Partially Separated Metamodels With Evolution Strategies for Well-Placement Optimization
- Zyed Bouzarkouna (IFP Energies Nouvelles) | Didier Y. Ding (IFP Energies Nouvelles) | Anne Auger (INRIA)
- Document ID
- Society of Petroleum Engineers
- SPE Journal
- Publication Date
- April 2013
- Document Type
- Journal Paper
- 1,003 - 1,011
- 2013. Society of Petroleum Engineers
- 1.6 Drilling Operations, 5.7.3 Deterministic Methods, 6.5 Environment, 5.5 Reservoir Simulation
- 1 in the last 30 days
- 259 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 10.00|
|SPE Non-Member Price:||USD 30.00|
The net present value (NPV) of a project can be significantly increased by finding the optimal location of non-conventional wells. This optimization problem is nowadays one of the most challenging problems in oil- and gas-field development. Suitable methods to tackle this problem include stochastic optimization algorithms, which are particularly robust and able to deal with complex reservoir geology with high heterogeneities. However, these methods require in general a considerable computational effort in terms of number of reservoir simulations, which are CPU-time-demanding.
This paper presents the use of the CMA-ES (covariance matrix adaptation-evolution strategy) optimizer, recognized as one of the mostpowerful derivative free optimizers, to optimize well locations and trajectories. A local-regression-based metamodel is incorporated into the optimization process in order to reduce the computational cost. The objective function (e.g., the NPV) can usually be split into local components, referring to each of the wells that moreover depends in general on a smaller number of principal parameters, and thus can be modeled as a partially separable function.
In this paper, we propose to exploit the partial separability of theobjective function into CMA-ES coupled with metamodels by building partiallyseparated metamodels. Thus, different metamodels are built for each well or setof wells, which results in a more accurate modeling.
An example is presented. Results show that taking advantage of the partialseparability of the objective function leads to a significant decrease in thenumber of reservoir simulations needed to find the "optimal" wellconfiguration, given a restricted budget of reservoir simulations. The proposedapproach is practical and promising to deal with the placement of a largenumber of wells.
|File Size||842 KB||Number of Pages||9|
Arnold, L., Auger, A., Hansen, N. et al. 2011.Information-Geometric Optimization Algorithms: A Unifying Picture viaInvariance Principles. Document arXiv:1106.3708v1 [math.OC], Cornell UniversityLibrary, arXiv.org (submitted 19 June 2011), http://arxiv.org/abs/1106.3708v1.
Beckner, B.L. and Song, X. 1995. Field Development PlanningUsing Simulated Annealing - Optimal Economic Well Scheduling and Placement.Paper SPE 30650 presented at the SPE Annual Technical Conference andExhibition, Dallas, 22-25 October. http://dx.doi.org/10.2118/30650-MS.
Bouzarkouna, Z., Auger, A., and Ding, D.Y. 2010a.Investigating the Local-Metamodel CMA-ES for Large Population Sizes.Proc., EvoApplications 2011: EvoCOMPLEX, EvoGAMES, EvoIASP,EvoINTELLIGENCE, EvoNUM, and EvoSTOC, Torino, Italy, April 2011, hal-00450238(ver. 1), 402-411.
Bouzarkouna, Z., Auger, A., and Ding, D.Y. 2010b. Usingevolution strategy with metamodels for well placement optimization. Paperpresented at the 12th European Conference on the Mathematics of Oil Recovery(ECMOR XII), Oxford, UK, 6-9 September.
Bouzarkouna, Z., Auger, A., and Ding, D.Y. 2011.Local-metamodel CMA-ES for partially separable functions. Proc., Geneticand Evolutionary Computation Conference (GECCO 2011), Dublin, Ireland, 12-16July, 869-876.
Bouzarkouna, Z., Ding, D., and Auger, A. 2012. Wellplacement optimization with the covariance matrix adaptation evolution strategyand metamodels. Comput. Geosci. 16 (1): 75-92. http://dx.doi.org/10.1007/s10596-011-9254-2.
Bukhamsin, A.Y., Farshi, M.M., and Aziz, K. 2010.Optimization of Multilateral Well Design and Location in a Real Field Using aContinuous Genetic Algorithm. Paper SPE 136944 presented at the SPE/DGS SaudiArabia Section Technical Symposium and Exhibition, Al-Khobar, Saudi Arabia, 4-7April. http://dx.doi.org/10.2118/136944-MS.
Chen, Y. and Oliver, D.S. 2010. Ensemble-Based Closed-LoopOptimization Applied to Brugge Field. SPE Res Eval & Eng 13 (1): 56-71. SPE-118926-PA. http://dx.doi.org/10.2118/118926-PA.
Chen, Y., Oliver, D.S., and Zhang, D. 2009. EfficientEnsemble-Based Closed-Loop Production Optimization. SPE J. 14 (4): 634-645. SPE-112873-PA. http://dx.doi.org/10.2118/112873-PA.
Ding, Y. 2008. Optimization of Well Placement UsingEvolutionary Algorithms. Paper SPE113525 presented at the Europec/EAGEConference and Exhibition, Rome, 9-12 June. http://dx.doi.org/10.2118/113525-MS.
Emerick, A.A., Silva, E., Messer, B. et al. 2009. WellPlacement Optimization Using a Genetic Algorithm With Nonlinear Constraints.Paper SPE 118808 presented at the SPE Reservoir Simulation Symposium, TheWoodlands, Texas, USA, 2-4 February. http://dx.doi.org/10.2118/118808-MS.
Floris, F.J.T., Bush, M.D., Cuypers, M. et al. 2001. Methodsfor quantifying the uncertainty of production forecasts: A comparative study.Pet. Geosci. 7 (Supplement, 1 May): 87-96.
Forouzanfar, F., Li, G., and Reynolds, A.C. 2010. ATwo-Stage Well Placement Optimization Method Based on Adjoint Gradient. PaperSPE 135304 presented at the SPE Annual Technical Conference and Exhibition,Florence, Italy, 19-22 September. http://dx.doi.org/10.2118/135304-MS.
Gelly, S., Ruette, S., and Teytaud, O. 2007.Comparison-Based Algorithms Are Robust and Randomized Algorithms Are Anytime.Evolutionary Computation 15 (4): 411-434. http://dx.doi.org/10.1162/evco.2007.15.4.411.
Guyaguler, B. and Horne, R. 2000. Optimization of WellPlacement. J. Energy Resour. Technol. 122 (2): 64-70. http://dx.doi.org/10.1115/1.483164.
Hansen, N. 2011. The CMA Evolution Strategy: A Tutorial.Report, (28 June 2011), http://www.lri.fr/~hansen/cmatutorial.pdf.
Hansen, N. and Kern, S. 2004. Evaluating the CMA evolutionstrategy on multimodal test functions. In Parallel Problem Solving fromNature - PPSN VIII, ed. X. Yao and J.A. Bullinaria and J. Rowe, et al.,LNCS 3242, 282-291. Berlin: Lecture Notes in Computer Science,Springer-Verlag.
Hansen, N. and Ostermeier, A. 2001. Completely DerandomizedSelf-Adaptation in Evolution Strategies. Evolutionary Computation 9 (2): 159-195. http://dx.doi.org/10.1162/106365601750190398.
Hansen, N., Auger, A., Ros, R. et al. 2010. Comparing Resultsof 31 Algorithms from the Black-Box Optimization Benchmarking BBOB-2009.Proc., Genetic and Evolutionary Computation Conference (GECCO 2010),Portland, Oregon, USA, 7-11 July, 1689-1696.
Kern, S., Hansen, N., and Koumoutsakos, P. 2006. Localmetamodels for optimization using evolution strategies. In Parallel ProblemSolving from Nature - PPSN IX, ed. T.P. Runarsson, H.-G. Beyer, E. Burke,J.J. Merelo-Guervós, L.D. Whitley, and X. Yao, LNCS 4193, 939-948. Berlin:Lecture Notes in Computer Science, Springer-Verlag.
Montes, G., Bartolome, P., and Udias, A.L. 2001. The Useof Genetic Algorithms in Well Placement optimization. Paper SPE 69439 presentedat the SPE Latin American and Caribbean Petroleum Engineering Conference,Buenos Aires, 25-28 March. http://dx.doi.org/10.2118/69439-MS.
Morales, A.N., Nasrabadi, H., and Zhu, D. 2010. AModified Genetic Algorithm for Horizontal Well Placement Optimization in GasCondensate Reservoirs. Paper SPE 135182 presented at the SPE Annual TechnicalConference and Exhibition, Florence, Italy, 19-22 September. http://dx.doi.org/10.2118/135182-MS.
Nocedal, J. 1996. Large Scale Unconstrained Optimization. InThe State of the Art in Numerical Analysis, 311-338. Oxford, UK: OxfordUniversity Press.
Onwunalu, J.E. and Durlofsky, L.J. 2010. Application of aparticle swarm optimization algorithm for determining optimum well location andtype. Comput. Geosci. 14 (1): 183-198. http://dx.doi.org/10.1007/s10596-009-9142-1.
Runarsson, T. 2004. Constrained evolutionary optimization byapproximate ranking and surrogate models. In Parallel Problem Solving fromNature - PPSN VIII, ed. X. Yao and J.A. Bullinaria and J. Rowe, et al.,LNCS 3242, 401-410. Berlin: Lecture Notes in Computer Science,Springer-Verlag.
Sarma, P. and Chen, W.H. 2008. Efficient Well PlacementOptimization with Gradient-based Algorithms and Adjoint Models. Paper SPE112257 presented at the Intelligent Energy Conference and Exhibition,Amsterdam, 25-27 February. http://dx.doi.org/10.2118/112257-MS.
Schulze-Riegert, R., Bagheri, M., Krosche, M. et al. 2011.Multiple-Objective Optimization Applied to Well Path Design under GeologicalUncertainty. Paper SPE 141712 presented at the SPE Reservoir SimulationSymposium, The Woodlands, Texas, USA, 21-23 February. http://dx.doi.org/10.2118/141712-MS.
Vlemmix, S., Joosten, G.J.P., Brouwer, R. et al. 2009.Adjoint-Based Well Trajectory Optimization in a Thin Oil Rim. Paper SPE 121891presented at the EUROPEC/EAGE Conference and Exhibition, Amsterdam, 8-11 June.http://dx.doi.org/10.2118/121891-MS.
Yeten, B., Durlofsky, L.J., and Aziz, K. 2003.Optimization of Nonconventional Well Type, Location, and Trajectory. SPEJ. 8 (3): 200-210. SPE-86880-PA. http://dx.doi.org/10.2118/86880-PA.
Zandvliet, M.J., Handels, M., Van Essen, G.M. et al. 2008.Adjoint-Based Well-Placement Optimization Under Production Constraints. SPEJ. 13 (4): 392-399. SPE-105797-PA. http://dx.doi.org/10.2118/105797-PA.
Zhao, H., Chen, C., Do, S.T. et al. 2011. Maximization of aDynamic Quadratic Interpolation Model for Production Optimization. Paper SPE141317 presented at the SPE Reservoir Simulation Symposium, The Woodlands,Texas, USA, 21-23 February. http://dx.doi.org/10.2118/141317-MS.