9.8.1735.

[32] M. Stundner, J.S. Al-Thuwaini, How data-driven modeling methods like neural networks can help to integrate different types of data into reservoir management, in: SPE Middle East Oil Show, Bahrain, 2001, https://doi.org/10.2118/68163-MS. SPE 68163.

[33] A. Ansari, M. Heras, J. Nones, M. Mohammadpoor, F. Torabi, Predicting the performance of steam assisted gravity drainage (SAGD) method utilizing artificial neural network (ANN), Petroleum 6 (4) (2020) 368-374, https://doi.org/10.1016/j.petlm.2019.04.001.

[34] I. Gupta, C. Rai, D. Devegowda, et al., Haynesville shale: predicting long-term production and residual analysis to identify well intereference and frac hits, SPE-195218-MS, in:SPE Oklahoma City Oil and Gas Symposium, 2019, https://doi.org/10.2118/195673-PA. Oklahoma City, Oklahoma, USA.

[35] D. Ilk, A.D. Perego, J.A. Rushing, et al., Integrating multiple production analysis techniques to assess tight gas sand reserves: defining a new paradigm for industry best practices, SPE 114947, in:CIPC/SPE Gas Technology Symposium 2008 Joint Conference, 2008, https://doi.org/10.2118/114947-MS. Calgary, Alberta, Canada.

[36] D. Ilk, J.A. Rushing, A.D. Perego, et al., Exponential vs. hyperbolic decline in tight gas sands: understanding the origin and implications for reserve estimates using Arps' decline curves, SPE 116731, in:SPE Annual Technical Conference and Exhibition, 2008, https://doi.org/10.2118/116731-MS. Denver, Colorado, USA.

[37] A.N. Duong, An unconventional rate decline approach for tight and fracturedominated gas wells, CSUG/SPE 137748, in:Canadian Unconventional Resources and International Petroleum Conference, 2010, https://doi.org/10.2118/137748-MS. Calgary, Alberta, Canada.

[38] A.N. Duong, Rate-decline analysis for fracture-dominated shale reservoirs: Part 2, SPE-171610-MS,in: SPE/CSUR Unconventional Resources Conference, 2014, https://doi.org/10.2118/137748-PA. Calgary, Alberta, Canada.

[39] D. Ilk, A.D. Perego, J.A. Rushing, et al., Integrating multiple production analysis techniques to assess tight gas sand reserves: defining a new paradigm for industry best practices, SPE 114947, in:CIPC/SPE Gas Technology Symposium 2008 Joint Conference, 2008, https://doi.org/10.2118/114947-MS. Calgary, Alberta, Canada.

[40] J. You, W. Ampomah, Q. Sun, Development and application of a machine learning based multi-objective optimization workflow for CO2-EOR projects, Fuel 264 (2020) 116758, https://doi.org/10.1016/j.fuel.2019.116758.