The Development, Application and Prospect of Oil and Gas Production Forecasting Technologys

Yupeng Zhuang; Haikun Zeng; Xin Cheng

doi:10.62051/ijnres.v6n3.02

Authors

Yupeng Zhuang
Haikun Zeng
Xin Cheng

DOI:

https://doi.org/10.62051/ijnres.v6n3.02

Keywords:

Oil and gas production forecast; Machine learning; Deep learning.

Abstract

In the practice of oil and gas field development, the heterogeneity of reservoir physical properties, the dynamic changes of fluid properties, and the diversity of process measures constitute a complex development environment. This complexity makes the analysis of oil and gas production data highly dependent on the professional experience of petroleum scientists and engineers. Traditional analysis methods not only consume a lot of computing resources and time costs, but also are difficult to meet the needs of efficient development of modern oil and gas reservoirs. Therefore, it is urgent to explore more efficient production forecasting methods. In recent years, with the rise of machine learning technologies such as deep neural networks and random forest algorithms, with their significant advantages in high - dimensional data processing, time - series feature capture, and development data feature mining, they have achieved fruitful results in the field of oil and gas production forecasting. This paper systematically combs the evolution context of oil and gas production forecasting technology, elaborates the principles, advantages and disadvantages of mainstream machine learning methods, summarizes the application status of machine learning methods in the field of oil and gas production forecasting, analyzes the potential problems in the application process, and looks forward to the future development trend. The study points out that in the future, we should focus on promoting two technological breakthroughs: first, organically integrating the physical mechanism of the reservoir with the machine learning model to enhance the interpretability of the model and ensure the reliability of the prediction results; second, developing algorithms and transfer learning technologies suitable for small sample scenarios to deeply tap the value of historical production data and provide more accurate and efficient technical support for oil and gas production forecasting.

References

[1]KUANG Lichun, LIU He, REN Yili, et al. Application and development trend of artificial intelligence in petroleum exploration and development[J]. Petroleum Exploration and Development, 2021, 48(1): 1-11.

[2]LU Jialiang, ZHAO Suping, SUN Yuping, et al. Natural gas production peaks in China: Research and strategic proposals[J]. Natural Gas Industry, 2018, 38(1): 1-9.

[3]Rushing J.A, Perego A.D, Sullivan R.E. Estimating reserves in tight gas sands at HP/HT reservoir conditions: use and misuse of an Arps decline curve methodology[C]. SPE Annual Technical Conference and Exhibition, 2007.

[4]Ilk D, Rushing J.A, Perego A.D, et al. Exponential vs. Hyperbolic Decline in Tight Gas Sands: Understanding the Origin and Implications for Reserve Estimates Using Arps' Decline Curves[C]. SPE Technical Conference and Exhibition, Society of Petroleum Engineers, 2008.

[5]Duong A.N. An Unconventional Rate Decline Approach for Tight and Fracture-Dominated Gas Wells[C]. Society of Petroleum Engineers, 2010.

[6]Clark A.J, Lake L.W, Patzek T.W. Production Forecasting with Logistic Growth Models[C]. Society of Petroleum Engineers, 2011.

[7]Chen Yuanqian, Hu Jianguo. The Weibull model of predicting oil product and reserve volume[J]. Xinjiang Petroleum Geology, 1995, 16(3): 250-255.

[8]Gao Wenjun. Theoretic basic of bi logarithmic water-drive characteristic curves[J]. Xinjiang Petroleum Geology, 2008, 29(3): 338-340.

[9]Hong A, Bratvold R.B, Lake L.W, et al. Integrating Model Uncertainty in Probabilistic Decline-Curve Analysis for Unconventional-Oil-Production Forecasting[J]. SPE Reservoir Evaluation and Engineering, 2019, 22(3): 861-876.

[10]Han Juange. Study on shale gas production decline analysis[D]. Beijing: China University of Petroleum (Beijing), 2018.

[11]Li Pingyou. Study on the production decline law of horizontal wells in low permeability reservoirs[D]. Beijing: China University of Geosciences (Beijing), 2018.

[12]Ji Bingyu. The seepage theoretical basis of the production decline equation[J]. Acta Petrolei Sinica, 1995, (03): 86-91.

[13]Xing Minghai, Chen Xiangguang, Wang Yu. Application of fuzzy neural network in oilfield production prediction[J]. Computer Simulation, 2005(02): 187-190.

[14]Lü Yumin, Tang Dazhen, Li Zhiping, et al. Dynamic productivity fitting and prediction model of coalbed methane wells[J]. Journal of China Coal Society, 2011, 36(09): 1481-1485.

[15]Tian Yapeng, Ju Binshan. Shale gas production decline prediction model based on genetic algorithm improved BP neural network[J]. China Science and Technology Paper, 2016, 11(15): 1710-1715.

[16]Cao Q, Banerjee R, Gupta S, et al. Data Driven Production Forecasting Using Machine Learning[C]. SPE Argentina Exploration & Production of Unconventional Resources Symposium, 2016.

[17]Fernández-Navarro, F, Angeles D.L.C.M, Gutiérrez, P.A, et al. Time series forecasting by recurrent product unit neural networks[J]. Neural Computing and Applications, 2018, 29(3): 779-791.

[18]Li Y, Han Y. Decline Curve Analysis for Production Forecasting Based on Machine Learning[C]. SPE Symposium: Production Enhancement and Cost Optimisation. 2017.

[19]Luo G, Yao T, By china M, et al. Production Optimization Using Machine Learning in Bakken Shale[C]. Unconventional Resources Technology Conference, 2018.

[20]Tang Lifang, Tang Wei. Oil demand prediction based on CPSO-SVM[J]. Computer Simulation, 2012, 29(06): 206-209.

[21]Zheng Minggui, Li Qi. Scenario prediction of China's petroleum resource demand from 2020 to 2030[J]. Advances in Earth Science, 2020, 35(03): 286-296.

[22]Sun Mei, Deng Xing, Dong Tongming. Application of neural network and Kalman algorithm in oil demand prediction[J]. Journal of Beijing Institute of Technology (Social Science Edition), 2013, 15(06): 10-15.

[23]Zhang Fangzhou, Yan Huyong, Yang Liquan, et al. Application of improved gray neural network model in oilfield production[J]. Computer Technology and Development, 2013, 23(06): 241-244+248.

[24]Li Zhenyu, Lu Hong, Ren Wenpo, et al. Analysis of the development trend of China's future oil consumption[J]. Chemical Industry and Engineering Progress, 2016, 35(06): 1739-1747.

[25]Xiang Yunfei, Kang Zhihong, Hao Weijun, et al. Composite method for reservoir parameter prediction based on linear regression and neural network[J]. Science Technology and Engineering, 2017, 17(31): 46-52.

[26]Wang Yong, Zhang Linxia. A study on combined model for predicting shale gas well production decline[J]. China Petroleum and Chemical Standard and Quality, 2019, 39(20): 121-123.

[27]Zheng Minggui, Yuan Xuemei. Demand prediction of China's chromium ore from 2020 to 2030 based on grey neural network[J]. Resource Development and Market, 2018, 34(06): 747-752.

[28]Fan D.Y, Sun H, Yao J, et al. Well production forecasting based on ARIMA-LSTM model considering manual operations[J]. Energy, 2021, 220: 119708.

[29]Liu Wei, Liu Wei, Gu Jianwei. Daily oil production prediction of oil wells based on machine learning methods[J]. Oil Drilling & Production Technology, 2020, 42(01): 70-75.

[30]ZHA Wenshu, LIU Yuping, WAN Yujin, et al. Forecasting monthly gas field production based on the CNN-LSTM model[J]. Energy, 2022, 260: 124889.

[31]LI Daolun, ZHOU Xia, XU Yanmei, et al. Deep learning-based analysis of the main controlling factors of different gas-fields recovery rate[J]. Energy, 2023, 285: 128767.

[32]HUANG Hu, GONG Bin, SUN Wenyue. A deep-learning based graph neural network-long-short-term memory model for reservoir simulation and optimization with varying well controls[J]. SPE Journal, 2023, 28(6): 2898-2916.

[33]DU Enda, LIU Yuetian, CHENG Ziyan, et al. Production forecasting with the inter well interference by integrating graph convolutional and long short-term memory neural network[J]. SPE Reservoir Evaluation & Engineering, 2022, 25(2): 197-213.

[34]LUNDBERG S M，ERION G G，LEE S I. Consistent individualized feature attribution for tree ensembles[J]. Ar Xiv Preprint, 2018, (2): 1706-1711.

[35]MILLER T. Explanation in artificial intelligence: Insights from the social sciences [J]. Artificial Intelligence, 2019, 267: 1-38.

[36]LI Daolun, ZHOU Xia, XU Yanmei, et al. Deep learning-based analysis of the main controlling factors of different gas-fields recovery rate[J]. Energy, 2023, 285: 128767.

[37]HUANG Hu, GONG Bin, SUN Wenyue. A deep-learning based graph neural network-long-short-term memory model for reservoir simulation and optimization with varying well controls[J]. SPE Journal, 2023, 28(6): 2898-2916.

[38]WANG Biao, HAN Guoqing, LU Xin, et al. The integrated geosciences and engineering production prediction in tight reservoir based on deep learning[J]. Geoenergy Science and Engineering, 2023, 223: 211571.

[39]YILDIRIM K, SHELDON B G, HOSSEIN E, et al. Improving the accuracy of short-term multiphase production forecasts in unconventional tight oil reservoirs using contextual Bi-directional long short-term memory[J]. Geoenergy Science and Engineering, 2024, 235: 212688.

[40]LIANG Bin, LIU Jiang, KANG Lixia, et al. A novel framework for predicting non-stationary production time series of shale gas based on BiLSTM - RF - MPA deep fusion model[J/OL]. Petroleum Science, 2024.