全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于metropolis抽样的非线性反演方法

Keywords: fft-ma,gdm,贝叶斯理论,非线性反演,高分辨率,metropolis抽样

Full-Text   Cite this paper   Add to My Lib

Abstract:

?基于metropolis抽样的非线性反演应用贝叶斯理论框架,是一种基于蒙特卡洛的非线性反演方法,能够有效地融合测井资料中的高频信息,提高反演结果的分辨率。首先通过快速傅里叶滑动平均模拟算法(fft-ma)和逐渐变形算法(gdm)得到基于地质统计学的先验信息;进而构建似然函数;最后利用metropolis算法对后验概率密度进行抽样,得到反演问题的解。其中fft-ma模拟作为一种高效的频率域模拟方法,融入gdm更新算法之后,可以在保持模拟空间结构不变的前提下,连续修改储层模型,保证反演结果有效地收敛,直至满足实际观测地震记录。模型试算和实际数据处理结果表明:基于metropolis抽样的非线性反演可以提供合理的弹性参数信息,尤其是提高纵波速度的分辨率,即使信噪比较小时,仍然可以反演出合理的弹性参数信息,从而证明了该方法的有效性;当不考虑噪声时,纵、横波阻抗的反演分辨率较弹性参数本身的反演分辨率更高。

References

[1]  francisa.limitationsofdeterministicandadvantagesofstochasticseismicinversion.canadiansocietyofexplorationgeophysicistsrecorder,2005,30(2):5-11.
[2]  francisa.understandingstochasticinversion:part1.firstbreak,2006,24(11):69-77.
[3]  francisa.understandingstochasticinversion:part2.firstbreak,2006,24(12):79-84.
[4]  dubruleo.workshopreport:“uncertaintyinreserveestimates”eageconference,amsterdam,2june1996.petroleumgeoscience,1996,2(4):351-352.
[5]  kj?nsbergh,hauger,kolbj?rnsenoetal.bayesianmontecarlomethodforseismicpredrillprospectassessment.geophysics,2010,75(2):o9-o19.
[6]  张繁昌,肖张波,印兴耀.地震数据约束下的贝叶斯随机反演[j].石油地球物理勘探.2014,49(1):176-182??浏览
[7]  zhangfanchang,xiaozhangbo,yinxingyao.baye-sianstochasticinversionconstrainedbyseismicdata[j].ogp.2014,49(1):176-182??浏览
[8]  leravalecm,noetingerb,huly.thefftmovingaverage(fft-ma)generator:anefficientnumericalmethodforgeneratingandconditioninggaussiansimulations[j].mathematicalgeology.2000,32(6):701-723
[9]  huly.gradualdeformationanditerativecalibrationofgaussian-relatedstochasticmodels[j].mathematicalgeology.2000,32(1):87-108
[10]  johnas,martinls,svent.introductorygeophysicalinversetheory.coloradoandnewengland:samizdatpress,2001.
[11]  duijndamajw.bayesianestimationinseismicinversion,partii:uncertaintyanalysis[j].geophysicalprospecting.1988,36(8):899-918
[12]  journelag,huijbregtscj.mininggeostatistics.london,newyorkandsanfrancisco:academicpress,1978.
[13]  mosegaardk,tarantolaa.montecarlosamplingofsolutionstoinverseproblems.journalofgeophysicalresearch:solidearth(1978-2012),1995,100(b7):12431-12447.
[14]  metropolisn,rosenbluthaw,rosenbluthmnetal.equationofstatecalculationsbyfastcomputingmachines[j].thejournalofchemicalphysics.1953,21(6):1087-1092
[15]  sanceveross,remacreaz,desouzaportugalretal.comparingdeterministicandstochasticseismicinversionforthin-bedreservoircharacterizationinaturbiditesyntheticreferencemodelofcamposbasin,brazil[j].theleadingedge.2005,24(11):1168-1172[2]moyenr,doyenpm.reservoirconnectivityuncertaintyfromstochasticseismicinversion.segtechnicalprogramexpandedabstracts,2009,28:2378-2382.[3]samsms,saussusd.comparisonofuncertaintyestimatesfromdeterministicandgeostatisticalinversion.the70theageconference&exhibition,2008.[4]francisa.limitationsofdeterministicandadvantagesofstochasticseismicinversion.canadiansocietyofexplorationgeophysicistsrecorder,2005,30(2):5-11.[5]francisa.understandingstochasticinversion:part1.firstbreak,2006,24(11):69-77.[6]francisa.understandingstochasticinversion:part2.firstbreak,2006,24(12):79-84.[7]dubruleo.workshopreport:“uncertaintyinreserveestimates”eageconference,amsterdam,2june1996.petroleumgeoscience,1996,2(4):351-352.[8]kj?nsbergh,hauger,kolbj?rnsenoetal.bayesianmontecarlomethodforseismicpredrillprospectassessment.geophysics,2010,75(2):o9-o19.[9]张繁昌,肖张波,印兴耀.地震数据约束下的贝叶斯随机反演[j].石油地球物理勘探.2014,49(1):176-182??浏览zhangfanchang,xiaozhangbo,yinxingyao.baye-sianstochasticinversionconstrainedbyseismicdata[j].ogp.2014,49(1):176-182??浏览[10]leravalecm,noetingerb,huly.thefftmovingaverage(fft-ma)generator:anefficientnumericalmethodforgeneratingandconditioninggaussiansimulations[j].mathematicalgeology.2000,32(6):701-723[11]huly.gradualdeformationanditerativecalibrationofgaussian-relatedstochasticmodels[j].mathematicalgeology.2000,32(1):87-108[12]johnas,martinls,svent.introductorygeophysicalinversetheory.coloradoandnewengland:samizdatpress,2001.[13]oliverds.movingaveragesforgaussiansimulationintwoandthreedimensions[j].mathematicalgeology.1995,27(8):939-960[14]duijndamajw.bayesianestimationinseismicinversion,partii:uncertaintyanalysis[j].geophysicalprospecting.1988,36(8):899-918[15]journelag,huijbregtscj.mininggeostatistics.london,newyorkandsanfrancisco:academicpress,1978.[16]huly,blancg.constrainingareservoirfaciesmodeltodynamicdatausingagradualdeformationmethod.the6theuropeanconferenceonthemathematicsofoilrecovery,1998.[17]huly,leravalecm,blancgetal.reducinguncertaintiesinproductionforecastsbyconstraininggeologicalmodelingtodynamicdata.speannualtechnicalconferenceandexhibition,societyofpetroleumengineers,1999.[18]mosegaardk,tarantolaa.montecarlosamplingofsolutionstoinverseproblems.journalofgeophysicalresearch:solidearth(1978-2012),1995,100(b7):12431-12447.[19]metropolisn,rosenbluthaw,rosenbluthmnetal.equationofstatecalculationsbyfastcomputingmachines[j].thejournalofchemicalphysics.1953,21(6):1087-1092[20]tarantolaa.inverseproblemtheoryandmethodsformodelparameterestimation.paris:siam,2005.
[16]  sanceveross,remacreaz,desouzaportugalretal.comparingdeterministicandstochasticseismicinversionforthin-bedreservoircharacterizationinaturbiditesyntheticreferencemodelofcamposbasin,brazil[j].theleadingedge.2005,24(11):1168-1172
[17]  moyenr,doyenpm.reservoirconnectivityuncertaintyfromstochasticseismicinversion.segtechnicalprogramexpandedabstracts,2009,28:2378-2382.
[18]  samsms,saussusd.comparisonofuncertaintyestimatesfromdeterministicandgeostatisticalinversion.the70theageconference&exhibition,2008.
[19]  oliverds.movingaveragesforgaussiansimulationintwoandthreedimensions[j].mathematicalgeology.1995,27(8):939-960
[20]  huly,blancg.constrainingareservoirfaciesmodeltodynamicdatausingagradualdeformationmethod.the6theuropeanconferenceonthemathematicsofoilrecovery,1998.
[21]  huly,leravalecm,blancgetal.reducinguncertaintiesinproductionforecastsbyconstraininggeologicalmodelingtodynamicdata.speannualtechnicalconferenceandexhibition,societyofpetroleumengineers,1999.
[22]  tarantolaa.inverseproblemtheoryandmethodsformodelparameterestimation.paris:siam,2005.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133