Journal of the Korean Institute of Gas (한국가스학회지)

The Korean Institute of GAS (한국가스학회)
권호 표지
DOI QR코드

DOI QR Code

Development of Productivity Prediction Model according to Choke Size and Gas Injection Rate by using ANN(Artificial Neural Network) at Oil Producer

오일 생산정에서 쵸크사이즈와 가스주입량에 따른 생산성 예측 인공신경망 모델 개발

  • Han, Dong-kwon (Dept. of Energy and Mineral Resources Engineering, Dong-A University) ;
  • Kwon, Sun-il (Dept. of Energy and Mineral Resources Engineering, Dong-A University)
  • 한동권 (동아대학교 에너지자원공학과) ;
  • 권순일 (동아대학교 에너지자원공학과)
  • Received : 2018.06.19
  • Accepted : 2018.12.18
  • Published : 2018.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

This paper presents the development of two ANN models which can predict an optimum production rate by controlling choke size in oil well, and gas injection rate in gas-lift well. The input data was solution gas-oil ratio, water cut, reservoir pressure, and choke size or gas injection rate. The output data was wellhead pressure and production rate. Firstly, a range of each parameters was decided by conducting sensitive analysis of input data for onshore oil well. In addition, 1,715 sets training data for choke size decision model and 1,225 sets for gas injection rate decision model were generated by nodal analysis. From the results of comparing between the nodal analysis and the ANN on the same reservoir system showed that the correlation factors were very high(>0.99). Mean absolute error of wellhead pressure and oil production rate was 0.55%, 1.05% with the choke size model, respectively. And the gas injection rate model showed the errors of 1.23%, 2.67%. It was found that the developed models had been highly accurate.

본 연구에서는 초크크기와 가스주입량을 조절함으로써 일반 유정이나 가스리프트가 적용된 유정에서 최적생산량을 산출할 수 있는 두 가지 인공신경망 모델을 개발하였다. 개발된 모델들의 입력자료는 용해가스-오일비, 물 생산 비율, 저류층압력, 초크크기 또는 가스주입량이고 출력자료는 정두압력과 오일 생산량으로 구성하였다. 먼저 육상 유정 시스템에 대하여 입력자료의 민감도 분석을 통해 각 변수의 범위를 결정하였고, 노달분석을 수행하여 초크크기 선정 모델에 1,715개, 가스주입량 선정 모델에 1,225개의 훈련자료를 각각 생성하였다. 동일한 저류층 자료에 대해 노달분석과 인공신경망 모델 결과를 비교해보면 두 모델 모두 결정계수 값이 0.99 이상으로 상관관계가 매우 높은 것으로 확인되었다. 또한 초크크기 선정 모델의 정두압력과 오일 생산량의 평균절대백분율오차는 각각 0.55%, 1.05%이고, 가스주입량 선정 모델의 정두압력과 오일 생산량의 평균절대백분율오차는 각각 1.23%, 2.67%로 개발된 모델의 정확도가 높은 것으로 확인되었다.

Keywords

GSGSBE_2018_v22n6_90_f0001.png 이미지

Fig. 1. Possible pressure losses in completed producing system.

GSGSBE_2018_v22n6_90_f0002.png 이미지

Fig. 2. Determination of system flow capacity using nodal analysis.

GSGSBE_2018_v22n6_90_f0003.png 이미지

Fig. 3. Schematic of gas lift well system

GSGSBE_2018_v22n6_90_f0004.png 이미지

Fig. 4. Graph to determine the optimum gas rate in gas-lift well[10].

GSGSBE_2018_v22n6_90_f0005.png 이미지

Fig. 5. Structure of basic neural network.

GSGSBE_2018_v22n6_90_f0006.png 이미지

Fig. 6. Structure of an ANN using error back-propagation algorithm.

GSGSBE_2018_v22n6_90_f0007.png 이미지

Fig. 7. Nodal analysis for selecting water cut range.

GSGSBE_2018_v22n6_90_f0008.png 이미지

Fig. 8. Nodal analysis for selecting choke size range.

GSGSBE_2018_v22n6_90_f0009.png 이미지

Fig. 9. Structure input-output data of ANN model in choke size seleciton.

GSGSBE_2018_v22n6_90_f0010.png 이미지

Fig. 10. Correlation of coefficient for choke size ANN model.

GSGSBE_2018_v22n6_90_f0011.png 이미지

Fig. 11. Structure input-output data of ANN model in gas injection rate selection.

GSGSBE_2018_v22n6_90_f0012.png 이미지

Fig. 12. Nodal analysis for selecting gas injection rate range.

GSGSBE_2018_v22n6_90_f0013.png 이미지

Fig. 13. Correlation of coefficient for gas lift ANN model.

Table 1. Statistical range of parameter for input data(choke size ANN model).

GSGSBE_2018_v22n6_90_t0001.png 이미지

Table 2. Statistical range of parameter for output data(choke size ANN model).

GSGSBE_2018_v22n6_90_t0002.png 이미지

Table 3. Statistical accuracy of choke size ANN model.

GSGSBE_2018_v22n6_90_t0003.png 이미지

Table 4. Validity verification of choke size ANN model.

GSGSBE_2018_v22n6_90_t0004.png 이미지

Table 5. Statistical range of parameter for input data(gas lift ANN model).

GSGSBE_2018_v22n6_90_t0005.png 이미지

Table 6. Statistical range of parameter for output data(gas lift ANN model).

GSGSBE_2018_v22n6_90_t0006.png 이미지

Table 7. Statistical accuracy of gas lift ANN model.

GSGSBE_2018_v22n6_90_t0007.png 이미지

Table 8. Validity verification of choke size ANN model.

GSGSBE_2018_v22n6_90_t0008.png 이미지

References

  1. Cramer, R., Gobel, D., Mueller, K., and Tulalian, R., "A measure of the digital oil fields status - Is it the end of the beginning", SPE Intelligent Energy International, Utrecht, The Netherlands, March 27-29. SPE 149957, (2012).
  2. Lim, J., Park, H. and Lim, J., "Classification and application of digital oil field System", J. Korean Soc. Miner. Energy Resour. Eng, Vol. 51, No. 5, 750-756, (2014). https://doi.org/10.12972/ksmer.2014.51.5.750
  3. Al-Jasmi, A., Goel, H.K., Nasr, H., Carvajal, G.A., Johnson, D.W., Cullick, A.S., Rodriguez, J.A., Moricca, G., Velasquez, G., Villamizar, M. and Querales, M., "A surveillance 'Smart Flow' for intelligent digital production operations", SPE Digital Energy Conference and Exhibition, Woodlands, Texas, USA, March 5-7. SPE 163697, (2013).
  4. Guo, B., Lyons, W.C., and Ghalambor, A., "Petroleum Production Engineering A Computer-Assisted Approach, Gulf Professional Publishing", Oxford, U.K., 182, (2007).
  5. 김의중., 알고리즘으로 배우는 인공지능, 머신러닝, 딥러닝 입문, 위키북스, (2016).
  6. Mohammed, A. Al-Khalifa. and Muhammad, A. and Al-Marhoun., "Application of neural network for two-phase flow through chokes", SPE Saudi Arabia section Annual Technical Symposium and Exhibition, Khobar, Saudi Arabia, May 19-22, SPE 169597, (2013).
  7. Khamehchi, A., Rashidi, F. and Rasouli, H., "Prediction of Gas Lift Parameters Using Artificial Neural Networks", Enhanced Oil Recovery-Iranian Chemical Engineering Journal (Special Issue), Vol. 8, No. 43, (2009).
  8. Ranjan, A., Verma, S. and Singh, Y., "Gas Lift Opimization using Artificial Neural Network", SPE Middle East Oil & Gas Show and Conference, Manama, Bahrain, March 8-11, SPE 172610, (2015).
  9. Beggs, H.D., "Production Optimization Using Nodal Analysis", 2th Ed., Vol. 1, OGCI and Petroskills Publications, Tulsa, USA, (2008).
  10. Robert, N.H. and Potter, K., "Optimization of Gas-Injected Oil Wells", SAS Gloval Forum, (2011).
  11. Sim, S.S.K., "Pressure-Volume-Temperature Correlations for Crud Oils from the Illinois Basin", Illinois State Geological Survey, 140, (1993).

Cited by

  1. 시추효율매개변수를 이용한 굴진율 예측 지도학습 모델 비교 연구 vol.25, pp.8, 2021, https://doi.org/10.6109/jkiice.2021.25.8.1032