셰일가스 생산량 예측 시 생산관련인자를 활용한 장단기기억 신경망 적용 연구
Application of Long Short-Term Memory Neural Networks in Shale Gas Production Prediction using Production-Related Factors
2
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
셰일 저류층의 생산량 자료를 활용하여 생산감퇴곡선분석법 적용 시 생산 초기의 충분하지 않은자료를 사용할 경우 예측 결과에 대한 불확실성이 존재하며, 수압파쇄와 유정완결 조건 등의 생산관련인자를 고려할 수 없다는 한계가 있다. 이 연구에서는 생산시점을 가정하여 셰일가스 생산량을 예측하기 위한 장단기기억 신경망을 구축하였으며, 예측모델이 여러 생산정의 감퇴거동을 학습하여 새로운 생산정에 대한 생산감퇴를 모사할 수 있음을 확인하였다. 또한, 실수직심도, 총천공간격, 수압파쇄 횟수, 초기 생산량에 대한 생산관련인자를 추가 입력자료로 활용함으로써 오차율을개선할 수 있음을 파악하였다. 이 결과를 통하여 생산감퇴곡선분석법 적용이 어려운 셰일가스 생산 초기에 장단기기억 신경망을 바탕으로 생산량 예측이 가능할 것으로 사료된다
Abstract
Insufficient data in the early stages of production results in uncertainty of predicted values on applying decline curve analysis (DCA) using production data of shale reservoir. Additionally, there are limits in production-related factors, such as hydraulic fracturing or well completion conditions, that cannot be considered. In this study, Long Short-Term Memory (LSTM) neural networks were constructed to predict shale gas production by assuming the production time. It was discovered that prediction models could simulate production decline for new wells by learning the decline behavior of multiple shale wells. Moreover, the error could be reduced by using production-related factors for true vertical depth, gross perforated interval, stage count of hydraulic fracturing, and initial production rate, as additional input was confirmed. It may be possible to predict production based on LSTM at the early stages of shale gas production, where it is challenging to apply DCA, through the results of this study.
참고문헌 (30)
- Alabboodi, M.J., "Conditioning the Estimating Ultimate Recovery of Shale Wells to Reservoir and Completion Parameters", The SPE Eastern Regional meeting, pp.1~21 (2016), SPE
- Alom, M. Z., "A State-of-the-Art Survey on Deep Learning Theory and Architectures", Electronics, 8(3), pp.1~66 (2019)
- Arps, J. J., "Analysis of Decline Curves", Transactions of the AIME, 160(1), pp.228~247 (1945)
- Biswas, D., "Adapting Shallow and Deep Learning Algorithms to Examine Production Performance – Data Analytics and Forecasting", The SPE, 2020, 1~21[DOI]
- Engelder, T., "Marcellus Shale Play’s Vast Resource Potential Creating Stir In Appalachia", American Oil and Gas Reporter, 51, pp.76~87 (2008)
- 김용식, "마르셀러스 셰일의 지질학적 특징 및 셰일가스 개발전략", 한국자원공학회지, 48(3), pp.371~382 (2011), 한국자원공학회
- Gupta, I., "Variable Exponential Decline: Modified Arps To Characterize Unconventional-Shale Production Performance", SPE Reservoir Evaluation and Engineering, 21(4), pp.1045~1057 (2018)
- Hochreiter, S., "Long short-term memory", Neural computation, 9(8), pp.1735~1780 (1997)
- 지민수, "물리검층자료를 활용한 딥러닝 기반 수포화도 예측", 한국자원공학회지, 58(3), pp.215~226 (2021), 한국자원공학회
- Kanfar, M., "Comparison of Empirical Decline Curve Methods for Shale Wells", The SPE Canadian Unconventional Resources Conference, pp.1~12 (2012), SPE
- 강판상, "셰일 저류층 물성 및 수압파쇄 조건에 따른 생산감퇴곡선 인자 영향 분석", 한국자원공학회지, 54(3), pp.223~232 (2017), 한국자원공학회
- 기세일, "순환신경망을 이용한 미측정 정두압 데이터 예측", 한국자원공학회지, 56(5), pp.416~426 (2019), 한국자원공학회
- 김지수, "확률론적 생산감퇴곡선분석법을 이용한 셰일가스 궁극가채량 추정", 한국자원공학회지, 51(6), pp.808~819 (2014), 한국자원공학회
- Kocoglu, Y., "Application of Bayesian Optimized Deep Bi-LSTM Neural Networks for Production Forecasting of Gas Wells in Unconventional Shale Gas Reservoirs", The SPE, SEG Unconventional Resources Technology Conference, URTeC, pp.2021[DOI]
- Lee, K., "Prediction of Shale-Gas Production at Duvernay Formation Using Deep-Learning Algorithm", SPE Journal, 24(6), pp.2423~2437 (2019)
- Li, X., "Application of Gated Recurrent Unit (GRU) Neural Network for Smart Batch Production Prediction", Energies, 13(22), pp.1~22 (2020)
- Luo, G., "Production-Strategy Insights Using Machine Learning: Application for Bakken Shale", SPE Reservoir Evaluation and Engineering, 22(3), pp.800~816 (2019)
- Luo, S., "Study on the Production Decline Characteristics of Shale Oil: Case Study of Jimusar Field", Frontiers in Energy Research, 10, pp.1~13 (2022)
- Mosobalaje, O. O., "Theories of Arps’ Decline Curve Exponent and Loss Ratio, for Saturated Reservoirs", The Abu Dhabi International Petroleum Exhibition and Conference, pp.1~16 (2014), SPE
- 오현택, "머신러닝 기반 생산기간에 따른 셰일가스 궁극가채량 예측의 불확실성 추이 분석", 한국자원공학회지, 58(5), pp.475~490 (2021), 한국자원공학회
- Paryani, M., "Using Improved Decline Curve Models for Production Forecasts in Unconventional Reservoirs", The SPE Eastern Regional Meeting, pp.1~15 (2016), SPE
- Rezaee, R., "Fundamentals of Gas Shale Reservoirs", pp.283~300 (2015), John Wiley & Sons
- Shin, H. J., "Probabilistic Prediction of Multi-Wells Production Based on Production Characteristics Analysis Using Key Factors in Shale Formations", Energies, 14(17), pp.1~30 (2021)
- Song, X., "Time-series well performance prediction based on Long Short-Term Memory (LSTM) neural network model", Journal of Petroleum Science and Engineering, 186, pp.1~11 (2020)
- Tan, L., "Methods of Decline Curve Analysis for Shale Gas Reservoirs", Energies, 11(3), pp.1~18 (2018)
- "U.S. Energy Information Administration(EIA)" (2022)[URL]
- Wang, S., "Applicability of deep neural networks on production forecasting in Bakken shale reservoirs", Journal of Petroleum Science and Engineering, 179, pp.112~125 (2019)
- Yao, B., "Prediction of Wax Deposits for Crude Pipelines Using Time-Dependent Data Mining", SPE Journal, 26(4), pp.1980~2001 (2021)
- Zhan, C., "Application of Machine Learning for Production Forecasting for Unconventional Resources", The SPE, SEG Unconventional Resources Technology Conference, URTeC, pp.2019[DOI]