This paper proposes a novel intelligent method for defining and solving the reservoir performance prediction problem within a manifold space, fully considering geological uncertainty and the dynamic characteristics of reservoirs under time-varying well control conditions, creating a surrogate model for reservoir performance prediction based on Conditional Evolutionary Generative Adversarial Networks (CE-GAN). The CE-GAN leverages conditional evolution in the feature space to direct the evolution of the generative network in previously uncontrollable directions, and transforms the problem of reservoir performance prediction into an image evolution problem based on permeability distribution, initial reservoir performance and time-varying well control, thereby enabling fast and accurate reservoir performance prediction under time-varying well control conditions. The experimental results in basic (egg model) and actual water-flooding reservoirs show that the model predictions align well with numerical simulations. The median of relative residuals of pressure and oil saturation for the basic reservoir model are 0.5% and 9.0%, respectively, while those for the actual reservoir model are both 4.0%. Regarding time efficiency, the surrogate model after training achieves approximately 160-fold and 280-fold increases in computational speed for the basic and actual reservoir models, respectively, compared with traditional numerical simulations. The reservoir performance prediction surrogate model based on the CE-GAN can effectively enhance the efficiency of production optimization.
LIYang, WUShenghe, HOUJiagen, et al. Progress and prospects of reservoir development geology[J]. Petroleum Exploration and Development, 2017, 44(4): 569-579.
ZHULihong, DUQinglong, JIANGXueyan, et al. Characteristics and strategies of three major contradictions for continental facies multi-layered sandstone reservoir at ultra-high water cut stage[J]. Acta Petrolei Sinica, 2015, 36(2): 210-216.
LIUHe, LIYanchun, JIADeli, et al. Application status and prospects of artificial intelligence in the refinement of waterflooding development program[J]. Acta Petrolei Sinica, 2023, 44(9): 1574-1586.
ZHOUR N, ZHANGD, WEIJ G. Experimental investigation on remaining oil distribution and recovery performances after different flooding methods[J]. Fuel, 2022, 322: 124219.
[5]
WANGS, XIANGJ, WANGX. A deep learning based surrogate model for reservoir dynamic performance prediction[J]. Geoenergy Science and Engineering, 2024, 233: 212516.
[6]
ANZ B, ZHOUK, HOUJ, et al. Accelerating reservoir production optimization by combining reservoir engineering method with particle swarm optimization algorithm[J]. Journal of Petroleum Science and Engineering, 2022, 208 (Part D): 109692.
[7]
GLADKOVA, SAKHIBGAREEVR, SALIMOVD, et al. Application of CRM for production and remaining oil reserves reservoir allocation in mature west Siberian waterflood field[R]. SPE 187841-MS, 2017.
[8]
AL-FADHLIW, KURMAR, KOVYAZIND, et al. Modeling and simulation to produce thin layers of remaining oil using downhole water sink technique for improved oil recovery. A case study in Greater Burgan field[R]. SPE 194839-MS, 2019.
ZHANGKai, ZHAOXinggang, ZHANGLiming, et al. Current status and prospect for the research and application of big data and intelligent optimization methods in oilfield development[J]. Journal of China University of Petroleum (Edition of Natural Science), 2020, 44(4): 28-38.
SUJian, LIUHe. Challenge and development of big data application in petroleum engineering[J]. Journal of China University of Petroleum (Edition of Social Sciences), 2020, 36(3): 1-6.
KUANGLichun, LIUHe, RENYili, et al. Application and development trend of artificial intelligence in petroleum exploration and development[J]. Petroleum Exploration and Development, 2021, 48(1): 1-11.
SONGLaiming, WANGChunqiu, LUChuan, et al. Research progress of data-driven injection production optimization of complex oil reservoirs[J]. Oil Drilling & Production Technology, 2022, 44(2): 253-260.
LIXiaobo, LIXin, YANLin, et al. A 3D attention U-Net network and its application in geological model parameterization[J]. Petroleum Exploration and Development, 2023, 50(1): 167-173.
SUNQifeng, LINa, DUANYouxiang, et al. Logging-while-drilling formation dip interpretation based on long short-term memory[J]. Petroleum Exploration and Development, 2021, 48(4): 843-850.
[15]
ZHANGH L, YUH, MENGS W, et al. Fast and accurate reconstruction of large-scale 3D porous media using deep learning[J]. Journal of Petroleum Science and Engineering, 2022, 217: 110937.
[16]
WUH, LUBBERSN, VISWANATHANH S, et al. A multi-dimensional parametric study of variability in multi-phase flow dynamics during geologic CO2 sequestration accelerated with machine learning[J]. Applied Energy, 2021, 287: 116580.
[17]
MOS X, ZABARASN, SHIX Q, et al. Deep autoregressive neural networks for high-dimensional inverse problems in groundwater contaminant source identification[J]. Water Resources Research, 2019, 55(5): 3856-3881.
ZHANGRui, JIAHu. Production performance forecasting method based on multivariate time series and vector autoregressive machine learning model for waterflooding reservoirs[J]. Petroleum Exploration and Development, 2021, 48(1): 175-184.
ZHANGLei, DOUHongen, WANGTianzhi, et al. A production prediction method of single well in water flooding oilfield based on integrated temporal convolutional network model[J]. Petroleum Exploration and Development, 2022, 49(5): 996-1004.
[20]
TANGJ Z, FANB, XIAOL Z, et al. A new ensemble machine-learning framework for searching sweet spots in shale reservoirs[J]. SPE Journal, 2021, 26(1): 482-497.
[21]
LIUY Y, MAX H, ZHANGX W, et al. A deep-learning-based prediction method of the estimated ultimate recovery (EUR) of shale gas wells[J]. Petroleum Science, 2021, 18(5): 1450-1464.
[22]
HUANGM C, XUH Y, YUH, et al. Fast prediction of methane adsorption in shale nanopores using kinetic theory and machine learning algorithm[J]. Chemical Engineering Journal, 2022, 446(Part 3): 137221.
[23]
GOODFELLOWI, BENGIOY, COURVILLEA. Deep learning[M]. Cambridge: The MIT Press, 2016.
[24]
XUEL, LIUY T, NANT C, et al. An efficient automatic history matching method through the probabilistic collocation based particle filter for shale gas reservoir[J]. Journal of Petroleum Science and Engineering, 2020, 190: 107086.
[25]
LIUP Y, ZHANGK, YAOJ. Reservoir automatic history matching: Methods, challenges, and future directions[J]. Advances in Geo-Energy Research, 2023, 7(2): 136-140.
JIADeli, LIUHe, ZHANGJiqun, et al. Data-driven optimization for fine water injection in a mature oil field[J]. Petroleum Exploration and Development, 2020, 47(3): 629-636.
[27]
KIMY D, DURLOFSKYL J. A recurrent neural network-based proxy model for well-control optimization with nonlinear output constraints[J]. SPE Journal, 2021, 26(4): 1837-1857.
[28]
ZHANGK, WANGZ Z, CHENG D, et al. Training effective deep reinforcement learning agents for real-time life-cycle production optimization[J]. Journal of Petroleum Science and Engineering, 2022, 208 (Part E): 109766.
[29]
SABERALIB, GOLSANAMIN, ZHANGK, et al. Simulating dynamics of pressure and fluid saturation at grid-scale by a deep learning-based surrogate reservoir modeling based on a fast-supply hybrid database and developing preliminary insights for future gas hydrate exploitations in China[J]. Geoenergy Science and Engineering, 2023, 222: 211415.
SHAHKARAMIA, MOHAGHEGHS. Applications of smart proxies for subsurface modeling[J]. Petroleum Exploration and Development, 2020, 47(2): 372-382.
[31]
ZHUY H, ZABARASN. Bayesian deep convolutional encoder-decoder networks for surrogate modeling and uncertainty quantification[J]. Journal of Computational Physics, 2018, 366: 415-447.
[32]
TRIPATHYR K, BILIONISI. Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification[J]. Journal of Computational Physics, 2018, 375: 565-588.
[33]
LALOYE, HÉRAULTR, JACQUESD, et al. Training-image based geostatistical inversion using a spatial generative adversarial neural network[J]. Water Resources Research, 2018, 54(1): 381-406.
[34]
WANGN Z, ZHANGD X, CHANGH B, et al. Deep learning of subsurface flow via theory-guided neural network[J]. Journal of Hydrology, 2020, 584: 124700.
[35]
ZHONGZ, SUNA Y, WANGY Y, et al. Predicting field production rates for waterflooding using a machine learning-based proxy model[J]. Journal of Petroleum Science and Engineering, 2020, 194: 107574.
[36]
ZHONGZ, SUNA Y, RENB, et al. A deep-learning-based approach for reservoir production forecast under uncertainty[J]. SPE Journal, 2021, 26(3): 1314-1340.
[37]
MAX P, ZHANGK, WANGJ, et al. An efficient spatial-temporal convolution recurrent neural network surrogate model for history matching[J]. SPE Journal, 2022, 27(2): 1160-1175.
[38]
MAX P, ZHANGK, ZHANGJ D, et al. A novel hybrid recurrent convolutional network for surrogate modeling of history matching and uncertainty quantification[J]. Journal of Petroleum Science and Engineering, 2022, 210: 110109.
[39]
JINZ L, LIUY M, DURLOFSKYL J. Deep-learning-based surrogate model for reservoir simulation with time-varying well controls[J]. Journal of Petroleum Science and Engineering, 2020, 192: 107273.
[40]
WEIC J, HUANGR J, DINGM M, et al. Characterization of saturation and pressure distribution based on deep learning for a typical carbonate reservoir in the Middle East[J]. Journal of Petroleum Science and Engineering, 2022, 213: 110442.
[41]
ZHANGK, WANGX Y, MAX P, et al. The prediction of reservoir production based proxy model considering spatial data and vector data[J]. Journal of Petroleum Science and Engineering, 2022, 208 (Part D): 109694.
[42]
HUANGH, GONGB, LIUY M, et al. Deep-learning based surrogate modeling for fast and accurate simulation in realistic 3D reservoir with varying well controls[J]. Geoenergy Science and Engineering, 2023, 222: 211418.
[43]
SUNN Z, SUNA. Model calibration and parameter estimation: For environmental and water resource systems[M]. New York: Springer, 2015.
[44]
PEACEMAND W. Representation of a horizontal well in numerical reservoir simulation[J]. SPE Advanced Technology Series, 1993, 1(1): 7-16.
[45]
JANSENJ D, FONSECAR M, KAHROBAEIS, et al. The egg model: A geological ensemble for reservoir simulation[J]. Geoscience Data Journal, 2014, 1(2): 192-195.
[46]
SUNA Y, SCANLONB R, ZHANGZ Z, et al. Combining physically based modeling and deep learning for fusing GRACE satellite data: Can we learn from mismatch?[J]. Water Resources Research, 2019, 55(2): 1179-1195.
[47]
SUNA Y. Optimal carbon storage reservoir management through deep reinforcement learning[J]. Applied Energy, 2020, 278: 115660.