卢艳,马宏宇,齐云峰,程梦薇,刘宇轩,王雪萍
LU Yan,MA Hongyu,QI Yunfeng,CHENG Mengwei,LIU Yuxuan,WANG Xueping

• 1:中国石油大庆油田有限责任公司勘探开发研究院

摘要(Abstract)：

特高含水后期地下流体性质越来越复杂，对测井曲线的影响也越来越大，针对传统解释方法测井曲线利用率低、解释符合率不能满足生产需求、解释效率低的问题，基于大数据分析技术提出“知识+数据”双驱动的水淹层智能测井评价技术。通过多源数据融合、异常曲线处理、自然电位（V_(sp)）泥岩基线构建等技术，形成了大数据分析预处理方法；创新采用专家经验、测井专业知识与数据相融合，通过测井机理指导特征工程，形成多维特征表征技术；优选10余种机器学习算法，构建水淹层识别模型，实现了水淹层智能评价。实际应用表明，新方法在23口密闭取心井中得到验证，储层参数预测精度得到显著提高，孔隙度平均相对误差为5.76%，目前含水饱和度平均绝对误差为7.03%。和以往方法相比储层参数的预测精度显著提高，孔隙度平均相对误差降低了2百分点，目前含水饱和度平均绝对误差降低了1百分点。薄层水淹级别符合率由常规解释方法的70.0%提高到77.1%；厚层水淹级别符合率由常规解释方法的75.0%提高到81.4%。研究成果为射孔及调剖措施提供依据，为剩余油评价及进一步提质提效挖潜提供了有力的技术支持。
In late stage of ultra-high water cut, underground fluid properties become increasingly complex, exerting a growing influence on well logging curves. In view of the problems of the traditional interpretation method including low utilization rate of well logging curves, unsatisfactory interpretation coincidence rate for production requirements and low interpretation efficiency, a “knowledge+data” dual driven intelligent logging evaluation technique for water flooded zone is proposed based on big-data analysis. Through techniques such as multi-sources data fusion, abnormal curves processing and spontaneous potential(V_(SP)) mudstone baseline construction, a preprocessing method for big data analysis is established. By innovatively integrating the expert experience, well logging professional knowledge and data, a multi-dimensional feature characterizing technique is developed based on feature engineering guided by logging mechanisms. Over 10 machine learning algorithms are selected to build a water-flooded reservoir identification model, achieving intelligent evaluation for water-flooded reservoirs. Actual application results indicate that the new method, validated in 23 sealed coring wells, significantly improves the prediction accuracy of reservoir parameters, with an average relative error of 5.76% for porosity and an average absolute error of 7.03% for current water saturation. Compared to previous methods, the prediction accuracy of reservoir parameters has been significantly improved, with average relative error of porosity decreased by 2 percentage points and current average absolute error of water saturation decreased by 1 percentage point. The coincidence rate of thin-layer water flooded grades increases from 70.0% with the traditional interpretation method to 77.1%, while the coincidence rate of thick-layer water flooded grades increases from 75.0% with the traditional interpretation method to 81.4%. The research provides a basis for the perforation and profile adjustment measures, offering strong technical support for remaining oil evaluation, further quality and efficiency improvements and potential tapping.

关键词(KeyWords)： 特高含水后期;水淹层智能解释;大数据分析;机器学习算法;参数预测
late stage of ultra-high watercut;intelligent interpretation of water-flooded reservoir;big data analysis;machine learning algorithm;parameter prediction

Abstract:

Keywords:

基金项目(Foundation): 中国石油天然气集团有限公司“十三五”科技开发基金项目“高-特高含水油田改善水驱效果关键技术”(2019B-1209)

作者(Author): 卢艳,马宏宇,齐云峰,程梦薇,刘宇轩,王雪萍
LU Yan,MA Hongyu,QI Yunfeng,CHENG Mengwei,LIU Yuxuan,WANG Xueping

DOI: 10.19597/J.ISSN.1000-3754.202504056

参考文献(References)：

• [1]魏玉梅，吴淑艳，李海燕，等.基于密闭取心资料的高含水油藏水淹层测井定量评价方法[J].工程地球物理学报，2019,16(1):1-6.WEI Yumei, WU Shuyan, LI Haiyan, et al. Discussion on quantitative evaluation method for logging of water-flooded layer in high water-containing reservoir based on closed coring data[J].Chinese Journal of Engineering Geophysics, 2019, 16(1):1-6.
• [2]卢艳，刘传平，杨青山，等.单井水淹层测井解释系统厚度划分方法研究[J].大庆石油地质与开发，2007,26(4),127-129.LU Yan,LIU Chuanping,YANG Qingshan,et al. Research on the thickness classification method of log interpretation system of water flooded zone in individual well[J]. Petroleum Geology&Oilfield Development in Daqing,2007,26(4),127-129.
• [3]韩志明，王宪成，李胜军.高含水期水淹层解释方法研究[J].测井技术，2000,24(5):333-336.HAN Zhiming,WANG Xiancheng,LI Shengjun. Study on waterflooded formation interpretation at high water-cut stage[J].Well Logging Technology,2000,24(5):333-336.
• [4]闫伟林，殷树军，马宏宇，等.大庆油田裸眼井测井技术进展与展望[J].大庆石油地质与开发，2024, 43(3):109-118.YAN Weilin, YIN Shujun, MA Hongyu, et al. Progress and prospects of open-hole well logging technology in Daqing Oilfield[J]. Petroleum Geology&Oilfield Development in Daqing,2024,43(3):109-118.
• [5]杨波，黄长兵，何岩，等.一种基于PSO-ELM的低渗透砂岩水淹层测井识别方法[J].断块油气田，2024,31(4):645-651.YANG Bo,HUANG Changbing,HE Yan,et al. A logging identification method of water-flooded layer of low permeability sandstone based on PSO-ELM[J]. Fault-Block Oil&Gas Field,2024,31(4):645-651.
• [6]王跃祥，赵佐安，唐玉林，等.基于知识驱动图版约束的致密砂岩气储层测井参数智能预测[J].天然气工业，2024,44(9):68-76.WANG Yuexiang,ZHAO Zuo’an,TANG Yulin,et al. Intelligent prediction of logging parameters of tight sandstone gas reservoirs based on knowledge-driven chart constraints[J]. Natural Gas Industry,2024,44(9):68-76.
• [7]段泽英，蔡贤明，滕卫卫，等.大数据分析技术在油田生产中的研究与应用[J].中国管理信息化，2015,18(18):64-65.DUAN Zeying,CAI Xianming,TENG Weiwei,et al. Research and application of big data analysis technology in oil production[J]. China Management Informationization, 2015, 18(18):64-65.
• [8]孙晓萍.大数据分析技术在油田生产中的应用研究[J].中国管理信息化，2019,22(9):171-172.SUN Xiaoping. Research on the application of big data analysis technology in oilfield production[J]. China Management Informationization,2019,22(9):171-172.
• [9]董文涛.大数据技术在未来石油企业中的应用探讨[J].中国化工贸易，2019,11(13):119.DONG Wentao. Application of big data technology in future petroleum enterprises[J]. China Chemical Trade, 2019, 11(13):119.
• [10]杨世刚.高含水期粉砂岩储层水淹层解释方法研究[J].江汉石油学院学报，2003,25(4):63-65.YANG Shigang. Water-out reservoir interpretation method in siltstone reservoirs at high watercut stage[J] Journal of Jianghan Petroleum Institute,2003,25(4):63-65.
• [11]刘传平，杨青山，杨景强，等.薄差层水淹层测井解释技术研究[J].大庆石油地质与开发，2004,23(5):118-120.LIU Chuanping, YANG Qingshan, YANG Jingqiang, et al.Well logging interpretation technique in thin and poor watered-out pay zone[J]. Petroleum Geology&Oilfield Development in Daqing,2004,23(5):118-120.
• [12]苏瑞.基于随机森林的F区块水淹层测井定性识别[D].大庆：东北石油大学，2019.SU Rui. Qualitative identification of flooded layers in F block based on random forest[D]. Daqing:Northeast Petroleum University,2019.
• [13]王如意，吴钧，杨向同，等.基于机器学习智能决策系统的古龙页岩油储层总有机碳含量定量表征及智能预测[J].大庆石油地质与开发，2022,41(3):172-182.WANG Ruyi, WU Jun, YANG Xiangtong, et al. Quantitative characterization and intelligent prediction of TOC of Gulong shale oil reservoir based on machine learning intelligent decision system[J]. Petroleum Geology&Oilfield Development in Daqing,2022,41(3):172-182.
• [14]李奎周，王团，赵海波，等.基于深度前馈神经网络的致密砂岩储层孔隙度预测[J].大庆石油地质与开发，2023,42(5):140-146.LI Kuizhou,WANG Tuan,ZHAO Haibo,et al. Pore permeability prediction of tight sandstone reservoir based on deep feedforward neural network[J]. Petroleum Geology&Oilfield Development in Daqing,2023,42(5):140-146.
• [15] GAO S Y,YAN W L,YIN S J. Study on a corrected calculation method of original oil saturation[C].Harbin:2nd International Conference on Applied Mathematics, Modelling and Intelligent Computing,2022.
• [16] MA H Y,XIAO L M,DING J W,et al. Prediction method of water cut in Changyuan transition zone oil-water layers based on correction of influencing factors[C]. Chengdu:12th Internation Filed Exploration and Development Conference,2022.
• [17] ANDERSON W G. Wettability literature survey-part 6:The effects of wettability on waterflooding[J]. Journal of Petroleum Technology,1987,39(12):1605-1622.
• [18] BOWERS M C,FITZ D E. A probabilistic approach to determine uncertainty in calculated water saturation[C]. Dallas:SPWLA41st Annual Logging Symposium,2000.
• [19] DE KUIJPER A,SANDOR R K J,HOFMAN J P,et al. Electrical conductivities in oil-bearing shaly sand accurately described with the satori saturation model[J]. Log Analyst, 1996, 37(5):22-32.