基于共轭方向置换的叠前三参数反演方法及其实现
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
叠前地震资料含有地层的纵、横波速度和密度等信息。利用叠前地震反演获得地震数据中的这些信息,即可揭示岩性及孔隙流体的性质。本文以平面波在层状弹性半空间传播时形成的地震共反射点道集数据与地层参数的关系为基础,提出了一种基于Powell共轭方向置换的叠前三参数反演方法。该方法不需要求解和存储庞大而复杂的一阶和二阶导数矩阵,通过目标函数的正则化,极大地改善了三参数反演的病态问题,并压缩了解空间。方向置换过程利用新的置换策略,有效防止了搜索方向的线性相关。同时,在反演实现过程中提出了一种高、低频分步反演策略,既保证了叠前地震数据信噪比较低时反演结果的稳定性和精度,又大大提高了收敛速度。本方法已利用理论模型进行了检验,并应用于油田实际工区,为充分利用叠前地震数据进行岩性识别和有利储层预测奠定了基础。
The information of strata such as P-and S-wave velocities and density is contained in prestack seismic data. Using prestack seismic inversion to acquire the information in seismic data is able to reveal the nature of lithology and pore fluid. Taking the relationship between seismic CRP gather data and strata parameters formed by plane waves propagating in layered elastic half-space,the paper presented prestack three-term inversion based on Powell's conjugate direction substitution. Without requiring solution and storage of large and complicated first-and second-order derivative,the method greatly improved the ill-conditioned problem of three-term inversion by normalization of objective function,and compressed the solution space as well. The usage of new substitution strategy in direction substitution process effectively prevented the linear correlation in searching direction. At the same time,the split inversion strategy for low and high frequencies was presented in implementing the inversion,which both guarantees the stability and precision of inverted results when prestack seismic data are lower S/N ratio and greatly improved convergent speed. The method has been tested with theoretical model and used in practical work zone of oilfield,laid the foundation for lithologic recognition and favorable reservoir prediction by fully using prestack seismic data.
引文
[1]Anderson P F,Gray F D.Using LMRfor dual attributeli-thology identification.Expanded Abstracts of71st SEG Mtg,2001,201~202
    [2]朱向阳,熊有伦.一种改进的Powell共轭方向算法.控制与决策,1996,11(2):304~308
    [3]孙成禹,倪长宽,李胜军等.广角地震反射数据特征及校正方法研究.石油地球物理勘探,2007,42(1):24~29
    [4]Buland A,Omre H.Bayesian linearized AVOinver-sion.Geophysics,2003,68:185~198
    [5]Debski W,Tarantola A.Information on elastic param-eters obtained fromthe amplitudes of reflected waves.Geophysics,1995,60:1426~1436
    [6]Castagna J P,Batzle M L,Eastwood R L.Relation-ships between compressional wave and shear wave ve-locities in clastic silicate rocks.Geophysics,1985,50:571~581
    [7]Brent R P.Algorithm for mini mization without de-rivatives.Englewood Cliffs:Prentice-Hall,1973

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