基于统计岩石物理学的直接岩石物理反演
|
摘要
统计岩石物理学目前已经成为量化地震储层描述中一种重要的新方法。统计岩石物理学把确定性的岩石物理学关系与地质统计学结合起来了的:进一步说,利用岩石物理学方程建立起储层参数和地震属性之间的联系,而统计技术被用来在岩石物性转换时传递不确定性和描述空间的变化。一般来讲传统的结合地震数据的地质统计学方法都是采用两步的序贯方法。其局限在于反演的弹性属性必须做油藏模型积分的时深转换,例如,岩石物理反演,由测井曲线建立的油藏弹性模型,能够的用到反演数据上,而并不需要正确解释反演结果的带限属性。一般来讲序贯反演的工作流程并不能够保证最终的油藏模型与地震数据完全一致。本篇文章提出的直接岩石物理学方法,并不采用传统的序贯反演的流程,其优势在于将深度转换进程结合到整个反演进程中,采用多尺度和多区域网格,收敛速度快,不确定性大大降低。
Stastical rock physics has recently emerged as an important new approach for quantitative seismic reservoir characterization.Statistical rock physics combines deterministic rock physics relations with geostatistics:the rock physics equations are used to establish the link between reservoir parameters and seismic properties,while statistical techniques are used to propagate uncertainty in the rock physics transform and describe spatial variability.Generally,the traditional Geostatistics take two-step sequential approach.There are limits that inverted elastic attributes must be converted from time to depth for integration with the reservoir model.For example,in rock-physics inversion,the PEM constructed at the log scale is applied sample-by-sample to the inverted data without properly accounting for the band-limited nature of the inversion results.Sequential inversion workflows do not guarantee that the final reservoir model is fully consistent with the seismic data.In this paper,Direcal rock physic techniques not take the sequential process,the advantage is that integrate the depth conversion into the inversion processing,by multi-scale and multi-domain stratigraphyic grid,the speed of the convergence is quick,the uncertainty is lowered greatly.
Stastical rock physics has recently emerged as an important new approach for quantitative seismic reservoir characterization.Statistical rock physics combines deterministic rock physics relations with geostatistics:the rock physics equations are used to establish the link between reservoir parameters and seismic properties,while statistical techniques are used to propagate uncertainty in the rock physics transform and describe spatial variability.Generally,the traditional Geostatistics take two-step sequential approach.There are limits that inverted elastic attributes must be converted from time to depth for integration with the reservoir model.For example,in rock-physics inversion,the PEM constructed at the log scale is applied sample-by-sample to the inverted data without properly accounting for the band-limited nature of the inversion results.Sequential inversion workflows do not guarantee that the final reservoir model is fully consistent with the seismic data.In this paper,Direcal rock physic techniques not take the sequential process,the advantage is that integrate the depth conversion into the inversion processing,by multi-scale and multi-domain stratigraphyic grid,the speed of the convergence is quick,the uncertainty is lowered greatly.
引文
[1]王家华,张团峰.油气储层随机建模[M].北京:石油工业出版社,2001.
[2]肖斌,赵鹏大,侯景儒.地质统计学新进展[J].地球科学进展,2000,15(03):492-692.
[3]张一伟,熊琦华,纪发华.地质统计学在油藏描述中的应用[M].东营:石油大学出版社,1992.
[4]刘文玲.地震约束储层地质建模技术[J].石油学报,2008,29(1):64.
[5]李黎,王永刚.地质统计学应用综述[J].勘探地球物理进展,2006,29(3):163-166.
[6]吴胜和,李文克.多点地质统计学—理论应用与展望[J].古地理学报,2005,7(1):137-143.
[7]Stastical rock physics:Combining rock physics,information theory,and geostatistics to reduce uncertainty in seismic reservoir characterization[M]TAPAN MUKERJI,PER AVESTHeta.
[8]Mapping lithofacies and pore-fluid probabilities in a North Sea reservoir:Seismic inversions and statiscal rock physics[M]T.Mukerji*G.Makvo and J.R.Granli.
[2]肖斌,赵鹏大,侯景儒.地质统计学新进展[J].地球科学进展,2000,15(03):492-692.
[3]张一伟,熊琦华,纪发华.地质统计学在油藏描述中的应用[M].东营:石油大学出版社,1992.
[4]刘文玲.地震约束储层地质建模技术[J].石油学报,2008,29(1):64.
[5]李黎,王永刚.地质统计学应用综述[J].勘探地球物理进展,2006,29(3):163-166.
[6]吴胜和,李文克.多点地质统计学—理论应用与展望[J].古地理学报,2005,7(1):137-143.
[7]Stastical rock physics:Combining rock physics,information theory,and geostatistics to reduce uncertainty in seismic reservoir characterization[M]TAPAN MUKERJI,PER AVESTHeta.
[8]Mapping lithofacies and pore-fluid probabilities in a North Sea reservoir:Seismic inversions and statiscal rock physics[M]T.Mukerji*G.Makvo and J.R.Granli.
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