Data-Driven End-To-End Production Prediction of Oil Reservoirs by EnKF-Enhanced Recurrent Neural Networks
- Anqi Bao (Texas A&M University) | Eduardo Gildin (Texas A&M University) | Jianhua Huang (Texas A&M University) | Emilio Jose Rocha Coutinho (Petrobras and Texas A&M University)
- Document ID
- Society of Petroleum Engineers
- SPE Latin American and Caribbean Petroleum Engineering Conference, 27-31 July, Virtual
- Publication Date
- Document Type
- Conference Paper
- 2020. Society of Petroleum Engineers
- 5.1.5 Geologic Modeling, 5.6 Formation Evaluation & Management, 7.6 Information Management and Systems, 5.1 Reservoir Characterisation, 5.5 Reservoir Simulation, 6.1.5 Human Resources, Competence and Training, 6.1 HSSE & Social Responsibility Management, 6 Health, Safety, Security, Environment and Social Responsibility, 5 Reservoir Desciption & Dynamics, 7 Management and Information, 5.4.1 Waterflooding, 5.3.2 Multiphase Flow, 7.6.6 Artificial Intelligence, 5.3 Reservoir Fluid Dynamics, 5.4 Improved and Enhanced Recovery, 7.6.7 Neural Networks, 5.6.9 Production Forecasting
- Production Prediction, Proxy Model, Recurrent Neural Network, EnKF
- 78 in the last 30 days
- 78 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 28.00|
Large data volumes and complex physical processes coupling multiphase flow, numerics and production assessment, as in the case of reservoir simulation, are difficult to analyze rapidly but are needed to guide operators for designing production strategies and subsequent reservoir management. The traditional reservoir simulation process is time consuming and alternatives such as data-driven proxy modeling can overcome the computation complexity drawbacks. A machine learning technique called recurrent neural network (RNN) has been proved useful for reservoir modeling with sequence data. In this paper, we develop a novel end-to-end production prediction workflow that can be used to fast guide reservoir development when production begins.
In this work, we apply RNN on analyzing control parameter data and synthetic historical production data for better reservoir characterization and production prediction. More specifically, we would like to build a model to directly link the control parameters (flow rate and bottom hole pressure) with desired production output, e.g. production rate and water cut. One immediate benefit of the model is to avoid the state variable calculation (pressure/saturation). In addition, as this is a data-driven end-to-end production prediction model, it will not require the numerical iteration and gradient calculation once the training is completed. We explore two types of RNN based structure: cascaded LSTM and Ensemble Kalman filter (EnKF) enhanced LSTM. The LSTM (long short-term memory) is used to compensate the weakness of standard RNN for preserving long time information dependencies. The structure of LSTM takes into account the memory of previous calculation when modeling the current response. Our cascaded LSTM is an improvement to regular LSTM as it incorporates physical quantities of interest such as water breakthrough. The model is conceived with two consecutive networks: one network for breakthrough time estimation with output being fed into a second network that reconciles other features important for oil production prediction. The EnKF enhanced LSTM has the capability of performing data assimilation based on real time production data, thus providing a way to update our model constantly.
In this work, we first show the methodology applied to the two-phase water flooding reservoir with five spot production scenarios. Then we conduct the comparison of Bayesian optimization tuned cascaded LSTM vs. standard LSTM. Finally, we showcase the usefulness of Ensemble Kalman filter in improving and updating current model.
The method presented in this paper uses RNN (specifically cascaded LSTM) to learn the pattern from sequence data and identify the reservoir simulation proxy model, which can accurately predict surface production rate and water cut without the state variable calculation. The study also shows improved accuracy (over standard methods) for EnKF trained RNN and its capability of updating flow rate prediction based on new observation data.
|File Size||1 MB||Number of Pages||21|
Aarnes, J., Krogstad, S., & Lie, K. (2006). A Hierarchical Multiscale Method for Two-Phase Flow Based upon Mixed Finite Elements and Nonuniform Coarse Grids. Multiscale Modeling & Simulation, 5(2), 337-363.doi:10.1137/050634566
Burgers, G., Jan van Leeuwen, P., & Evensen, G. (1998). Analysis Scheme in the Ensemble Kalman Filter. Monthly Weather Review, 126(6), 1719-1724.doi:10.1175/1520-0493(1998)126<1719:ASITEK>2.0.CO;2
Canchumuni, S. A., Emerick, A. A., & Pacheco, M. A. (2017). Integration of Ensemble Data Assimilation and Deep Learning for History Matching Facies Models. Paper presented at the OTC Brasil, Rio de Janeiro, Brazil.doi:10.4043/28015-MS.
Chen, H., Onishi, T., Olalotiti-Lawal, F., & Datta-Gupta, A. (2018). Streamline Tracing and Applications in Naturally Fractured Reservoirs Using Embedded Discrete Fracture Models. Paper presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, USA.doi:10.2118/191475-MS.
Cheng, H., Kharghoria, A., Zhong, H., & Datta-Gupta, A. (2004). Fast History Matching of Finite-Difference Models Using Streamline-Derived Sensitivities. Paper presented at the SPE/DOE Symposium on Improved Oil Recovery, Tulsa, Oklahoma.doi:10.2118/89447-MS.
Engle, R. F. (1982). Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation. Econometrica, 50(4), 987-1007.doi:10.2307/1912773
Evensen, G. (1994). Sequential data assimilation with a nonlinear quasi-geostrophic model using Monte Carlo methods to forecast error statistics. Journal of Geophysical Research: Oceans, 99(C5), 10143-10162.doi:10.1029/94JC00572
Ho, S. L., & Xie, M. (1998). The use of ARIMA models for reliability forecasting and analysis. Computers & Industrial Engineering, 35(1), 213-216.doi:10.1016/S0360-8352(98)00066-7
Hochreiter, S., & Schmidhuber, J. (1997). Long Short-Term Memory. Neural Comput., 9(8), 1735-1780.doi:10.1162/neco.19184.108.40.2065
Houtekamer, P. L., & Mitchell, H. L. (1998). Data Assimilation Using an Ensemble Kalman Filter Technique. Monthly Weather Review, 126(3), 796-811.doi:10.1175/1520-0493(1998)126<0796:DAUAEK>2.0.CO;2
Kemajou, V. N., Bao, A., & Germain, O. (2019). Wellbore Schematics to Structured Data Using Artificial Intelligence Tools. Paper presented at the Offshore Technology Conference, Houston, Texas.doi:10.4043/29490-MS.
Liu, Y., & Horne, R. N. (2012). Interpreting Pressure and Flow-Rate Data From Permanent Downhole Gauges by Use of Data-Mining Approaches. SPE Journal, 18(01), 69-82.doi:10.2118/147298-PA
Ma, X., & Liu, Z. (2018). Predicting the oil production using the novel multivariate nonlinear model based on Arps decline model and kernel method. Neural Computing and Applications, 29(2), 579-591.doi:10.1007/s00521-016-2721-x
Madasu, S., & Rangarajan, K. P. (2018). Deep Recurrent Neural Network DRNN Model for Real-Time Multistage Pumping Data. Paper presented at the OTC Arctic Technology Conference, Houston, Texas, USA.doi:10.4043/29145-MS.
Ng, A. (2018). Sequence Models. Retrieved from https://www.coursera.org/learn/nlp-sequence-models/home/welcome
Pan, Y., Bi, R., Zhou, P., Deng, L., & Lee, J. (2019). An Effective Physics-Based Deep Learning Model for Enhancing Production Surveillance and Analysis in Unconventional Reservoirs. Paper presented at the SPE/AAPG/SEG Unconventional Resources Technology Conference, Denver, Colorado, USA.doi:10.15530/urtec-2019-145.
Pan, Y., Deng, L., & Lee, J. (2019). Data-Driven Deconvolution Using Echo-State Networks Enhances Production Data Analysis in Unconventional Reservoirs. Paper presented at the SPE Eastern Regional Meeting, Charleston, West Virginia, USA.doi:10.2118/196598-MS.
Pan, Y., Zhou, P., Deng, L., & Lee, J. (2019). Production Analysis and Forecasting for Unconventional Reservoirs Using Laplacian Echo-State Networks. Paper presented at the SPE Western Regional Meeting, San Jose, California, USA.doi:10.2118/195243-MS.
Sun, Z., Shi, J., Wu, K., Zhang, T., Feng, D., & Li, X. (2019). Effect of Pressure-Propagation Behavior on Production Performance: Implication for Advancing Low-Permeability Coalbed-Methane Recovery. SPE Journal, 24(02), 681-697.doi:10.2118/194021-PA
Tang, J., & Wu, K. (2018). A 3-D model for simulation of weak interface slippage for fracture height containment in shale reservoirs. International Journal of Solids and Structures, 144-145, 248-264.doi:10.1016/j.ijsolstr.2018.05.007
Tian, C., & Horne, R. N. (2017). Recurrent Neural Networks for Permanent Downhole Gauge Data Analysis. Paper presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA.doi:10.2118/187181-MS.
Wu, Y., Cheng, L., Huang, S., Fang, S., Killough, J. E., Jia, P., & Wang, S. (2019). A Transient Two-Phase Flow Model for Production Prediction of Tight Gas Wells with Fracturing Fluid-Induced Formation Damage. Paper presented at the SPE Western Regional Meeting, San Jose, California, USA.doi:10.2118/195327-MS.
Xiong, H., Huang, S., Devegowda, D., Liu, H., Li, H., & Padgett, Z. (2019). Influence of Pressure Difference Between Reservoir and Production Well on Steam-Chamber Propagation and Reservoir-Production Performance. SPE Journal, 24(02), 452-476.doi:10.2118/190107-PA
Zarate-Losoya, E., Cunningham, T., El-Sayed, I., Noynaert, S. F., & Florence, F. (2018). Lab-Scale Drilling Rig Autonomously Mitigates Downhole Dysfunctions and Geohazards Through Bit Design, Control System and Machine Learning. Paper presented at the IADC/SPE Drilling Conference and Exhibition, Fort Worth, Texas, USA.doi:10.2118/189630-MS.
Zhang, D., Chen, Y., & Meng, J. (2018). Synthetic well logs generation via Recurrent Neural Networks. Petroleum Exploration and Development, 45(4), 629-639.doi:10.1016/S1876-3804(18)30068-5
Zhang, F., Saputra, I. W. R., Niu, G., Adel, I. A., Xu, L., & Schechter, D. S. (2018). Upscaling Laboratory Result of Surfactant-Assisted Spontaneous Imbibition to the Field Scale through Scaling Group Analysis, Numerical Simulation, and Discrete Fracture Network Model. Paper presented at the SPE Improved Oil Recovery Conference, Tulsa, Oklahoma, USA.doi:10.2118/190155-MS.
Zhang, T., Li, X., Li, J., Feng, D., Li, P., Zhang, Z., … Wang, S. (2017). Numerical investigation of the well shut-in and fracture uncertainty on fluid-loss and production performance in gas-shale reservoirs. Journal of Natural Gas Science and Engineering, 46, 421-435.doi:10.1016/j.jngse.2017.08.024
Zhao, Y., & Forouzanfar, F. (2017). A Simultaneous Bayesian Estimation of Channelized Facies and Reservoir Properties under Prior Uncertainty. Paper presented at the SPE Europec featured at 79th EAGE Conference and Exhibition, Paris, France.doi:10.2118/185800-MS.
Zhao, Y., Forouzanfar, F., & Reynolds, A. C. (2017). History matching of multi-facies channelized reservoirs using ES-MDA with common basis DCT. Computational Geosciences, 21(5), 1343-1364.doi:10.1007/s10596-016-9604-1
Zhou, P., Pan, Y., Sang, H., & Lee, W. J. (2018). Criteria for Proper Production Decline Models and Algorithm for Decline Curve Parameter Inference. In Unconventional Resources Technology Conference, Houston, Texas, 23-25 July 2018 (pp. 3535-3551): Society of Exploration Geophysicists, American Association of Petroleum Geologists, Society of Petroleum Engineers.