Simplified Dynamic Modeling of Faulted Turbidite Reservoirs: A Deep-Learning Approach to Recovery-Factor Forecasting for Exploration
- Faruk O. Alpak (Shell International Exploration and Production Inc.) | Mauricio Araya-Polo (Shell International Exploration and Production Inc.) | Kachi Onyeagoro (Shell International Exploration and Production Inc.)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Evaluation & Engineering
- Publication Date
- November 2019
- Document Type
- Journal Paper
- 1,240 - 1,255
- 2019.Society of Petroleum Engineers
- faults, structural uncertainty, deep learning, recovery factor, turbidite reservoirs
- 9 in the last 30 days
- 238 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
Faults are geological features that are essential in considering the development of hydrocarbon reservoirs. Significant resources (appraisal costs and manpower) are often deployed to locate them and assess their connectivity more accurately so that adequate investment decisions can be made. Early in a project’s life, data might be sparse or the time that can be dedicated to data processing might be limited. An estimate of the impact of faults can be derived by considering the impact of analogous fault patterns in similar reservoir environments.
A significant number of discoveries consist of channelized turbidite reservoirs draped over deepwater toe-thrust anticlines. To understand the effects of various fault patterns on the recovery factors of structurally complex turbidite reservoirs, we first perform extensive flow-simulation-based multidimensional sensitivity studies using realistic channelized stratigraphic architectures. Many of these reservoirs contain high-porosity, high-permeability channels, light oil, limited aquifer drive, and a limited number of fault populations. To constrain the potential impacts of the faults, a generic reservoir model was constructed and simulated with varying reservoir and fault properties. Because the input parameters are known (e.g., fault-zone thickness and permeability, shale-drape coverage, and oil viscosity), the relative importance of each variable can be quantified. The simulation results show that the impacts of the faults on reservoir performance vary systematically with the fault length and orientation, the undeformed-reservoir permeability, and the fault permeability. The recovery factors are relatively insensitive to fault-zone thickness, net/gross ratio, and porosity. The time-delay effect of the faults is significant in reservoirs with a permeability of approximately 100 md but not in darcy-quality reservoirs. In most fault scenarios, the recovery factors from simulations with high-permeability faults range from approximately 30% in 100-md reservoirs to approximately 45% in 3,000-md reservoirs, equivalent to 75 to 96% of the recovery factors of the unfaulted reservoirs. In comparison, assuming the most likely fault permeabilities, the recovery factors decrease to circa 20% in 100-md reservoirs and circa 41% in 3,000-md reservoirs, equivalent to 40 to 85% of the values in unfaulted reservoirs.
To learn and predict the effects of faults on the recovery factor in a very fast and reasonably generic fashion in deepwater toe-thrust anticlines, we have developed an accurate deep-learning (DL) based surrogate model that predicts the structure connectivity factors as a function of recovery time. These factors are then used for discounting the recovery factors computed by simpler and easier-to-construct unfaulted models. The surrogate model is a very fast and advanced proxy for the multidimensional structure connectivity factor function and operates without resorting to any new flow simulation with a faulted model. We describe the novel DL architecture used for constructing the surrogate model and quantitatively prove its effectiveness. We finally demonstrate an example real-life application of the surrogate model on an exploration-stage prospect-ranking exercise.
|File Size||1 MB||Number of Pages||16|
Abadi, M., Barham, P., Chen, J. et al. 2016. TensorFlow: A System for Largescale Machine Learning. Presented at the 12th USENIX Conference on Operating Systems Design and Implementation, OSDI’16. USENIX Association, Berkeley, California, 265–283.
Adler, A., Boublil, D., and Zibulevsky, M. 2017. Block-Based Compressed Sensing of Images Via Deep Learning. Presented at the IEEE 19th International Workshop on Multimedia Signal Processing (MMSP), Luton, UK, 16–18 October, 1–6. https://doi.org/10.1109/MMSP.2017.8122281.
Alpak, F. O. 2015. Quasiglobal Multiphase Upscaling of Reservoir Models With Nonlocal Stratigraphic Heterogeneities. SPE J. 20 (2): 277–293. SPE-170245-PA. https://doi.org/10.2118/170245-PA.
Alpak, F. O. and Barton, M. D. 2014. Dynamic Impact and Flow-Based Upscaling of the Estuarine Point-Bar Stratigraphic Architecture. J Pet Sci Eng 120: 18–38. https://doi.org/10.1016/j.petrol.2014.05.020.
Alpak, F. O., Barton, M. D., and Naruk, S. 2013. The Impact of Fine-Scale Turbidite Channel Architecture on Deepwater Reservoir Performance. AAPG Bull 97 (2): 251–284. https://doi.org/10.1306/04021211067.
Alpak, F. O., Barton, M. D., van der Vlugt, F. F. et al. 2010. Simplified Modeling of Turbidite Channel Reservoirs. SPE J. 15 (2): 480–494. SPE-114854-PA. https://doi.org/10.2118/114854-PA.
Alpak, F. O., Jin, L., and Ramirez, B. A. 2016. Robust Optimization of Well-Placement in Geologically Complex Reservoirs. Int J Pet Eng 2 (4): 247–264. SPE-175106-MS. https://doi.org/10.2118/175106-MS.
Alpak, F. O. and van der Vlugt, F. F. 2014. Shale-Drape Modeling for the Geologically Consistent Simulation of Clastic Reservoirs. SPE J. 19 (5): 832–844. SPE-169820-PA. https://doi.org/10.2118/169820-PA.
Araya-Polo, M., Dahlke, T., Frogner, C. et al. 2017. Automated Fault Detection Without Seismic Processing. Leading Edge 36 (3): 194–280. https://doi.org/10.1190/tle36030208.1.
Araya-Polo, M., Jennings, J., Adler, A. et al. 2018. Deep-Learning Tomography. Leading Edge 37 (1): 58–66. https://doi.org/10.1190/tle37010058.1.
Barton, M., O’Byrne, C., Pirmez, C. et al. 2010. Turbidite Channel Architecture: Recognizing and Quantifying the Distribution of Channel Base Drapes Using Core and Dipmeter Data. In Dipmeter and Borehole Image Log Technology, AAPG Memoir 92, ed. M. Pöppelreiter, C. García-Carballido, and M. A. Kraaijveld, 195–211. Tulsa, Oklahoma, USA: AAPG Publications.
Blenkinsop, T. G. 1989. Thickness—Displacement Relationships for Deformation Zones: Discussion. Journal of Structural Geology 11 (8): 1051–1053. https://doi.org/10.1016/0191-8141(89)90056-4.
C&C Digital Reservoir Database, www.ccreservoirs.com (accessed on 19 April 2019).
Chetlur, S., Woolley, C., Vandermersch, P. et al. 2014. cuDNN: Efficient Primitives for Deep Learning. arXiv:1410.0759 [cs.NE].
Dickey, P. A. 1992. La Cira-Infantas Field-Colombia, Middle Magdalena Basin. In Structural Traps VII: AAPG Treatise of Petroleum Geology, Atlas of Oil and Gas Fields, ed. E. A. Beaumont and N. H. Foster, 323–347. Tulsa, Oklahoma, USA: AAPG Publications.
Edwards, E. B. 1998. Redevelopment of Carpinteria Field, Santa Barbara Channel, California. In Structure and Petroleum Geology, Santa Barbara Channel, California, Miscellaneous Publication 46, ed. D. S. Kunitomi, T. E. Hopps, and J. M. Galloway, 239–246. Bakersfield, California, USA: AAPG Publications, Pacific Section. https://doi.org/10.32375/1998-MP46.20.
Foxford, K. A., Walsh, J. J., Watterson, J. et al. 1998. Structure and Content of Moab Fault Zone, Utah, USA, and Its Implication for Fault Seal Prediction. Geological Society, London, Special Publications 147 (1): 87–103. https://doi.org/10.1144/GSL.SP.1998.147.01.06.
Goodfellow, I., Bengio, Y., and Courville, A. 2016. Deep Learning. Cambridge, Massachusetts: MIT Press.
Gulli, A. and Pal, S. 2017. Deep Learning With Keras. Birmingham, UK: Packt Publishing.
Haykin, S. 1999. Neural Networks, A Comprehensive Foundation, second edition. Upper Saddle River, New Jersey: Prentice Hall Inc.
Hornik, K., Stinchcombe, M., and White, H. 1989. Multilayer Feedforward Networks are Universal Approximators. Neural Networks 2 (5): 359–366. https://doi.org/10.1016/0893-6080(89)90020-8.
Jolley, S. J., Dijk, H., Lamens, J. H. et al. 2007. Faulting and Fault Sealing in Production Simulation Models: Brent Province, Northern North Sea. Pet Geosci 13 (4): 321–340. https://doi.org/10.1144/1354-079306-733.
LeCun, Y., Bengio, Y., and Hinton, G. 2015. Deep Learning. Nature 521: 436–444. https://doi.org/10.1038/nature14539.
Manzocchi, T., Walsh, J. J., Nell, P. et al. 1999. Fault Transmissibility Multipliers for Flow Simulation Models. Pet Geosci 5 (1): 53–63. https://doi.org/10.1144/petgeo.5.1.53.
Mitra, S., Figueroa, G. C., Garcia, J. H. et al. 2005. Three-Dimensional Structural Model of the Cantarell and Sihil Structures, Campeche Bay, Mexico. AAPG Bull 89 (1): 1–26. https://doi.org/10.1306/08310403108.
Mitra, S. and Leslie, W. 2003. Three-Dimensional Structural Model of the Rhourde el Baguel Field, Algeria. AAPG Bull 87 (2): 231–250. https://doi.org/10.1306/07120201114.
Montgomery, S. L. 1996. Brady Unit, Rock Springs Uplift, Wyoming: Migration and Structural History. AAPG Bull 80 (10): 1535–1546. https://doi.org/10.1306/64EDA0A0-1724-11D7-8645000102C1865D.
Robertson, E. C. 1983. Relationship of Fault Displacement to Gouge and Breccia Thicknesses. Min Eng 35 (10): 1426–1432. https://doi.org/10.1016/0191-8141(88)90020-X.
Ruder, S. 2016. An Overview of Gradient Descent Optimization Algorithms. arXiv:1609.04747 [cs.LG].
Sutskever, I., Martens, J., Dahl, G. et al. 2013. On the Importance of Initialization and Momentum in Deep Learning. Proceedings of the 30th International Conference on Machine Learning, PMLR 28 (3): 1139–1147.
Wang, G. 2016. A Perspective on Deep Imaging. IEEE Access 4: 8914–8924. https://doi.org/10.1109/ACCESS.2016.2624938.
Walsh, J. J., Watterson, J., Heath, A. E. et al. 1998. Representation and Scaling of Faults in Fluid Flow Models. Pet Geosci 4 (3): 241–251. https://doi.org/10.1144/petgeo.4.3.241.
Valcarce, G.Z., Zapata, T., Ansa, A. et al. 2006. Three-Dimensional Structural Modeling and Its Application for Development of the El Porton Field, Argentina. AAPG Bull 90 (3): 307–319. https://doi.org/10.1306/09300504142.