时移地震资料贝叶斯AVO波形反演
|
摘要
针对时移地震差异数据,给出了一种基于贝叶斯理论的AVO波形反演方法.该方法可以利用时移地震差异数据同时反演出纵波阻抗、横波阻抗和密度的变化.利用时移地震资料进行反演,由于采集和处理过程中存在一定的差异,不同年份地震资料在非注采过程影响区域也会存在一定的变化,而该变化会导致反演结果在非注采区域有较大的变化.针对这一问题,本文采用贝叶斯理论框架,将待求的纵横波阻抗、密度变化的先验信息和包含在地震数据中的信息结合起来,对于纵横波阻抗和密度变化,假设其服从Gauss分布,并以时移地震分别反演的结果作为其期望,同时,为了更好地表征储层属性变化,提高分辨率和抑制非注采区域弹性参数的变化,假设弹性参数变化的导数服从改进的Cauchy分布.数值模拟试验和实际资料处理结果皆表明,本文提出的反演方法能够有效地抑制假象,突出储层性质的变化,得到高分辨率的弹性参数变化信息,为研究储层属性的变化和优化开采方案提供更多的有效的信息.
Aiming at the difference of time-lapse seismic data of different vintages,a Bayesian AVO waveform inversion method is given in this paper.This method can obtain the changes of compressional/shear impedance and density simultaneously using the pre-stack seismic data in different vintages.Because of the differences in the process of data acquisition and processing in the seismic data of different vintages,when the time-lapse seismic inversion is made,changes exist in the region where the process of injection and production does not have impacts,and these changes will lead to some obvious changes in the inversion results in this region.To solve the problem,in this paper,Bayesian theoretical framework is used to combine the prior information of the changes of the elastic parameters with information contained in the seismic data.We assume the changes of compressional/shear impedance and density follow the Gaussian distribution,whose expectation is the answer of time-lapse seismic uncoupled inversion.Meanwhile in order to better characterize the changes in reservoir properties and improve the resolution of the results and inhibit the artifacts in the non-production region,we also assume that the derivatives of changes of elastic parameters follow the improved Cauchy distribution.Tests on synthetic data and practical data show that the proposed inversion method can effectively inhibit the artifacts,highlight the changes of reservoir properties,get the changes of elastic parameters of high-resolution and provide more effective information for the study of reservoir properties and the optimization of recovery scheme.
Aiming at the difference of time-lapse seismic data of different vintages,a Bayesian AVO waveform inversion method is given in this paper.This method can obtain the changes of compressional/shear impedance and density simultaneously using the pre-stack seismic data in different vintages.Because of the differences in the process of data acquisition and processing in the seismic data of different vintages,when the time-lapse seismic inversion is made,changes exist in the region where the process of injection and production does not have impacts,and these changes will lead to some obvious changes in the inversion results in this region.To solve the problem,in this paper,Bayesian theoretical framework is used to combine the prior information of the changes of the elastic parameters with information contained in the seismic data.We assume the changes of compressional/shear impedance and density follow the Gaussian distribution,whose expectation is the answer of time-lapse seismic uncoupled inversion.Meanwhile in order to better characterize the changes in reservoir properties and improve the resolution of the results and inhibit the artifacts in the non-production region,we also assume that the derivatives of changes of elastic parameters follow the improved Cauchy distribution.Tests on synthetic data and practical data show that the proposed inversion method can effectively inhibit the artifacts,highlight the changes of reservoir properties,get the changes of elastic parameters of high-resolution and provide more effective information for the study of reservoir properties and the optimization of recovery scheme.
引文
[1]Landr M.Discrimination between pressure and fluidsaturation changes from time-lapse seismic data.Geophysics,2001,66(3):836-844.
[2]Landr M,Veire H H,Duffaut K,et al.Discriminationbetween pressure and fluid saturation changes from marinemulticomponent time-lapse seismic data.Geophysics,2003,68(5):1592-1599.
[3]Batzle M,Wang Z.Seismic properties of pore fluid.Geophysics,1992,57(11):1396-1408.
[4]Sarkar S,Gouveia W P,Johnston D H.On the inversion oftime-lapse seismic data.73th Annual International Meeting,SEG,Expanded Abstract,2003:1489-1492.
[5]Tura A,Lumley D E.Estimating pressure and saturationchanges from time-lapse AVO data.69th Annual InternationalMeeting,SEG,Expanded Abstract,1999:1655-1658.
[6]Gluck S,Deschizeaux B,Mignot A,et al.Time-lapseimpedance inversion of post-stack seismic data.70th AnnualInternational Meeting,SEG,Expanded Abstract,2000:1509-1512.
[7]Abubakar A,Berg P M,Fokkema J T.A feasibility study onnonlinear inversion of time-lapse seismic data.71th AnnualInternational Meeting,SEG,Expanded Abstract,2001:1664-1667.
[8]Buland A,Ouair Y E.Bayesian time-lapse inversion.Geophysics,2006,71(3):43-48.
[9]李来林,牟永光,陈小宏.计算时移波阻抗的新方法.大庆石油地质与开发,2001,20(6):54-55.Li L L,Mou Y G,Chen X H.A new method for calculatingtime shift wave impedance.Petroleum Geology&Oil FieldDevelopment in Daqing(in Chinese),200l,20(6):54-55.
[10]李景叶,陈小宏,金龙.时移地震AVO反演在油藏定量解释中的应用.石油学报,2005,26(3):68-73.Li J Y,Chen X H,Jin L.Application of time-lapse seismicAVO inversion to quantitative interpretation of reservoir.Acta Petrolei Sinica(in Chinese),2005,26(3):68-73.
[11]李景叶,陈小宏,郝振江等.多波时移地震AVO反演研究.地球物理学报,2005,48(4):902-908.Li J Y,Chen X H,Hao Z J,et al.A study on multiple time-lapse seismic AVO inversion.Chinese J.Geophys.(inChinese),2005,48(4):902-908.
[12]陈小宏,赵伟,马继涛等.时移地震非线性反演压力与饱和度变化研究.地球物理学进展,2006,21(3):830-836.Chen X H,Zhao W,Ma J T,et al.Pressure and saturationchange nonlinear inversion method using time-lapse seismicdata.Progress in Geophysics(in Chinese),2006,21(3):830-836.
[13]陈勇,韩波,肖龙等.多尺度全变分法及其在时移地震中的应用.地球物理学报,2010,53(8):1883-1892.Chen Y,Han B,Xiao L,et al.Multiscale total variationmethod and its application on time-lapse seismic.Chinese J.Geophys.(in Chinese),2010,53(8):1883-1892.
[14]Eidsvik J.Blocky inversion of time-lapse seismic AVO data.79th Annual International Meeting,SEG,Expanded Abstract,2009:3820-3824.
[15]Kato A,Stewart R R.Joint AVO inversion for time-lapseelastic reservoir properties.Hangingstone heavy oilfield,Alberta.80th Annual International Meeting,SEG,ExpandedAbstract,2010:4432-4437.
[16]Hansteen F,Wills P B,Hornman K,et al.Time-lapserefraction seismic monitoring.80th Annual InternationalMeeting,SEG,2010:4170-4174.
[17]Suman A,Fernandes-Martinez J L,Mukerji T.Jointinversion of time-lapse seismic and production data for Nornefield.81th Annual International Meeting,SEG,2011:4102-4108.
[18]Zhao L X,Geng J H,Cheng J B,et al.Combined BayesianAVO inversion with rock physics to predict gas carbonatereservoir.81th Annual International Meeting,SEG,2011:2747-2751.
[19]Li J Y,Wang S D,Chen X H.Time-lapse seismic elasticimpedance difference inversion and application.81th AnnualInternational Meeting,SEG,2011:2492-2496.
[20]Zhu Z Y,Jiang X D,Zhao W,et al.Time-lapse prestackelastic impedance inversion based on seismic difference data.81th Annual International Meeting,SEG,2011:2497-2501.
[21]Asnaashari A,Brossier R,Garambois S,et al.Sensitivityanalysis of time-lapse images obtained by differentialwaveform inversion with respect to reference model.81thAnnual International Meeting,SEG,2011:2482-2486.
[22]张繁昌,刘杰,印兴耀等.修正柯西约束地震盲反褶积方法.石油地球物理勘探,2008,43(4):391-396.Zhang F C,Liu J,Yin X Y,et al.Modified Cauchy-constrained seismic blind deconvolution.Oil GeophysicalProspecting(in Chinese),2008,43(4):391-396.
[23]Gidlow P M,Smith G C,Vail P J.Hydrocarbon detectionusing fluid factor traces,a case study:How useful is AVOanalysis?Joint SEG/EAEG Summer Research Workshop,Technical Program and Abstracts,1992:78-89.
[24]Todoeschuck J P,Jensen O G,Labonte S.Gaussian scalingnoise model of seismic reflection sequences:Evidence fromwell logs.Geophysics,1990,55(4):480-484.
[25]Walden A T,Hosken J W.The nature of the non-Gaussianity of primary reflection coefficients and itssignificance for deconvolution.Geophys.Prosp.,1986,34(7):1038-1066.
[2]Landr M,Veire H H,Duffaut K,et al.Discriminationbetween pressure and fluid saturation changes from marinemulticomponent time-lapse seismic data.Geophysics,2003,68(5):1592-1599.
[3]Batzle M,Wang Z.Seismic properties of pore fluid.Geophysics,1992,57(11):1396-1408.
[4]Sarkar S,Gouveia W P,Johnston D H.On the inversion oftime-lapse seismic data.73th Annual International Meeting,SEG,Expanded Abstract,2003:1489-1492.
[5]Tura A,Lumley D E.Estimating pressure and saturationchanges from time-lapse AVO data.69th Annual InternationalMeeting,SEG,Expanded Abstract,1999:1655-1658.
[6]Gluck S,Deschizeaux B,Mignot A,et al.Time-lapseimpedance inversion of post-stack seismic data.70th AnnualInternational Meeting,SEG,Expanded Abstract,2000:1509-1512.
[7]Abubakar A,Berg P M,Fokkema J T.A feasibility study onnonlinear inversion of time-lapse seismic data.71th AnnualInternational Meeting,SEG,Expanded Abstract,2001:1664-1667.
[8]Buland A,Ouair Y E.Bayesian time-lapse inversion.Geophysics,2006,71(3):43-48.
[9]李来林,牟永光,陈小宏.计算时移波阻抗的新方法.大庆石油地质与开发,2001,20(6):54-55.Li L L,Mou Y G,Chen X H.A new method for calculatingtime shift wave impedance.Petroleum Geology&Oil FieldDevelopment in Daqing(in Chinese),200l,20(6):54-55.
[10]李景叶,陈小宏,金龙.时移地震AVO反演在油藏定量解释中的应用.石油学报,2005,26(3):68-73.Li J Y,Chen X H,Jin L.Application of time-lapse seismicAVO inversion to quantitative interpretation of reservoir.Acta Petrolei Sinica(in Chinese),2005,26(3):68-73.
[11]李景叶,陈小宏,郝振江等.多波时移地震AVO反演研究.地球物理学报,2005,48(4):902-908.Li J Y,Chen X H,Hao Z J,et al.A study on multiple time-lapse seismic AVO inversion.Chinese J.Geophys.(inChinese),2005,48(4):902-908.
[12]陈小宏,赵伟,马继涛等.时移地震非线性反演压力与饱和度变化研究.地球物理学进展,2006,21(3):830-836.Chen X H,Zhao W,Ma J T,et al.Pressure and saturationchange nonlinear inversion method using time-lapse seismicdata.Progress in Geophysics(in Chinese),2006,21(3):830-836.
[13]陈勇,韩波,肖龙等.多尺度全变分法及其在时移地震中的应用.地球物理学报,2010,53(8):1883-1892.Chen Y,Han B,Xiao L,et al.Multiscale total variationmethod and its application on time-lapse seismic.Chinese J.Geophys.(in Chinese),2010,53(8):1883-1892.
[14]Eidsvik J.Blocky inversion of time-lapse seismic AVO data.79th Annual International Meeting,SEG,Expanded Abstract,2009:3820-3824.
[15]Kato A,Stewart R R.Joint AVO inversion for time-lapseelastic reservoir properties.Hangingstone heavy oilfield,Alberta.80th Annual International Meeting,SEG,ExpandedAbstract,2010:4432-4437.
[16]Hansteen F,Wills P B,Hornman K,et al.Time-lapserefraction seismic monitoring.80th Annual InternationalMeeting,SEG,2010:4170-4174.
[17]Suman A,Fernandes-Martinez J L,Mukerji T.Jointinversion of time-lapse seismic and production data for Nornefield.81th Annual International Meeting,SEG,2011:4102-4108.
[18]Zhao L X,Geng J H,Cheng J B,et al.Combined BayesianAVO inversion with rock physics to predict gas carbonatereservoir.81th Annual International Meeting,SEG,2011:2747-2751.
[19]Li J Y,Wang S D,Chen X H.Time-lapse seismic elasticimpedance difference inversion and application.81th AnnualInternational Meeting,SEG,2011:2492-2496.
[20]Zhu Z Y,Jiang X D,Zhao W,et al.Time-lapse prestackelastic impedance inversion based on seismic difference data.81th Annual International Meeting,SEG,2011:2497-2501.
[21]Asnaashari A,Brossier R,Garambois S,et al.Sensitivityanalysis of time-lapse images obtained by differentialwaveform inversion with respect to reference model.81thAnnual International Meeting,SEG,2011:2482-2486.
[22]张繁昌,刘杰,印兴耀等.修正柯西约束地震盲反褶积方法.石油地球物理勘探,2008,43(4):391-396.Zhang F C,Liu J,Yin X Y,et al.Modified Cauchy-constrained seismic blind deconvolution.Oil GeophysicalProspecting(in Chinese),2008,43(4):391-396.
[23]Gidlow P M,Smith G C,Vail P J.Hydrocarbon detectionusing fluid factor traces,a case study:How useful is AVOanalysis?Joint SEG/EAEG Summer Research Workshop,Technical Program and Abstracts,1992:78-89.
[24]Todoeschuck J P,Jensen O G,Labonte S.Gaussian scalingnoise model of seismic reflection sequences:Evidence fromwell logs.Geophysics,1990,55(4):480-484.
[25]Walden A T,Hosken J W.The nature of the non-Gaussianity of primary reflection coefficients and itssignificance for deconvolution.Geophys.Prosp.,1986,34(7):1038-1066.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心