History Matching of Gas Production Rates Integrated an Artificial Neural Network with Distance-based Candidate Selection

Jaejun Kim; Joe M. Kang; Changhyup Park; Seongin Ahn; Baehyun Min

doi:10.12972/ksmer.2017.54.4.416

All Issue

2017 Vol.54, Issue 4 Preview Page Next Page

Research Paper

History Matching of Gas Production Rates Integrated an Artificial Neural Network with Distance-based Candidate Selection 거리기반 후보군 선정과 인공신경망을 결합한 가스생산량 히스토리매칭

31 August 2017. pp. 416-428

PDF XML

Abstract

This paper presents a reliable history matching method that improves supervised-learning performances of an artificial neural network using the candidate models selected by distance-based clustering. The artificial neural network is used as an inverse method instead of a proxy model. The reservoir models within the cluster containing the true dataset update the training module. The developed method matches 2-year production histories and then estimates 3-year performances followed at shale gas reservoirs. The training performance improves the mean absolute error of daily gas rates from 29.2% to 4.3% while the optimum solutions, obtained by k-medoids clustering, show about 1.2% error of daily gas rates and 0.7% error of total production. The sensitivity analysis made by the optimal solutions reveals that the matrix porosity and its permeability influence the shale-gas production. The developed approach is able to estimate unknown production rates with reliable predictability.

Keywords

Artificial neural network

Distance-based clustering

History matching

Gas production

이 연구에서는 거리기반 후보군 선정법을 인공신경망과 결합하여 생산 히스토리매칭을 수행하고, 생산추이를 신뢰도 높게 예측하는 기법을 개발하였다. 생산추이를 입력층으로 한 역산모델 인공신경망을 구성하고, 거리기반 군집화를 통해 실제 생산자료와 유사한 추이를 보이는 후보모델을 인공신경망의 학습집단으로 활용하여 지도학습효과를 개선하였다. 4개의 수압파쇄대를 갖는 셰일가스전에서 2년간 생산이력을 히스토리매칭하여 향후 3년간의 생산이력을 예측하였다. 인공신경망 학습을 통해 매칭한 학습집단의 일가스생산량 오차는 초기 29.2%에서 마지막 세대에는 4.3%로 개선되었다. 최적해는 참값과 비교하여 일가스생산량은 1.2%, 총생산량은 0.7%의 오차를 보였다. 최적해를 이용한 민감도 분석결과, 모암의 공극률과 유체투과도가 생산추이예측에 민감하였다. 이 연구의 유사생산추이 후보군을 이용한 인공신경망 학습능력 개선기술은 효율적인 저류층 특성화와 신뢰도 높은 생산량 예측기법으로 활용할 수 있다.

키워드

인공신경망

거리기반 군집화

히스토리매칭

가스생산량

References

Ahn, S., 2017. Characterization of Channelized Reservoir using the Neural Network Incorporated with Deep Autoencoder, MS Thesis, Seoul National University, Seoul, Korea, 70p.

Anderson, D.M., Liang, P., and Okouma, V., 2012. Probabilistic forecasting of unconventional resources using rate transient analysis: Case studies. The SPE Americas Unconventional Resources Conference, SPE, Pittsburgh, Pennsylvania, USA, SPE-155737-MS.

Batycky, R.P., Blunt, M.J., and Thiele, M.R., 1997. A 3D field-scale streamline-based reservoir simulator. SPE Reserv. Eng, 12, 246-254.

10.2118/36726-PA

Berkhin, P., 2006. A survey of clustering data mining techniques. In: Kogan, J., Nicholas, C., and Teboulle, M. (eds) Grouping Multidimensional Data, Springer, Berlin, Heidelberg, Germany, p.25-71.

10.1007/3-540-28349-8_2

Centilmen, A., Ertekin, T., and Grader, A.S., 1999. Applications of neural networks in multiwell field development. The SPE Annual Technical Conference and Exhibition, SPE, Houston, USA, SPE-56433-MS.

10.2118/56433-MS

Fenwick, D., Scheidt, C., and Caers, J., 2014. Quantifying asymmetric parameter interactions in sensitivity analysis: Application to reservoir modeling. Mathematical Geosciences, 46(4), 493-511.

10.1007/s11004-014-9530-5

German, G.M., Navarro-Rosales, R.A., and Dubost, F.X., 2012. Production forecasting for shale gas exploration prospects based on statistical analysis and reservoir simulation. The SPE Latin America and Caribbean Petroleum Engineering Conference, SPE, Mexico City, Mexico, SPE- 152639-MS.

Kam, D., Min, B., Chung, S., Park, C., Kang, J.M., Kim, J., Jang, I., and Choi, Y., 2009. Optimization of steam injection pressure in SAGD using artificial neural network. J. of the Korean Society for Geosystem Engineering, 46(2), 143-150.

Kam, D., Park, C., Min, B., and Kang, J.M., 2013. An optimal operation strategy of injection pressures in solvent-aided thermal recovery for viscous oil in sedimentary reservoirs. Petroleum Science and Technology, 31(22), 2378-2387.

10.1080/10916466.2011.569816

Kim, J., Kang, J.M., Park, C., Park, Y., Park, J., and Lim, S., 2017. Multi-objective history matching with a proxy model for the characterization of production performances at the shale gas reservoir. Energies, 10(4), 579.

10.3390/en10040579

Kim, J., Kang, J.M., Park, Y., Lim, S., Park C., and Park, J., 2016. Probabilistic estimation of shale gas reserves implementing fast marching method and Monte Carlo simulation. The ASME 2016 35th International Conference on Ocean, Offshore and Arctic Engineering, ASME, Busan, Korea, OMAE2016-54167.

10.1115/OMAE2016-54167

Kim, K., Ju, S., Ahn, J., Shin, H., Shin, C., and Choe, J., 2015. Determination of key parameters and hydraulic fracture design for shale gas productions. The Twenty-fifth International Ocean and Polar Engineering Conference, ISOPE, Kona, Hawaii, USA, ISOPE-I-15-812.

Lim, S., 2016. On the Study of a Robust History Matching of Facies Model by Distance-based Method and Ensemble Kalman Filter, MS Thesis, Seoul National University, Seoul, Korea, 68p.

Min, B., Park, C., Kang, J.M., Park, H.J., and Jang, I., 2011. Optimal well-placement based on artificial neural network incorporating the productivity potential. Energy Sources, Part A, 33(18), 1726-1738.

Park, H., Lim, J., and Kang, J.M., 2005. Analysis of multi-well gas reservoir performance using polynomial neural network (PNN) with layer over-passing structure. J. of the Korean Society for Geosystem Engineering, 42(3), 152-159.

Park, H.S. and Jun, C.H., 2009. A simple and fast algorithm for k-medoids clustering. Expert Systems with Applications, 36(2), 3336-3341.

10.1016/j.eswa.2008.01.039

Park, J., 2015. A Characterization of Shale Gas Reservoir using Fast Marching Method Combined with Model Selection Approach, MS Thesis, Seoul National University, Seoul, Korea, 82p.

Park, J., Yang, G., Satija, A., Scheidt, C., and Caers, J., 2016. DGSA: A matlab toolbox for distance-based generalized sensitivity analysis of geoscientific computer experiments. Computers & Geosciences, 97, 15-29.

10.1016/j.cageo.2016.08.021

Scheidt, C. and Caers, J., 2009. Uncertainty quantification in reservoir performance using distances and kernel methods: Application to a west Africa deepwater turbidite reservoir. SPE Journal, 14(4), 680-692.

10.2118/118740-PA

Sethian, J.A. and Vladimirsky, A., 2000. Fast methods for the eikonal and related Hamilton-Jacobi equations on unstructured meshes. Proceedings of the National Academy of Sciences, 97(11), 5699-5703.

10.1073/pnas.09006009710811874PMC18495

Shahkarami, A., Mohaghegh, S.D., Gholami, V., and Haghighat, S.A., 2014. Artificial intelligence (AI) assisted history matching. The SPE Western North American and Rocky Mountain Joint Meeting, SPE, Denver, Colorado, USA, SPE-169507-MS.

Velmurugan, T. and Santhanam, T., 2010. Computational complexity between k-means and k-medoids clustering algorithms for normal and uniform distributions of data points. Journal of Computer Science, 6(3), 363-368.

10.3844/jcssp.2010.363.368

Xie, J., Yang, C., Gupta, N., King, M.J., and Datta-Gupta, A., 2015. Depth of investigation and depletion in unconventional reservoirs with fast-marching methods. SPE Journal, 20(4), 831-841.

10.2118/154532-PA

Information

Publisher :The Korean Society of Mineral and Energy Resources Engineers
Publisher(Ko) :한국자원공학회
Journal Title :Journal of the Korean Society of Mineral and Energy Resources Engineers
Journal Title(Ko) :한국자원공학회지
Volume : 54
No :4
Pages :416-428
Received Date : 2017-05-09
Revised Date : 2017-08-24
Accepted Date : 2017-08-30
DOI :https://doi.org/10.12972/ksmer.2017.54.4.416

Journal of the Korean Society of Mineral and Energy Resources Engineers ISSN:2288-0291(Print) 2288-2790(Online) 한국자원공학회지

All Issue