大庆油田杏南开发区低效循环带测井识别方法研究
|
摘要
注水低效循环已经成为油田开发后期最主要问题之一。大庆油田已经进入到高含水期开采阶段,长期持续注水导致油层早期水淹、水窜,从而降低注入水的波及面积,造成低效或无效循环。低效循环问题是油田开发中最近几年才出现的,目前还没有引起国内外的广泛重视。杏南开发区已经进入到高含水和特高含水开发期。检查井岩芯分析、室内物模试验和现场动态监测资料均表明,以厚油层为主的非均质多油层砂岩油田综合含水上升到90%以后,存在着严重的无效和低效循环问题。大量的注入水沿高渗透、高含水的便捷通道(简称“大孔道”)无效或低效循环。低效循环的存在,导致油田采收率低、生产成本上升,开发效益下降,给油田开发带来巨大压力,因此低效循环带识别技术的研究日益紧迫。
对低效循环带的识别首先要在单井上找到大孔道部位,进而在平面上追溯低效循环带。本研究首先依据大孔道形成机理,利用其微观变化在测井曲线上的反映,找到大孔道与油层水淹特征相区别的电测曲线特征,即:自然电位上升、自然伽玛降低、声波时差增大、三侧向电阻率降低、井径扩大、微电极降低。认识到“特高水淹油层”的电测特征有可能就是大孔道的电测特征。建立具有地区特征的解释模型,计算大孔道试验区储层参数,并分析这些参数在大孔道位置的变化特征,即:泥质含量下降、含水饱和度大幅度增加、含油饱和度大幅降低、渗透率和孔隙度增大、岩石骨架胶结强度降低、粒度中值变大;然后制定出大孔道判别的标准,采用大孔道参数法和BP神经网络法对研究区域大孔道进行判别;最后,综合地质和动态资料,利用现有井的资料分析油水井以及邻井之间的注采和连通关系,再利用沉积相和高渗透带进行约束,进行低效循环带平面展布,简化并提高低效循环带识别技术。
The problem of low effective circulation channel has become one of the most important problems in the later stage of oilfield development. Daqing Oilfield has entered a period of high water cut exploitation stage, the long-term sustainability of reservoir water is leading to the early flooding, water channeling, thereby is reducing the spread of water into the area, so that has resulted in inefficient or ineffective circulation. Low effective circulation is found at the only oilfield development in recent years, has not attracted widespread attention at home and abroad. Southern Xingshugang district has entered a high water cut or super-high water cut development phase. Core analysis of inspecting well, indoor objects dynamic modulus test and field monitoring data show that inhomogeneous multi-oil layer sandstone oilfields mainly to the thick oil layer rise to 90% comprehensive saturation, there is a serious problem of ineffective and inefficient circulation. Inject water is circulated ineffectively or inefficiently in the large number of high permeability, high water cut a convenient channel ("The large pore path"). Low effective circulation channel leads to a low recovery rate, a rising production cost and a low efficiency, brings enormous pressure to the oilfield development. Therefore identifying low effective circulation channel is increasingly significant.
Identifying low effective circulation channel firstly should find the large pore path in a single well log, and then bring back low effective circulation channel in the plane. Firstly the thesis studies the formation mechanism of the large pore path. Based on its microscopic changes in the logging curves, the big differences between the characteristics of the large pore path and reservoir flooded electric log characteristics are found, namely: spontaneous-potential log curve (SP) increases, natural gamma ray log curve (GR) decreases, acoustic log curve (AC) increases, lateral resisitivity log curve (RLLD/RLLS) decreases, caliper log curve (CAL) expanded, minilog curve decreases. It is recognized that the electric characteristics of super-high water cut reservoir may have great sounding features of the large pore paths. Establish interpretation model with regional characteristic in the experimental areas, we calculate reservoir parameters. The characteristics of these parameters in the large pore path are analyzed, namely: shale content, saturation of oil and the cementation intensity of the rock framework reduce; saturation of water, porosity, permeability and median grain diameter decrease. The large pore path discrimination standards are established, we identify the large pore path using a large pore path parameters method and BP neural network in the exploit region. Finally, Comprehensive geological and dynamic data, the analysis of existing oil wells and adjacent and connected relations between the injection and production wells are studied, we bring back low effective circulation channel in the plane using sedimentary facies and high permeability channels. Technology of identifying low effective circulation channel is simplified and improved.
对低效循环带的识别首先要在单井上找到大孔道部位,进而在平面上追溯低效循环带。本研究首先依据大孔道形成机理,利用其微观变化在测井曲线上的反映,找到大孔道与油层水淹特征相区别的电测曲线特征,即:自然电位上升、自然伽玛降低、声波时差增大、三侧向电阻率降低、井径扩大、微电极降低。认识到“特高水淹油层”的电测特征有可能就是大孔道的电测特征。建立具有地区特征的解释模型,计算大孔道试验区储层参数,并分析这些参数在大孔道位置的变化特征,即:泥质含量下降、含水饱和度大幅度增加、含油饱和度大幅降低、渗透率和孔隙度增大、岩石骨架胶结强度降低、粒度中值变大;然后制定出大孔道判别的标准,采用大孔道参数法和BP神经网络法对研究区域大孔道进行判别;最后,综合地质和动态资料,利用现有井的资料分析油水井以及邻井之间的注采和连通关系,再利用沉积相和高渗透带进行约束,进行低效循环带平面展布,简化并提高低效循环带识别技术。
The problem of low effective circulation channel has become one of the most important problems in the later stage of oilfield development. Daqing Oilfield has entered a period of high water cut exploitation stage, the long-term sustainability of reservoir water is leading to the early flooding, water channeling, thereby is reducing the spread of water into the area, so that has resulted in inefficient or ineffective circulation. Low effective circulation is found at the only oilfield development in recent years, has not attracted widespread attention at home and abroad. Southern Xingshugang district has entered a high water cut or super-high water cut development phase. Core analysis of inspecting well, indoor objects dynamic modulus test and field monitoring data show that inhomogeneous multi-oil layer sandstone oilfields mainly to the thick oil layer rise to 90% comprehensive saturation, there is a serious problem of ineffective and inefficient circulation. Inject water is circulated ineffectively or inefficiently in the large number of high permeability, high water cut a convenient channel ("The large pore path"). Low effective circulation channel leads to a low recovery rate, a rising production cost and a low efficiency, brings enormous pressure to the oilfield development. Therefore identifying low effective circulation channel is increasingly significant.
Identifying low effective circulation channel firstly should find the large pore path in a single well log, and then bring back low effective circulation channel in the plane. Firstly the thesis studies the formation mechanism of the large pore path. Based on its microscopic changes in the logging curves, the big differences between the characteristics of the large pore path and reservoir flooded electric log characteristics are found, namely: spontaneous-potential log curve (SP) increases, natural gamma ray log curve (GR) decreases, acoustic log curve (AC) increases, lateral resisitivity log curve (RLLD/RLLS) decreases, caliper log curve (CAL) expanded, minilog curve decreases. It is recognized that the electric characteristics of super-high water cut reservoir may have great sounding features of the large pore paths. Establish interpretation model with regional characteristic in the experimental areas, we calculate reservoir parameters. The characteristics of these parameters in the large pore path are analyzed, namely: shale content, saturation of oil and the cementation intensity of the rock framework reduce; saturation of water, porosity, permeability and median grain diameter decrease. The large pore path discrimination standards are established, we identify the large pore path using a large pore path parameters method and BP neural network in the exploit region. Finally, Comprehensive geological and dynamic data, the analysis of existing oil wells and adjacent and connected relations between the injection and production wells are studied, we bring back low effective circulation channel in the plane using sedimentary facies and high permeability channels. Technology of identifying low effective circulation channel is simplified and improved.
引文
[1]王学忠.孤东油田高含水期井间大孔道特征研究.中国西部油气地质,2006,2(1):101-105.
[2]韩镇石,具海啸,刘刚,等.应用测井资料识别喇嘛甸油田大孔道.大庆石油地质与开发,2006,25(4):72-74.
[3]宋考平,杨秋荣,付青春,等.模糊综合评判方法判断低效循环井层.钻井工业,2006,29(4):35-37.
[4]王学忠.井间大孔道预测方法研究.石油知识勘探开发,2005(1):18-19.
[5]尹文军,陈永生,王华,等.水力探测大孔道和剩余油饱和度解释模型的建立.油气地质与采收率.2005,12(1):63-65.
[6]金山.“大孔道”测试方法的优化.国外测井技术,2004,19(5):61-64.
[7]刘月田,孙保利,于永生.大孔道模糊识别与定量计算方法。石油钻采工艺,2003,25(5):54-59.
[8]史有刚,曾庆辉,周晓俊.大孔道试井理论解释模型。石油钻采工艺,2003,25(3):48-50.
[9]曾流芳,陈柏平,王学忠.疏松砂岩油藏大孔道定量描述初步研究.油气地质与采收率.2002,9(4):53-54.
[10]曾流芳,赵国景,张子海,等.疏松砂岩油藏大孔道形成机理及判别方法.应用基础与工程科学学报.2002,10(3):268-275.
[11]窦之林,曾流芳.大孔道诊断和描述技术研究.石油勘探与开发,2001,28(1):75-77.
[12]何长,李平,汪正勇,等.大孔道的表现特征及调剖对策.石油钻采艺,2000,22(5):63-66
[13]商志英,万新德,何长虹,等.应用水力探测方法确定储层大孔道及剩余油分布状况的研究。大庆石油地质与开发,2004,23(2):30-32.
[14]陈伟,马如雄,郝颜红.基于MATLAB的BP人工神经网络设计.电脑学习,2005,4(2):30-31.
[15]雍世和,孙宝佃.用滑动平均滤波法消除测井曲线上的毛刺干扰.华东石油学院学报,1983(1):11-19.
[16]雍世和。最优化测井解释.石油大学出版社,1995年5月第1版.
[17]雍世和,张超谟.测井数据处理与综合解释.石油大学出版社,1996年9月第1版.
[18]雍世和,孙宝佃.用计算机对放射性测井曲线进行环境影响校正.测井技术,1984年第4期.
[19]雍世和,靳秀英.用计算机对双感应-浅聚焦测井曲线进行泥浆浸入影响校正.测井技术,1985年第3期.
[20]雍世和.用计算机对双侧向曲线进行泥浆浸入影响校正.华东石油学院学报,1985年第3期.
[21]王群,庞彦明,郭洪岩,等编著.矿场地球物理测井.石油工业出版社,2002年4月第1版.
[22]李周波主编.钻井地球物理勘探.地质出版社,1979年12月第1版.
[23]宋延杰,刘宪伟,黄宝华.混合泥质砂岩有效介质通用HB电阻率模型在高泥储层中的应用.测井技术,2003,27(6):508-512.
[24]宋延杰,王群,印长海.S-B电导率模型:确定泥质砂岩储层含水饱和度的新方法.测井技术,1995,19(4):244-249.
[25]徐守余.孤东油田七区中水淹层测井资料综合解释及评价.石油大学学报:自然科学版, 1999,23(5):24-27.
[26]吴长虹,李敬功,陈彬,等.高含水期产水率解释水淹层方法研究.河南石油,2004,18(5):30-34.
[27]侯连华,王京红,刘泽容.水淹层测井评价方法.石油学报,1999,20(3):49-55.
[28]韩清忠,雍世和.用常规电阻率测井资料确定水淹层的剩余油饱和度.测井技术,1996,20(5):351-355.
[29]韩志明,王宪成,李胜军.高含水期水淹层解释方法研究.测井技术,2000,24(5):333-336.
[30]陈一鸣,朱德怀,任康,等.矿场地球物理测井技术测井资料解释.北京,石油工业出版社,1994年9月第1版.
[31]刘红歧,彭仕宓,陈烨菲,等.高含水期水淹层的定量识别.新疆石油学院学报,2003,15(2):38-42.
[32]吴浩江,周芳德.应用RBF神经网络智能识别油气水多相流模型[J].石油学报,2000,21(3):57-60.
[33]王智平,刘在德,高成秀,等.遗传算法在BP网络权值学习中的应用[J].甘肃工业大学学报,2001,27(2):20-22.
[34]刘争平著.人工神经网络在测井-地震资料联合反演中的应用研究.科学出版社,2003年4月第1版.
[35]贾春雨,杨军,周群,等.喇嘛甸油田水淹层解释方法改进.大庆石油地质与开发，2002，21(3)：47-48.
[36]焦李成主编.神经网络系统理论[M].西安电子科大出版社,1995.
[37] TKlaus Cozzolino and Jadir da Conceicao da Silva. TSynthetic focusing andsimulation of dual laterolog tool in axisymmetric subsurface modelsT. Journal ofApplied Geophysics, Volume 61, Issue 2, February 2007, Pages 102-110.
[38] Martin K. Dubois, Geoffrey C. Bohling and Swapan Chakrabarti.Comparison of fourapproaches to a rock facies classification problem.Computers & Geosciences,Volume 33, Issue 5, May 2007, Pages 599-617.
[39] TCarl J. Koizumi.T Neutron capture loggingT calibration and data analysis forenvironmental contaminant assessmentT. Journal of Applied Geophysics, Volume 61,Issue 2, February 2007, Pages 111-131.
[40] TLarry N. Smith. TLate Pleistocene stratigraphy and implications for deglaciationand subglacial processes of the Flathead Lobe of the Cordilleran Ice Sheet, FlatheadValley, Montana, USA.T Sedimentary Geology, Volume 165, Issues 3-4, 15 March2004, Pages 295-332.
[41] TS. H. Al-Mahrooqi, C. A. Grattoni, A. K. Moss and X. D. Jing. TAn investigationof the effect of wettability on NMR characteristics of sandstone rock and fluidsystems.T Journal of Petroleum Science and Engineering, Volume 39, Issues3-4, September 2003, Pages 389-398.
[42] Larson, RGetal.Percolation Theory of Two-phase Flowin Porous Media.Chem.EngSci,36(1981)75-85.
[43] U.Woz′nicka，J.Jarzyna，E.Krynicka. Evaluation of uncertainty of thecomprehensive interpretation of borehole logs by the multiple re-runstatistical method. Applied Radiation and Isotopes 62 (2005) 817-827.
[44] Xin Yao, Yong Liu.A New Evolutionary System for Evolving Artificial NeuralNetworks. Natural Networks,1997,8(3):694-712.
[45] Du Zongjun, Jing Wanxue. Quantitatively Calculating Formation WaterResisitivity of Water-Flooded Zone with Log Data. Well Logging Technol1999,23(1):43-45.
[46] Tang Wengsheng, Ouyang Jian, Liu Xingli. Log Analysis of Natural Watered-out Reservoir. Well Logging Technol, 1995,19(6):421-427.
[47] Rogers S J et al. Determination of lithology from well logs using a neural network.AAPG, 1992,76(5):731-739.
[48] Fox G G..Empirically derived relationship between fractal dimension and power law from frenquency spectra [J].Pure and Applied Geophysics,1989,131(4):211-239
[49] Bremer M H, Kulenkam pff J ,Schopper J R. Lithological and fracture response of common logs in crystalline rock. In: Hurst A,Griffiths C M,Worthington P F eds. Geological Society Special Publication,1992,(65):221-234.
[50] Adler, F., Attribute Analysis Using Locally Scaled Regression : 61st Mtg. Eur. Assoc. Expl Geophys., Extended Abstracts, European Association Of Geophysical Exploration,1998, Session: P139.
[51] HEDGES P L.Infill Well Water-Cut Estimates Based on Open Hole Log Data in a mineralogically complex Reservoir : Kuparuk River Field Alaska.SPE 35684.
[52] QUINTERO L F . Estimation of Residual Gas Saturation from Neutron-Density Log Data. 47th Annu Cim Petrol Soc Tech Mtg Proc V 2,1996.
[2]韩镇石,具海啸,刘刚,等.应用测井资料识别喇嘛甸油田大孔道.大庆石油地质与开发,2006,25(4):72-74.
[3]宋考平,杨秋荣,付青春,等.模糊综合评判方法判断低效循环井层.钻井工业,2006,29(4):35-37.
[4]王学忠.井间大孔道预测方法研究.石油知识勘探开发,2005(1):18-19.
[5]尹文军,陈永生,王华,等.水力探测大孔道和剩余油饱和度解释模型的建立.油气地质与采收率.2005,12(1):63-65.
[6]金山.“大孔道”测试方法的优化.国外测井技术,2004,19(5):61-64.
[7]刘月田,孙保利,于永生.大孔道模糊识别与定量计算方法。石油钻采工艺,2003,25(5):54-59.
[8]史有刚,曾庆辉,周晓俊.大孔道试井理论解释模型。石油钻采工艺,2003,25(3):48-50.
[9]曾流芳,陈柏平,王学忠.疏松砂岩油藏大孔道定量描述初步研究.油气地质与采收率.2002,9(4):53-54.
[10]曾流芳,赵国景,张子海,等.疏松砂岩油藏大孔道形成机理及判别方法.应用基础与工程科学学报.2002,10(3):268-275.
[11]窦之林,曾流芳.大孔道诊断和描述技术研究.石油勘探与开发,2001,28(1):75-77.
[12]何长,李平,汪正勇,等.大孔道的表现特征及调剖对策.石油钻采艺,2000,22(5):63-66
[13]商志英,万新德,何长虹,等.应用水力探测方法确定储层大孔道及剩余油分布状况的研究。大庆石油地质与开发,2004,23(2):30-32.
[14]陈伟,马如雄,郝颜红.基于MATLAB的BP人工神经网络设计.电脑学习,2005,4(2):30-31.
[15]雍世和,孙宝佃.用滑动平均滤波法消除测井曲线上的毛刺干扰.华东石油学院学报,1983(1):11-19.
[16]雍世和。最优化测井解释.石油大学出版社,1995年5月第1版.
[17]雍世和,张超谟.测井数据处理与综合解释.石油大学出版社,1996年9月第1版.
[18]雍世和,孙宝佃.用计算机对放射性测井曲线进行环境影响校正.测井技术,1984年第4期.
[19]雍世和,靳秀英.用计算机对双感应-浅聚焦测井曲线进行泥浆浸入影响校正.测井技术,1985年第3期.
[20]雍世和.用计算机对双侧向曲线进行泥浆浸入影响校正.华东石油学院学报,1985年第3期.
[21]王群,庞彦明,郭洪岩,等编著.矿场地球物理测井.石油工业出版社,2002年4月第1版.
[22]李周波主编.钻井地球物理勘探.地质出版社,1979年12月第1版.
[23]宋延杰,刘宪伟,黄宝华.混合泥质砂岩有效介质通用HB电阻率模型在高泥储层中的应用.测井技术,2003,27(6):508-512.
[24]宋延杰,王群,印长海.S-B电导率模型:确定泥质砂岩储层含水饱和度的新方法.测井技术,1995,19(4):244-249.
[25]徐守余.孤东油田七区中水淹层测井资料综合解释及评价.石油大学学报:自然科学版, 1999,23(5):24-27.
[26]吴长虹,李敬功,陈彬,等.高含水期产水率解释水淹层方法研究.河南石油,2004,18(5):30-34.
[27]侯连华,王京红,刘泽容.水淹层测井评价方法.石油学报,1999,20(3):49-55.
[28]韩清忠,雍世和.用常规电阻率测井资料确定水淹层的剩余油饱和度.测井技术,1996,20(5):351-355.
[29]韩志明,王宪成,李胜军.高含水期水淹层解释方法研究.测井技术,2000,24(5):333-336.
[30]陈一鸣,朱德怀,任康,等.矿场地球物理测井技术测井资料解释.北京,石油工业出版社,1994年9月第1版.
[31]刘红歧,彭仕宓,陈烨菲,等.高含水期水淹层的定量识别.新疆石油学院学报,2003,15(2):38-42.
[32]吴浩江,周芳德.应用RBF神经网络智能识别油气水多相流模型[J].石油学报,2000,21(3):57-60.
[33]王智平,刘在德,高成秀,等.遗传算法在BP网络权值学习中的应用[J].甘肃工业大学学报,2001,27(2):20-22.
[34]刘争平著.人工神经网络在测井-地震资料联合反演中的应用研究.科学出版社,2003年4月第1版.
[35]贾春雨,杨军,周群,等.喇嘛甸油田水淹层解释方法改进.大庆石油地质与开发，2002，21(3)：47-48.
[36]焦李成主编.神经网络系统理论[M].西安电子科大出版社,1995.
[37] TKlaus Cozzolino and Jadir da Conceicao da Silva. TSynthetic focusing andsimulation of dual laterolog tool in axisymmetric subsurface modelsT. Journal ofApplied Geophysics, Volume 61, Issue 2, February 2007, Pages 102-110.
[38] Martin K. Dubois, Geoffrey C. Bohling and Swapan Chakrabarti.Comparison of fourapproaches to a rock facies classification problem.Computers & Geosciences,Volume 33, Issue 5, May 2007, Pages 599-617.
[39] TCarl J. Koizumi.T Neutron capture loggingT calibration and data analysis forenvironmental contaminant assessmentT. Journal of Applied Geophysics, Volume 61,Issue 2, February 2007, Pages 111-131.
[40] TLarry N. Smith. TLate Pleistocene stratigraphy and implications for deglaciationand subglacial processes of the Flathead Lobe of the Cordilleran Ice Sheet, FlatheadValley, Montana, USA.T Sedimentary Geology, Volume 165, Issues 3-4, 15 March2004, Pages 295-332.
[41] TS. H. Al-Mahrooqi, C. A. Grattoni, A. K. Moss and X. D. Jing. TAn investigationof the effect of wettability on NMR characteristics of sandstone rock and fluidsystems.T Journal of Petroleum Science and Engineering, Volume 39, Issues3-4, September 2003, Pages 389-398.
[42] Larson, RGetal.Percolation Theory of Two-phase Flowin Porous Media.Chem.EngSci,36(1981)75-85.
[43] U.Woz′nicka，J.Jarzyna，E.Krynicka. Evaluation of uncertainty of thecomprehensive interpretation of borehole logs by the multiple re-runstatistical method. Applied Radiation and Isotopes 62 (2005) 817-827.
[44] Xin Yao, Yong Liu.A New Evolutionary System for Evolving Artificial NeuralNetworks. Natural Networks,1997,8(3):694-712.
[45] Du Zongjun, Jing Wanxue. Quantitatively Calculating Formation WaterResisitivity of Water-Flooded Zone with Log Data. Well Logging Technol1999,23(1):43-45.
[46] Tang Wengsheng, Ouyang Jian, Liu Xingli. Log Analysis of Natural Watered-out Reservoir. Well Logging Technol, 1995,19(6):421-427.
[47] Rogers S J et al. Determination of lithology from well logs using a neural network.AAPG, 1992,76(5):731-739.
[48] Fox G G..Empirically derived relationship between fractal dimension and power law from frenquency spectra [J].Pure and Applied Geophysics,1989,131(4):211-239
[49] Bremer M H, Kulenkam pff J ,Schopper J R. Lithological and fracture response of common logs in crystalline rock. In: Hurst A,Griffiths C M,Worthington P F eds. Geological Society Special Publication,1992,(65):221-234.
[50] Adler, F., Attribute Analysis Using Locally Scaled Regression : 61st Mtg. Eur. Assoc. Expl Geophys., Extended Abstracts, European Association Of Geophysical Exploration,1998, Session: P139.
[51] HEDGES P L.Infill Well Water-Cut Estimates Based on Open Hole Log Data in a mineralogically complex Reservoir : Kuparuk River Field Alaska.SPE 35684.
[52] QUINTERO L F . Estimation of Residual Gas Saturation from Neutron-Density Log Data. 47th Annu Cim Petrol Soc Tech Mtg Proc V 2,1996.