综述

机器学习方法预测油气产量技术发展现状及前景展望

  • 谢坤 ,
  • 田轩硕 ,
  • 刘长龙 ,
  • 邵明 ,
  • 刘延春 ,
  • 高铭宣 ,
  • 袁世亮 ,
  • 张宝岩
展开
  • 1.东北石油大学提高油气采收率教育部重点实验室,黑龙江 大庆 163318;
    2.中海石油(中国)有限公司天津分公司渤海石油研究院,天津 300459;
    3.中国石油大庆油田有限责任公司采油工程研究院,黑龙江 大庆 163453
谢坤(1991—),男,副教授,2013年毕业于东北石油大学石油工程专业,2019年毕业于该校石油与天然气工程专业,获博士学位,现从事提高采收率理论和技术研究工作。

收稿日期: 2024-08-02

  修回日期: 2025-05-06

  网络出版日期: 2025-09-03

基金资助

国家自然科学基金青年基金“高矿化度油藏智能响应微凝胶构建及时变液流转向机制研究”(52204037);中央支持地方高校改革发展资金优秀青年项目“智能响应复合相微凝胶构建及时变液流转向效果研究”

Status and prospects of machine learning methods for predicting hydrocarbon production

  • XIE Kun ,
  • TIAN Xuanshuo ,
  • LIU Changlong ,
  • SHAO Ming ,
  • LIU Yanchun ,
  • GAO Mingxuan ,
  • YUAN Shiliang ,
  • ZHANG Baoyan
Expand
  • 1. Key Laboratory for Improving Oil and Gas Recovery (Northeast Petroleum University), Ministry of Education, Daqing, Heilongjiang 163318, China;
    2. CNOOC (China) Tianjin Company, Tianjin Branch, Bohai Oil Research Institute, Tianjin 300459, China;
    3. PetroChina Daqing Oilfield Co., Ltd., Oil Production Engineering Research Institute, Daqing, Heilongjiang 163453, China

Received date: 2024-08-02

  Revised date: 2025-05-06

  Online published: 2025-09-03

摘要

受油气开发过程中储层物性、流体性质和工艺措施的复杂多变性影响,生产数据的分析利用程度多取决于石油科技工作者的专业经验,计算成本和时间成本高,难以满足油气藏高效开发需求,亟待发现更加高效的油气产量预测方法。近年来,以深度神经网络、随机森林算法和迁移学习为代表的机器学习方法凭借处理高维数据、捕捉时序数据长期依赖关系和提取开发数据特征等方面的独特优势,在油气产量预测中取得了显著应用效果。该文通过对常用油气产量预测机器学习方法的原理及其优缺点进行分析,阐述了机器学习方法在油气产量预测领域的应用现状,总结了应用过程中潜在的问题,同时对油气产量预测方法的发展前景进行展望。未来,一方面应加强对物理约束融入机器学习模型的研究,提高模型的可解释性,避免过于理想化的预测结果;另一方面要开发适合小样本情况下的算法和迁移学习技术,充分利用历史生产数据,为油气产量预测提供更好的数据分析技术支持。该研究对油气产量的智能预测技术完善具有理论指导意义。

本文引用格式

谢坤 , 田轩硕 , 刘长龙 , 邵明 , 刘延春 , 高铭宣 , 袁世亮 , 张宝岩 . 机器学习方法预测油气产量技术发展现状及前景展望[J]. 特种油气藏, 2025 , 32(4) : 14 -24 . DOI: 10.3969/j.issn.1006-6535.2025.04.002

Abstract

Affected by the complex variability of reservoir physical properties, fluid characteristics, and technological measures during hydrocarbon development, the utilization of production data largely depends on the professional experience of petroleum scientists and engineers, resulting in high computational and time costs. This makes it difficult to meet the demands of efficient modern hydrocarbon reservoir development, necessitating the discovery of more efficient methods for hydrocarbon production prediction. In recent years, machine learning methods represented by deep neural networks, random forest algorithms, and transfer learning have achieved significant application results in hydrocarbon production prediction due to their unique advantages in handling high-dimensional data, capturing long-term dependencies in time-series data, and extracting development data features. This paper analyzes the principles, advantages, and disadvantages of commonly used machine learning methods for hydrocarbon production prediction, elaborates on the current application status of these methods in the field of hydrocarbon production prediction, summarizes potential issues during application, and prospects the development trends of hydrocarbon production prediction methods. In the future, on one hand, research on integrating physical constraints into machine learning models should be strengthened to enhance model interpretability and avoid overly idealized prediction results; on the other hand, algorithms and transfer learning techniques suitable for small-sample scenarios should be developed to fully utilize historical production data, providing better data analysis and technical support for hydrocarbon production prediction. This research has theoretical significance for improving intelligent hydrocarbon production prediction technology.

参考文献

[1] 祝启康,林伯韬,杨光,等.低压低产页岩气井智能生产优化方法[J].石油勘探与开发,2022,49(4):770-777.
ZHU Qikang,LIN Botao,YANG Guang,et al.Intelligent production optimization method for a low pressure and low productivity shale gas well[J].Petroleum Exploration and Development,2022,49(4):770-777.
[2] 刘巍,刘威,谷建伟.基于机器学习方法的油井日产油量预测[J].石油钻采工艺,2020,42(1):70-75.
LIU Wei,LIU Wei,GU Jianwei.Oil production prediction based on a machine learning method[J].Oil Drilling & Production Technology,2020,42(1):70-75.
[3] ARPS J J.Analysis of decline curves[J].Transactions of the AIME,1945,160(1):228-247.
[4] DUONG A N.Rate-decline analysis for fracture-dominated shale reservoirs[C].SPE137748,2014:377-387.
[5] LIANG Bin,LIU Jiang,YOU Junyu,et al.Hydrocarbon production dynamics forecasting using machine learning:a state-of-the-art review[J].Fuel,2023,337:127067.
[6] 谢坤,吴湛奇,李彦阅,等.机器学习在油气开发领域的应用及展望[J].西安石油大学学报(自然科学版),2023,38(5):58-67.
XIE Kun,WU Zhanqi,LI Yanyue,et al.Application of machine learning in oil and gas development field and its prospect[J].Journal of Xi′an Shiyou University(Natural Science Edition),2023,38(5):58-67.
[7] SIRCAR A,YADAV K,RAYAVARAPU K,et al.Application of machine learning and artificial intelligence in oil and gas industry[J].Petroleum Research,2021,6(4):379-391.
[8] 何佑伟,贺质越,汤勇,等.基于机器学习的页岩气井产量评价与预测[J].石油钻采工艺,2021,43(4):518-524.
HE Youwei,HE Zhiyue,TANG Yong,et al.Shale gas well production evaluation and prediction based on machine learning[J].Oil Drilling & Production Technology,2021,43(4):518-524.
[9] 孔鹏,李忠城,杨晨曦,等.基于PCA-NB方法的煤层气井压后产量预测研究[J].特种油气藏,2019,26(6):108-112.
KONG Peng,LI Zhongcheng,YANG Chenxi,et al.Production forecast of fractured coalbed methane wells based on PCA-NB method[J].Special Oil & Gas Reservoirs,2019,26(6):108-112.
[10] LIU Yazhou,ZENG Jianhui,QIAO Juncheng,et al.An advanced prediction model of shale oil production profile based on source-reservoir assemblages and artificial neural networks[J].Applied Energy,2023,333:120604.
[11] 谷建伟,周梅,李志涛,等.基于数据挖掘的长短期记忆网络模型油井产量预测方法[J].特种油气藏,2019,26(2):77-81,131.
GU Jianwei,ZHOU Mei,LI Zhitao,et al.Oil well production forecast with long- short term memory network model based on data mining[J].Special Oil & Gas Reservoirs,2019,26(2):77-81,131.
[12] DUQUE M C M A,CHAVES G S,MONTEIRO D O,et al.Machine learning models to automatically validate petroleum production tests[C].SPE199112-MS,2020:1-5.
[13] 匡立春,刘合,任义丽,等.人工智能在石油勘探开发领域的应用现状与发展趋势[J].石油勘探与开发,2021,48(1):1-11.
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.
[14] LI Hong,YU Haiyang,CAO Nai,et al.Applications of artificial intelligence in oil and gas development[J].Archives of Computational Methods in Engineering,2021,28(3):937-949.
[15] GOODFELLOW I,BENGIO Y,COURVILLE A.Deep learning[M].Cambridge:MIT press,2016:13-52.
[16] LECUN Y,BENGIO Y,HINTON G.Deep learning[J].Nature,2015,521(7553):436-444.
[17] LEE K,LIM J,YOON D,et al.Prediction of shale-gas production at Duvernay Formation using deep-learning algorithm[J].SPE Journal,2019,24(6):2423-2437.
[18] FARGALLA M A M,YAN Wei,DENG Jingen,et al.TimeNet:Time2Vec attention-based CNN-BiGRU neural network for predicting production in shale and sandstone gas reservoirs[J].Energy,2024,290:130184.
[19] ZHANG Lei,DOU Hongen,WANG Hongliang,et al.Neural network optimized by genetic algorithm for predicting single well production in high water cut reservoir[C].IEEE9513395,2021:297-306.
[20] 王金甲,周雅倩,郝智.基于注意力模型的多传感器人类活动识别[J].计量学报,2019,40(6):958-969.
WANG Jinjia,ZHOU Yaqian,HAO Zhi.Multi-sensor human activity recognition based on attention model[J].Acta Metrologica Sinica,2019,40(6):958-969.
[21] ALOMAIR O A,GARROUCH A A.A general regression neural network model offers reliable prediction of CO2 minimum miscibility pressure[J].Journal of Petroleum Exploration and Production Technology,2016,6(3):351-365.
[22] ASANTE J,AMPOMAH W,TU Jiawei,et al.Data-driven modeling for forecasting oil recovery:a timeseries neural network approach for tertiary CO2 WAG EOR[J].Geoenergy Science and Engineering,2024,233:212555.
[23] MAHZARI P,EMAMBAKHSH M,TEMIZEL C,et al.Oil production forecasting using deep learning for shale oil wells under variable gas-oil and water-oil ratios[J].Petroleum Science and Technology,2022,40(4):445-468.
[24] GUO Zixi,ZHAO Jinzhou,YOU Zhenjiang,et al.Prediction of coalbed methane production based on deep learning[J].Energy,2021,230:120847.
[25] 刘合,李艳春,杜庆龙,等.基于多变量时间序列模型的高含水期产量预测方法[J].中国石油大学学报(自然科学版),2023,47(5):103-114.
LIU He,LI Yanchun,DU Qinglong,et al.Prediction of production during high water-cut period based on multivariate time series model[J].Journal of China University of Petroleum(Edition of Natural Sciences),2023,47(5):103-114.
[26] WANG Hongliang,MU Longxin,SHI Fugeng,et al.Production prediction at ultra-high water cut stage via recurrent neural network[J].Petroleum Exploration and Development,2020,47(5):1084-1090.
[27] YANG Ruiyue,QIN Xiaozhou,LIU Wei,et al.A physics-constrained data-driven workflow for predicting coalbed methane well production using artificial neural network[J].SPE Journal,2022,27(3):1531-1552.
[28] LI Xuechen,MA Xinfang,XIAO Fengchao,et al.A physics-constrained long-term production prediction method for multiple fractured wells using deep learning[J].Journal of Petroleum Science and Engineering,2022,217:110844.
[29] 张弘斌,袁奇,赵柄锡,等.采用多通道样本和深度卷积神经网络的轴承故障诊断方法[J].西安交通大学学报,2020,54(8):58-66.
ZHANG Hongbin,YUAN Qi,ZHAO Bingxi,et al.Bearing fault diagnosis with multi-channel sample and deep convolutional neural network[J].Journal of Xi′an Jiaotong University,2020,54(8):58-66.
[30] PAN Shaowei,YANG Bo,WANG Shukai,et al.Oil well production prediction based on CNN-LSTM model with self-attention mechanism[J].Energy,2023,284:128701.
[31] ZHOU Wei,LI Xiangchengzhen,QI Zhongli,et al.A shale gas production prediction model based on masked convolutional neural network[J].Applied Energy,2024,353:122092.
[32] 张蕾,窦宏恩,王天智,等.基于集成时域卷积神经网络模型的水驱油田单井产量预测方法[J].石油勘探与开发,2022,49(5):996-1004.
ZHANG Lei,DOU Hongen,WANG Tianzhi,et al.A production prediction method of single well in water flooding oilfield based on integrated temporal convolutional network model[J].Petroleum Exploration and Development,2022,49(5):996-1004.
[33] SYED F I,ALNAQBI S,MUTHER T,et al.Smart shale gas production performance analysis using machine learning applications[J].Petroleum Research,2022,7(1):21-31.
[34] ZHA Wenshu,LIU Yuping,WAN Yujin,et al.Forecasting monthly gas field production based on the CNN-LSTM model[J].Energy,2022,260:124889.
[35] ZHANG Lei,DOU Hongen,ZHANG Kun,et al.CNN-LSTM model optimized by Bayesian optimization for predicting single-well production in water flooding reservoir[J].Geofluids,2023,2023:5467956.
[36] 李媛,郭大立,康芸玮.融合注意力机制的煤层气产量动态预测[J].科学技术与工程,2023,23(2):550-557.
LI Yuan,GUO Dali,KANG Yunwei.Dynamic prediction of coalbed methane production with attention mechanisms[J].Science Technology and Engineering,2023,23(2):550-557.
[37] ZHOU Guangzhao,GUO Zanquan,SUN Simin,et al.A CNN-BiGRU-AM neural network for AI applications in shale oil production prediction[J].Applied Energy,2023,344:121249.
[38] LYU Zhihan,WANG Nana,LOU Ranran,et al.Towards carbon neutrality:prediction of wave energy based on improved GRU in maritime transportation[J].Applied Energy,2023,331:120394.
[39] NIU Zhaoyang,ZHONG Guoqiang,YU Hui.A review on the attention mechanism of deep learning[J].Neurocomputing,2021,452:48-62.
[40] ZHAO Zhibo,CHEN Yuhua,ZHANG Yi,et al.A deep learning model for predicting the production of coalbed methane considering time,space,and geological features[J].Computers & Geosciences,2023,173:105312.
[41] LI Qi,LI Yaxiong,GAO Shusheng,et al.Well network optimization and recovery evaluation of tight sandstone gas reservoirs[J].Journal of Petroleum Science and Engineering,2021,196:107705.
[42] LI Yong,YANG Jianghao,PAN Zhejun,et al.Nanoscale pore structure and mechanical property analysis of coal:an insight combining AFM and SEM images[J].Fuel,2020,260:116352.
[43] HE Youwei,HE Zhiyue,TANG Yong,et al.Shale gas production evaluation framework based on data-driven models[J].Petroleum Science,2023,20(3):1659-1675.
[44] LUO Guofan,TIAN Yao,SHARMA A,et al.Eagle ford well insights using data-driven approaches[C].IPTC19260-MS,2019:1-10.
[45] YU Zhichao,WANG Zhizhang,JIANG Qingping,et al.Analysis of factors of productivity of tight conglomerate reservoirs based on random forest algorithm[J].ACS Omega,2022,7(23):20390-20404.
[46] LI Hangyu,GONG Changping,LIU Shuyang,et al.Machine learning-assisted prediction of oil production and CO2 storage effect in CO2-water-alternating-gas injection(CO2-WAG)[J].Applied Sciences,2022,12(21):10958.
[47] 李文倚,侯明雨,全航,等.一种基于知识图谱和随机森林算法的致密气井产能预测方法[J].特种油气藏,2024,31(5):77-84.
LI Wenyi,HOU Mingyu,QUAN Hang,et al.A productivity prediction method for tight gas wells based on knowledge graph and random forest algorithm[J].Special Oil & Gas Reservoirs,2024,31(5):77-84.
[48] XUE Liang,LIU Yuetian,XIONG Yifei,et al.A data-driven shale gas production forecasting method based on the multi-objective random forest regression[J].Journal of Petroleum Science and Engineering,2021,196:107801.
[49] ZHU Jie,ZHAO Yuhan,HU Qiujia,et al.Coalbed methane production model based on random forests optimized by a genetic algorithm[J].ACS Omega,2022,7(15):13083-13094.
[50] LONG Mingsheng,WANG Jianmin,DING Guiguang,et al.Transfer feature learning with joint distribution adaptation[C].IEEE6751384,2013:2200-2207.
[51] WEISS K,KHOSHGOFTAAR T M,WANG Dingding.A survey of transfer learning[J].Journal of Big Data,2016,3:1-40.
[52] MA Yuchi,CHEN Shuo,ERMON S,et al.Transfer learning in environmental remote sensing[J].Remote Sensing of Environment,2024,301:113924.
[53] KHEDDAR H,HIMEUR Y,AL-MAADEED S,et al.Deep transfer learning for automatic speech recognition:towards better generalization[J].Knowledge-Based Systems,2023,277:110851.
[54] LOTFOLLAHI M,NAGHIPOURFAR M,LUECKEN M D,et al.Mapping single cell data to reference atlases by transfer learning[J].Nature biotechnology,2022,40(1):121-130.
[55] 刘合,李艳春,贾德利,等.人工智能在注水开发方案精细化调整中的应用现状及展望[J].石油学报,2023,44(9):1574-1586.
LIU He,LI Yanchun,JIA Dеli,et al.Application status and prospects of artificial intelligence in the refinement of waterflooding development program[J].Acta Petrolei Sinica,2023,44(9):1574-1586.
[56] NIU Wente,SUN Yuping,ZHANG Xiaowei,et al.An ensemble transfer learning strategy for production prediction of shale gas wells[J].Energy,2023,275:127443.
[57] 姜春雷,方硕,刘伟,等.基于卷积神经网络和迁移学习的特高含水油井生产预测[J].中国石油大学学报(自然科学版),2023,47(6):162-170.
JIANG Chunlei,FANG Shuo,LIU Wei,et al.Production prediction of extra high water cut oil well based on convolution neural network and transfer learning[J].Journal of China University of Petroleum(Edition of Natural Sciences),2023,47(6):162-170.
[58] DU Shuyi,WANG Meizhu,YANG Jiaosheng,et al.An enhanced prediction framework for coalbed methane production incorporating deep learning and transfer learning[J].Energy,2023,282:128877.
[59] CORNELIO J,MOHD R S,JAHANDIDEH A,et al.Investigating transfer learning for characterization and performance prediction in unconventional reservoirs[C].SPE204563-MS,2021:12-16.
[60] 田建华,朱博华,卢志强,等.基于井控多属性机器学习的缝洞型储层预测方法[J].油气地质与采收率,2023,30(1):86-92.
TIAN Jianhua,ZHU Bohua,LU Zhiqiang,et al.Fracture-cavity reservoir prediction based on well-controlled multi-attribute machine learning[J].Petroleum Geology and Recovery Efficiency,2023,30(1):86-92.
[61] 盖建.基于自动机器学习的采油井压裂效果预测方法[J].油气地质与采收率,2023,30(1):161-170.
GAI Jian.Prediction method for hydraulic fracturing effect of oil production well based on automatic machine learning technology[J].Petroleum Geology and Recovery Efficiency,2023,30(1):161-170.
[62] 陈林,黎棚武,张绍俊,等. 基于机器学习的岩芯渗透率及裂缝开度预测[J]. 西南石油大学学报(自然科学版),2023,45(4):155-163.
CHEN Lin,LI Pengwu,ZHANG Shaojun,et al.Prediction method of core permeability and fracture aperture based on machine learning[J]. Journal of Southwest Petroleum University (Science & Technology Edition), 2023, 45(4):155-163.
[63] 朱博含,单玄龙,衣健,等.基于机器学习的火山岩识别方法及应用[J].特种油气藏,2024,31(5):41-49.
ZHU Bohan,SHAN Xuanlong,YI Jian,et al.Volcanic rock identification method based on machine learning and its application[J].Special Oil & Gas Reservoirs,2024,31(5):41-49.
[64] 柳洁,田冷,刘士鑫,等. 基于复合机器算法的致密气井产能预测模型[J].大庆石油地质与开发,2024,43(5):69-78.
LIU Jie,TIAN Leng,LIU Shixin,et al.Prediction model of tight gas well productivity based on composite machine algorithm:taking Block SM of Ordos Basin as an example[J].Petroleum Geology & Oilfield Development in Daqing,2024,43(5):69-78.
[65] 庞兰苏,王杨,蒋薇,等.基于机器学习的短生产周期碳酸盐岩气井产量预测研究[J]. 特种油气藏,2023,30(2):134-141.
PANG Lansu,WANG Yang,JIANG Wei,et al.Research on yield prediction of carbonatite gas well with a short production cycle based on machine learning[J].Special Oil & Gas Reservoirs,2023,30(2):134-141.
[66] 蒋文超.基于机器学习与模型融合的大庆油田SN区块油井压裂效果预测技术[J].大庆石油地质与开发,2023,42(1):64-72.
JIANG Wenchao.Prediction model for production well hydraulic fracturing effect of Block SN in Daqing Oilfield based on machine learning and model ensemble[J].Petroleum Geology & Oilfield Development in Daqing,2023,42(1):64-72.
文章导航

/