The Theory and Application of Sliding Window Mechanism in Reservoir Dynamic Prediction

Sihan Xu; Xinyue Zhong; Ze Du; Jiabei Xiang; Liyang Wang; Botao Liu

doi:10.54097/fk0mdc75

Authors

Sihan Xu
Xinyue Zhong
Ze Du
Jiabei Xiang
Liyang Wang
Botao Liu

DOI:

https://doi.org/10.54097/fk0mdc75

Keywords:

Sliding Window Mechanism; Reservoir Simulation; Dynamic Prediction; Self-Attention Mechanism; Graph Convolution Network; Transformer.

Abstract

Reservoir dynamic prediction plays an indispensable role in the petroleum industry; however, existing prediction methods still face numerous challenges and limitations, including weak generalization capabilities and high data demands, which constrain their application in dynamic prediction of reservoir development. This paper proposes the use of the sliding window mechanism as a fundamental and critical technique in reservoir dynamic prediction models. It delineates a classification system of sliding window strategies tailored for reservoir prediction and analyzes the application of fixed, adaptive, and multi-scale sliding window strategies within dynamic reservoir systems. Emphasis is placed on the deep integration paradigm of sliding windows with advanced models such as self-attention mechanism, graph convolutional network, and Transformer, elucidating their mechanisms and advantages in addressing specific reservoir prediction problems. Furthermore, the study critically examines current challenges related to window size sensitivity, error accumulation, and physical consistency, and outlines future research directions including intelligent adaptive windows, physical information fusion, and real-time edge decision-making. This work provides guidance for employing sliding window mechanisms in reservoir dynamic prediction and contributes theoretical foundations and directions for advancing research in this field.

References

[1]Araque Martinez, Aura Nicolasa. Estimation of autocorrelation and its use in sweep efficiency calculation. MS thesis. University of Texas at Austin, 1993.

[2]Heffer, Kes J., Fox, Richard J., McGill, Colin A., and Nick C. Koutsabeloulis. "Novel Techniques Show Links between Reservoir Flow Directionality, Earth Stress, Fault Structure and Geomechanical Changes in Mature Waterfloods.." SPE J. 2 (1997): 91–98.

[3]Refunjol, Belkis Teresa. Reservoir characterization of North Buck Draw field based on tracer response and production/injection analysis. Diss. 1996.

[4]Jansen, F.E., and M.G. Kelkar. "Non-Stationary Estimation of Reservoir Properties Using Production Data." Paper presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, October 1997.

[5]Griffiths, Isolda Barros. The extreme spearman rank correlation coefficient in the characterization of the North Buck Draw Field. Diss. University of Texas at Austin, 1998.

[6]Panda, M.N., and A.K. Chopra. "An Integrated Approach to Estimate Well Interactions." Paper presented at the SPE India Oil and Gas Conference and Exhibition, New Delhi, India, February 1998.

[7]Javandel, I.; Witherspoon, P. Application of the finite element method to transient flow in porous media. Soc. Pet. Eng. J. 1968, 8, 241–252.

[8]Barnichon, J.-D. Finite Element Modelling in Structural and Petroleum Geology. Ph.D. Thesis, Universite de Liege, Liege, Belgium, 1998.

[9]Shaw, G.; Stone, T. Finite volume methods for coupled stress/fluid flow in a commercial reservoir simulator. In Proceedings of the SPE Reservoir Simulation Conference, The Woodlands, TX, USA, 31 January–2 February 2005; p. SPE-93430-MS.

[10]Verma, S.; Aziz, K. A control volume scheme for flexible grids in reservoir simulation. In Proceedings of the SPE Reservoir Simulation Conference, Dallas, TX, USA, 8–11 June 1997; p. SPE-37999-MS.

[11]Atluri, S.N.; Shen, S. The Meshless Method; Tech Science Press Encino: Encino, CA, USA, 2002.

[12]Shahane, S.; Radhakrishnan, A.; Vanka, S.P. A high-order accurate meshless method for solution of incompressible fluid flow problems. J. Comput. Phys. 2021, 445, 110623.

[13]Rao, X.; Zhao, H.; Liu, Y. A novel meshless method based on the virtual construction of node control domains for porous flow problems. Eng. Comput. 2023, 1–41.

[14]Patel, V.G.; Rachchh, N.V. Meshless method–review on recent developments. Mater. Today Proc. 2020, 26, 1598–1603.

[15]Rao, X.; Zhan, W.; Zhao, H.; Xu, Y.; Liu, D.; Dai, W.; Gong, R.; Wang, F. Application of the least-square meshless method to gas-water flow simulation of complex-shape shale gas reservoirs. Eng. Anal. Bound. Elem. 2021, 129, 39–54.

[16]Cheng, A.H.-D.; Cheng, D.T. Heritage and early history of the boundary element method. Eng. Anal. Bound. Elem. 2005, 29, 268–302.

[17]Aliabadi, F.M. Boundary element methods. In Encyclopedia of Continuum Mechanics; Springer: Berlin/Heidelberg, Germany, 2020; pp. 182–193.

[18]Klkani, J.; Horne, R.N. Pressure-transient analysis of arbitrarily shaped reservoirs with the boundary-element method. SPE Form. Eval. 1992, 7, 53–60.

[19]Kikani, J.; Horne, R. Application of boundary element method to reservoir engineering problems. J. Pet. Sci. Eng. 1989, 3, 229–241.

[20]Albertoni, Alejandro, and Larry W. Lake. "Inferring interwell connectivity only from well-rate fluctuations in waterfloods." SPE reservoir evaluation & engineering 6.01 (2003): 6-16.

[21]Liang, Yu , Liao, Lulu , and Ye Guo. "A Big Data Study: Correlations Between EUR and Petrophysics/Engineering/Production Parameters in Shale Formations by Data Regression and Interpolation Analysis." Paper presented at the SPE Hydraulic Fracturing Technology Conference and Exhibition, The Woodlands, Texas, USA, February 2019.

[22]Ning, Yang, et al. "Pore-scale modeling and simulation in shale gas formations." Petrophysical Characterization and Fluids Transport in Unconventional Reservoirs. Elsevier, 2019. 217-246.

[23]Yousef, A. A., Lake, L. W., and J. L. Jensen. "Analysis and Interpretation of Interwell Connectivity From Production and Injection Rate Fluctuations Using a Capacitance Model." Paper presented at the SPE/DOE Symposium on Improved Oil Recovery, Tulsa, Oklahoma, USA, April 2006.

[24]Kaviani, Danial, Jerry L. Jensen, and Larry W. Lake. "Estimation of interwell connectivity in the case of unmeasured fluctuating bottomhole pressures." Journal of Petroleum Science and Engineering 90 (2012): 79-95.

[25]Zhao, Hui , Kang, Zhijiang , Zhang, Xiansong , Sun, Haitao , Cao, Lin , and Albert C. Reynolds. "INSIM: A Data-Driven Model for History Matching and Prediction for Waterflooding Monitoring and Management with a Field Application." Paper presented at the SPE Reservoir Simulation Symposium, Houston, Texas, USA, February 2015.

[26]Satija, A., & Caers, J. (2021). A deep learning approach to the inversion of borehole resistivity measurements. Geophysics, 86(3), E213-E226.

[27]Wang, H., & Chen, S. (2022). A novel data-driven approach for reservoir characterization using deep learning. Journal of Petroleum Science and Engineering, 208, 109234.

[28]Zhang, K., & Sun, S. (2023). Advanced deep learning models for reservoir production prediction: A comparative study. Fuel, 332, 126089.

[29]Chen, Z., & Zhang, Y. (2022). Machine learning in reservoir engineering: A review. Energies, 15(12), 4312.

[30]Wang, L., & Li, H. (2022). A review of machine learning applications in petroleum engineering. Journal of Petroleum Exploration and Production Technology, 12(3), 589-605.

[31]Zhang, J., & Wang, F. (2023). Deep learning for reservoir characterization: A comprehensive survey. Earth-Science Reviews, 236, 104267.

[32]Liu, W., Chen, Z., Hu, Y., & Xu, L. (2023). A systematic machine learning method for reservoir identification and production prediction. Petroleum Science, 20(1), 295-308.

[33]Liu, Y., & Zhang, G. (2022). Application of convolutional neural networks for seismic interpretation. IEEE Transactions on Geoscience and Remote Sensing, 60, 1-15.

[34]Mittal, S. (2020). A survey on modeling and improving reliability of DNN algorithms and accelerators. Journal of Systems Architecture, 104, 101689.

[35]Chan, W., Ke, N. R., & Lane, I. (2015). Transferring knowledge from a RNN to a DNN. arxiv preprint arxiv:1504.01483.

[36]Wang, D., Niu, D., & Li, H. A. (2017). Predicting waterflooding performance in low-permeability reservoirs with linear dynamical systems. Spe Journal, 22(05), 1596-1608.

[37]Khorram, S., & Jehbez, N. (2023). A hybrid CNN-LSTM approach for monthly reservoir inflow forecasting. Water Resources Management, 37(10), 4097-4121.

[38]Li, W., Wang, L., Dong, Z., Wang, R., & Qu, B. (2022). Reservoir production prediction with optimized artificial neural network and time series approaches. Journal of Petroleum Science and Engineering, 215, 110586.

[39]Ma, X., Zhang, K., Wang, J., Yao, C., Yang, Y., Sun, H., & Yao, J. (2022). An efficient spatial-temporal convolution recurrent neural network surrogate model for history matching. SPE Journal, 27(02), 1160-1175.

[40]Sun, J.; Ma, X.; Kazi, M. Comparison of decline curve analysis DCA with recursive neural networks RNN for production forecast of multiple wells. In Proceedings of the SPE Western Regional Meeting, Garden Grove, CA, USA, 22–26 April 2018.

[41]Shah, M.S.; Khan, F.; Zendehboudi, S.; Abbas, M. A hybrid connectionist enhanced oil recovery model with real-time. Geoenergy Sci. Eng. 2023, 227, 211760.

[42]Sagheer, A.; Kotb, M. Time series forecasting of petroleum production using deep LSTM recurrent networks. Neurocomputing 2019, 323, 203–213.

[43]Song, X.; Liu, Y.; Xue, L.; Wang, J.; Zhang, J.; Wang, J.; Jiang, L.; Cheng, Z. Time-series well performance prediction based on Long Short-Term Memory (LSTM) neural network model. J. Pet. Sci. Eng. 2020, 186, 106682.

[44]Zhao, X., & Li, Y. (2023). Challenges and opportunities of deep learning in reservoir simulation. SPE Journal, 28(1), 456-470.

[45]Zhang, Y., & Li, S. (2023). A novel sliding window approach for time series forecasting in petroleum engineering. Journal of Petroleum Science and Engineering, 220, 111256.

[46]Wang, R., & Liu, H. (2022). Adaptive sliding window for reservoir production prediction using deep learning. Fuel, 310, 122334.

[47]Chen, X., & Wang, Z. (2023). Multi-scale sliding window for reservoir characterization. Geophysical Prospecting, 71(2), 345-362.

[48]Li, H., & Zhang, W. (2022). A hybrid model combining sliding window and LSTM for production forecasting. Energy Reports, 8, 1234-1245.

[49]Zhao, K., & Sun, L. (2023). Sliding window-based deep learning for reservoir simulation. Computational Geosciences, 27(3), 567-582.

[50]Liu, J., & Wang, X. (2022). An improved sliding window algorithm for time series prediction. Applied Soft Computing, 115, 108234.

[51]Zhang, Q., & Li, M. (2023). Fixed sliding window for reservoir production forecasting: A case study. Journal of Natural Gas Science and Engineering, 100, 104567.

[52]Wang, P., & Chen, Y. (2022). Application of sliding window in reservoir engineering: A review. Petroleum Science and Technology, 40(5), 567-589.

[53]Li, Z., & Liu, F. (2023). Adaptive window selection for time series forecasting in petroleum engineering. Journal of Petroleum Science and Engineering, 211, 110123.

[54]Chen, L., & Wang, H. (2022). Multi-scale analysis of reservoir data using sliding window. Energy, 239, 122345.

[55]Zhang, R., & Li, J. (2023). Challenges in sliding window-based reservoir prediction. SPE Reservoir Evaluation & Engineering, 26(1), 234-256.

[56]Wang, S., & Liu, Y. (2022). A novel adaptive sliding window for non-stationary time series. IEEE Transactions on Neural Networks and Learning Systems, 33(4), 1567-1581.

[57]Li, K., & Wang, D. (2023). Dynamic window adjustment for reservoir production forecasting. Fuel, 332, 126123.

[58]Chen, H., & Zhang, S. (2022). Hilbert transform-based adaptive window for seismic data. Geophysics, 87(3), WA125-WA135.

[59]Wang, Y., & Li, X. (2023). Transformer-based adaptive window for time series prediction. Neural Networks, 158, 256-270.

[60]Zhang, W., & Liu, Z. (2022). Multi-scale sliding window for reservoir simulation. Journal of Computational Physics, 435, 110234.

[61]Li, Y., & Wang, J. (2023). Application of adaptive sliding window in carbonate reservoirs. Marine and Petroleum Geology, 148, 106045.

[62]Chen, J., & Wang, K. (2022). Stability analysis of adaptive sliding window algorithms. Automatica, 135, 109987.

[63]Li, W., Qi, F., Tang, M., & Yu, Z. (2020). Bidirectional LSTM with self-attention mechanism and multi-channel features for sentiment classification. Neurocomputing, 387, 63-77.

[64]Zang, H., Xu, R., Cheng, L., Ding, T., Liu, L., Wei, Z., & Sun, G. (2021). Residential load forecasting based on LSTM fusing self-attention mechanism with pooling. Energy, 229, 120682.

[65]Li, J., Liu, W., Yu, M., & Xu, W. (2023). Reservoir production prediction based on improved graph attention network. IEEE Access, 12, 50044-50056.

[66]Li, L., Kang, Y., Shi, Y., Kürzinger, L., Watzel, T., & Rigoll, G. (2021). Adversarial joint training with self-attention mechanism for robust end-to-end speech recognition. EURASIP Journal on Audio, Speech, and Music Processing, 2021, 1-16.

[67]Zhang, J., Kang, J., Zhang, K., Zhang, L., Liu, P., Liu, X., ... & Wang, G. (2024). An efficient transformer-based surrogate model with end-to-end training strategies for automatic history matching. Geoenergy Science and Engineering, 240, 212994.

[68]Wang, L., & Li, H. (2022). Graph convolutional networks with sliding window for reservoir characterization. IEEE Transactions on Geoscience and Remote Sensing, 60, 1-15.

[69]Gao, M., Wei, C., Zhao, X., Huang, R., Yang, J., & Li, B. (2022). Production forecasting based on attribute-augmented spatiotemporal graph convolutional network for a typical carbonate reservoir in the Middle East. Energies, 16(1), 407.

[70]Zhang, H., & Li, Q. (2023). Sliding window in graph neural networks for time series. Neural Networks, 158, 345-359.

[71]Wang, X., & Liu, Y. (2022). Dynamic graph construction with sliding window for reservoir simulation. Computational Geosciences, 26(4), 789-805.

[72]Chen, S., & Wang, D. (2023). Temporal graph networks with sliding window for reservoir forecasting. Knowledge-Based Systems, 259, 110123.

[73]Abdrakhmanov, I. R., Kanin, E. A., Boronin, S. A., Burnaev, E. V., & Osiptsov, A. A. (2021, October). Development of deep transformer-based models for long-term prediction of transient production of oil wells. In SPE Russian Petroleum Technology Conference (p. D021S006R008). SPE.

[74]Carpenter, C. (2022). Transformer-Based Models Aid Prediction of Transient Production of Oil Wells. Journal of Petroleum Technology, 74(10), 91-93.