«Oilfield engineering»

INFLUENCE OF THE OBJECTIVE FUNCTION WEIGHTING METHODS ON THE AUTOMATED RESERVOIR MODEL HISTORY MATCHING EFFICIENCY

UDC: 622.276:532:519.876
DOI: 10.33285/0207-2351-2021-1(625)-33-40

Authors:

¹ Petroleum Learning Centre of Tomsk Polytechnic University, Tomsk, Russian Federation

Keywords: weight coefficients, objective function, automated history matching, automated history matching of a model, optimization algorithms, history matching quality, history matching efficiency, reservoir geological-hydrodynamic simulation, oil field, numerical model

Annotation:

The article considers one of the aspects of the targeted objective function formulation for carrying out an automated history matching of hydrocarbon reservoirs geological-hydrodynamic models, namely, choice of weight coefficients. Numerical modeling of a field includes three basic stages: development of a static geological model, building it on the basis of a dynamic filtration model as well as iterative process of the developed geological-hydrodynamic model matching. To carry out history auto-matching, it is necessary to define the objective function that describes the discrepancy between simulation results and observed data. The aim of the work is to study the influence of the targeted function weighting methods on history matching efficiency based on a synthetic model of an oil reservoir to develop recommendations for setting the objective function. The novelty of the study lies in comprehensive study of three types of weighting factors and identification of conditions for their reasonable use. Based on the results of this research, recommendations were made regarding the use of objective function weighting methods in history matching of hydrocarbon reservoir models.

Bibliography:

1. Design of Objective Function for Interference Well Testing / R. Booth, A.C. Bertolini, K.L. Morton, A.J. Fitzpatrick // OTC-24513-MS. - 2013.
2. Ding Y.D., McKee F. Using partial separability of the objective function for gradient-based optimizations in history matching // SPE-140811-MS. - 2011.
3. How does the definition of the objective function influence the outcome of history matching? / G. Eremyan, I. Matveev, G. Shishaev [et al.] // Conference Proceedings, ECMOR XVII. - Sep. 2020. - Vol. 2020. - P. 1-14.
4. Ferreira C.J., Davolio A., Schiozer D.J. Use of a Probabilistic and Multi-Objective History Matching for Uncertainty Reduction for the Norne Benchmark Case // SPE-185837-MS. - 2017.
5. Hajizadeh Y., Christie M.A., Demyanov V. Towards multiobjective history matching: Faster convergence and uncertainty quantification // SPE-141111-MS. - 2011.
6. Hutahaean J.J., Demyanov V.V., Christie M.A. On Optimal Selection of Objective Grouping for Multiobjective History Matching // SPE-185957-PA. - 2017.
7. Ishibuchi H., Tsukamoto N., Nojima Y. Evolutionary many-objective optimization: A short review // Proceedings of 2008 IEEE Congress on Evolutionary Computation. - Hong Kong, June 1-6, 2008. - P. 2424-2431.
8. Geology realism control in automated history matching / I. Matveev, G. Shishaev, G. Eremyan [et al.] // Conference Proceedings, ECMOR XVII. - Sep. 2020. - Vol. 2020. - P. 1-9.
9. Geology Driven History Matching / I.V. Matveev, G.Y. Shishaev, G.A. Eremyan [et al.] // SPE-196881-MS. - 2019.
10. Multiobjective Optimization With Application to Model Validation and Uncertainty / R.W. Schulze-Riegert, M. Krosche, A. Fahimuddin, S.G. Ghedan // SPE-105313. - 2007.