程翊珊,李治平,许龙飞,史华
CHENG Yishan,LI Zhiping,XU Longfei,SHI Hua

• 1:中国地质大学(北京)能源学院
• 2:非常规天然气地质评价及开发北京市重点实验室
• 3:中国石油长庆油田分公司油气工艺研究院
• 4:低渗透油气田勘探开发国家工程实验室

摘要(Abstract)：

油田注水开采过程中注入水与地层水不配伍导致无机积垢阻塞储层和管道,预测油层无机积垢对稳定油井长期正常生产具有积极意义。以影响积垢因素作为输入变量,设计积垢等级概念作为输出变量,建立了预测油层无机积垢的BP神经网络模型,对实际结垢情况进行预测并控制变量分析环境因素对积垢结果的影响。结果表明:神经网络预测结果与实际实验浊度算得的积垢等级相符,配伍性良好的2种地层水混合时预测积垢等级均小于0.1,不配伍的2种地层水混合时预测积垢等级均大于10;神经网络对各影响因素的预测与积垢理论相符,积垢趋势与积垢离子浓度、温度、pH值正相关,积垢趋势与压力负相关。应用BP神经网络模型预测油层无机积垢具有预测精度高、计算速度快、需要资料少等特点,为油层无机积垢预测提供了一种高效快捷的方法。
In the process of water-flooded oilfield exploitation, the inorganic scale caused by the incompatibility between the injected water and formation water will result in the blocking of reservoirs and pipelines, so the prediction of inorganic scaling in the reservoir is of great significance for the long-term normal production of oil wells. Taking the scaling influencing factors as the input variables and the designed concept of scale grade as the output variable, a BP neural network model for predicting the inorganic scaling in the reservoir was established, which can predict the actual scaling and control the variables to analyze the influences of environmental factors on the scaled results. The results show that the predicted results by the neural network are coincident with the actual experimental scaling grade calculated by the turbidity, and their predicted scale grades are both less than 0.1 when the two types of formation water with good compatibility are mixed, while they are higher than 10 when two types of the formation water without compatibility are mixed. The prediction of neural network for each influencing factor is consistent with the scaling theory, and the scale trend is positively correlated with the fouling ion concentration, temperature and pH value, while negatively correlated with the pressure. The inorganic scaling predication of the reservoir with BP neural network model is characterized by high precision, rapid calculation speed and less data, providing a high-efficiency and rapid prediction method for inorganic scaling prediction in the reservoir.

关键词(KeyWords)： 油层无机积垢;BP神经网络;影响因素;预测积垢等级;地层水
inorganic scaling in the reservoir;BP neural network;the prediction of influencing factor;scaling grade;formation water

Abstract:

Keywords:

基金项目(Foundation): 国家科技重大专项“致密油气藏数值模拟新方法与开发设计”(2017ZX05009-005)

作者(Author): 程翊珊,李治平,许龙飞,史华
CHENG Yishan,LI Zhiping,XU Longfei,SHI Hua

DOI: 10.19597/j.issn.1000-3754.202005034

参考文献(References)：

• [1]高辉，朱耿博仑，王泫懿，等.鄂尔多斯盆地延长组致密砂岩储层微观孔喉特征差异及其成因[J].石油与天然气地质，2019,40(2):302-312.GAO Hui,ZHUGENG Bolun,WANG Xuanyi,et al.Differences and origin of micro-pore throat characteristics for tight sandstone reservoir of Yanchang Formation,Ordos Basin[J].Oil & Gas Geology,2019,40(2):302-312.
• [2]李承龙，苗志国，李照永，等.大庆长垣外围特低渗透油藏开发效果综合评价方法[J].特种油气藏，2019,26(4):96-102.LI Chenglong,MIAO Zhiguo,LI Zhaoyong,et al.A new comprehensive evaluation of development performance for the peripheral ultra-low permeability oil reservoirs in Changyuan of Daqing Oilfield [J].Special Oil & Gas Reservoirs,2019,26(4):96-102.
• [3]吕文雅，苗凤彬，张本键，等.四川盆地剑阁地区须家河组致密砾岩储层裂缝特征及对天然气产能的影响[J].石油与天然气地质，2020,41(3):484-491.Lü Wenya,MIAO Fengbin,ZHANG Benjian,et al.Fracture characteristics and their influence on natural gas production:A case study of the tight conglomerate reservoir in the Upper Triassic Xujiahe Formation in Jiange area,Sichuan Basin[J].Oil & Gas Geology,2020,41(3):484-491.
• [4]安永生，周大可，欧阳铁兵，等.基于热力学模型的油井井筒析蜡规律[J].断块油气田，2019,26(5):649-652.AN Yongsheng,ZHOU Dake,OUYANG Tiebing,et al.Research on patterns of wax precipitation in wellbore based on thermodynamic model[J].Fault-Block Oil & Gas Field,2019,26(5):649-652.
• [5]黄亮.松辽盆地龙西地区扶余油层特低渗透储层孔隙结构[J].大庆石油地质与开发，2019,38(4):15-21.HUANG Liang.Pore structure of Fuyu ultra-low-permeability oil reservoir in Longxi area of Songliao Basin[J].Petroleum Geology & Oilfield Development in Daqing,2019,38(4):15-21.
• [6]王朋，孙灵辉，王核，等.库车坳陷下侏罗统阿合组致密砂岩储层孔隙微观结构特征及其对致密气富集的控制作用[J].石油与天然气地质，2020,41(2):295-304.WANG Peng,SUN Linghui,WANG He,et al.Microscopic pore structure of the tight sand gas reservoirs of the Low Jurassic in Kuqa Depression and its controls on tight gas enrichment[J].Oil & Gas Geology,2020,41(2):295-304.
• [7]LANGELIER W F.The analytical control of anti-corrosion water treatment[J].Journal-American Water Works Association,1936,28(10):1500-1521.
• [8]王琪，刘江红，韩露，等.油田主要结垢机理及结垢预测模型研究进展[J].长春理工大学学报(自然科学版),2016,39(1):129-133.WANG Qi,LIU Jianghong,HAN Lu,et al.Research progress of the main fouling mechanism and fouling prediction model in oilfield[J].Journal of Changchun University of Science and Technology,2016,39(1):129-133.
• [9]崔付义，方颖，杨明，等.胜利油田纯九区注水开发过程中无机结垢趋势预测[J].地学前缘，2012,19(4):301-306.CUI Fuyi,FANG Ying,YANG Ming,et al.Prediction of inorganic scaling tendency due to injection of incompatible water in Chunjiu area of Shengli Oilfield[J].Earth Science Frontiers,2012,19(4):301-306.
• [10]尹先清，伍家忠，王正良.油田注入水碳酸钙垢结垢机理分析与结垢预测[J].石油勘探与开发，2002,29(3):85-87.YIN Xianqing,WU Jiazhong,WANG Zhengliang.Analysis and prediction of scaling mechanism of the CaCO3 from oilfield injection waters[J].Petroleum Exploration and Development,2002,29(3):85-87.
• [11]肖曾利，蒲春生，时宇，等.油田水无机结垢及预测技术研究进展[J].断块油气田，2004,11(6):76-78.XIAO Zengli,PU Chunsheng,SHI Yu,et al.Review on inorganic scaling and prediction of field water[J].Fault-Block Oil & Gas Field,2004,11(6):76-78.
• [12]游靖，王志坤，余吉良，等.华北油田地层水与注入水结垢趋势预测及影响因素[J].岩性油气藏，2017,29(4):162-168.YOU Jing,WANG Zhikun,YU Jiliang,et al.Scaling tendency prediction and influencing factors of formation and injected water in Huabei Oilfield[J].Lithologic Reservoirs,2017,29(4):162-168.
• [13]朱达江，孙洪亮，周峰，等.川东石炭系气藏气井结垢预测及防治[J].断块油气田，2012,19(6):792-795.ZHU Dajiang,SUN Hongliang,ZHOU Feng,et al.Prediction and prevention on scaling of gas wells in Carboniferous gas reservoirs in east Sichuan[J].Fault-Block Oil & Gas Field,2012,19(6):792-795.
• [14]YEBOAH Y D,SAMUAH S K,SAEED M R.Prediction of carbonate and sulfate scales in oilfields[R].SPE 25586,1993.
• [15]崔刚，沈建军，唐洪明，等.QK18-1油田注水开发储层损害研究[J].油田化学，2017,34(3):456-462.CUI Gang,SHEN Jianjun,TANG Hongming,et al.Research on reservoir damage of water flooding in QK18-1 Oilfield[J].Oilfield Chemistry,2017,34(3):456-462.
• [16]罗明良，蒲春生，王得智，等.油水井近井带无机结垢动态预测数学模型[J].石油学报，2002,23(1):61-66.LUO Mingliang,PU Chunsheng,WANG Dezhi,et al .Mathematics model on inorganic scale dynamic prediction near well-bore formation[J].Acta Petrolei Sinica,2002,23(1):61-66.
• [17]ODDO J E,TOMSON M B.Why scale forms and how to predict[R].SPE 21710,1994.
• [18]李沁，伊向艺，卢渊，等.镇泾油田结垢原因分析及趋势预测[J].石油地质与工程，2009,23(4):128-130.LI Qin,YI Xiangyi,LU Yuan,et al.Cause analysis and trend forecast of scaling in Zhenjing Oilfield[J].Petroleum Geology and Engineering,2009,23(4):128-130.
• [19]蒋建方，黄勇，熊婷，等.R语言在特低渗透孔隙结构与可动流体关系研究中的应用[J].大庆石油地质与开发，2019,38(3):46-53.JIANG Jianfang,HUANG Yong,XIONG Ting,et al.Application of R language in studying the relationship between the super-low permeability pore structure and movable fluid[J].Petroleum Geology & Oilfield Development in Daqing,2019,38 (3):46-53.
• [20]王量，买皓，李勇.基于在线字典学习算法的地震数据去噪应用[J].断块油气田，2019,26(2):177-180.WANG Liang,MAI Hao,LI Yong.Application of seismic data denoising based on online dictionary learning algorithm[J].Fault-Block Oil & Gas Field,2019,26(2):177-180.
• [21]许玲，陈德民，赖枫鹏.气井新井产能预测的神经网络方法研究[J].油气井测试，2008,17(3):12-14.XU Ling,CHEN Demin,LAI Fengpeng,et al.Study on method of nervous network about productivity forecast in new gas well[J].Well Testing,2008,17(3):12-14.
• [22]石创，朱俊章，龙祖烈，等.基于概率神经网络的烃源岩TOC预测：以珠江口盆地陆丰南区为例[J].断块油气田，2019,26(5):561-565.SHI Chuang,ZHU Junzhang,LONG Zulie,et al.Prediction of total organic carbon in source rocks by probabilistic neural network:A case study of southern Lufeng area in Pearl River Mouth Basin[J].Fault-Block Oil & Gas Field,2019,26(5):561-565.
• [23]侯春华.基于长短期记忆神经网络的油田新井产油量预测方法[J].油气地质与采收率，2019,26(3):105-110.HOU Chunhua.New well oil production forecast method based on long-term and short-term memory neural network[J].Petroleum Geology and Recovery Efficiency,2019,26(3):105-110.
• [24]巨晓昕.定边采油厂多层水混合注入结垢及防垢技术研究[D].西安：西安石油大学，2018.JU Xiaoxin.Study on scale inbibition and scale prevention technology in multi-layered water of Dingbian oil production plant[D].Xi’an:Xi’an Shiyou University,2018.
• [25]但松林，刘尚奇，罗艳艳，等.基于BP神经网络预测高含水层对 SAGD 开发效果的影响[J].大庆石油地质与开发，2019,38(2):73-80.DAN Songlin,LIU Shangqi,LUO Yanyan,et al.Predicted SAGD development effects by BP neural network for the high-watercut reservoir[J].Petroleum Geology & Oilfield Development in Daqing,2019,38(2):73-80.