叠前多级优化联合偏移速度建模  

Pre-stack joint migration velocity modeling with multi-stage optimization

作　　者：秦宁[1] 李振春[1] 杨晓东[2] 贾正虹 岳玉波 

机构地区：[1]中国石油大学(华东)地球科学与技术学院,山东青岛266580 [2]中石化胜利石油管理局石油开发中心,山东东营257000 [3]中石油东方地球物理公司物探技术研究中心,河北涿州072751

出　　处：《地球物理学进展》2013年第1期320-328,共9页Progress in Geophysics

基　　金：国家863课题(2010AA0603223002);国家科技重大专项课题(2011ZX05006-002);中央高校基本科研业务费专项资金(12CX04008A);中国石油大学(华东)研究生创新基金重点项目(CXZD11-01)联合资助

摘　　要：推导了基于角度域共成像点道集的叠前深度层析速度建模公式,提出了一种叠前多级优化联合偏移速度建模方法.通过基于共散射点(CSP)道集的叠前时间偏移速度分析获取初始速度,利用基于角度域共成像点道集(ADCIGs)的叠前深度层析速度反演进行速度更新建模.实现步骤可以概括为:首先,将叠前地震数据映射为CSP道集,利用CSP道集的叠加速度谱进行速度分析得到均方根速度场;其次,通过Dix公式将均方根速度场转换为层速度场,以此进行层析初始速度建模,基于ADCIGs实现叠前深度层析速度反演,最终得到高精度的叠前偏移速度场.断层模型和实际资料试算结果验证了该方法的正确性和有效性.This paper derived the formula of pre-staek depth migration tomographie velocity analysis hased on angle domain commorL imaging gathers （ADCIGs） and proposed a pre-stack joint migration velocity modeling method with multi-stage optimization which gets the initial velocity through the pre-stack time migration velocity analysis on the basis of common scattering point （CSP） gathers and obtains the velocity update through the pre-stack depth migration velocity analysis on the basis of AIX21Gs. The implement steps can be summarized as follows： First, make the pre-stack seismic data mapped to CSP gathers and get the root-mean-square velocity by stack velocity spectrum of CSP gathers; then transform the root-mean-square velocity to interval velocity in order to model the initial tomography velocity field; finally obtain the pre-stack migration velocity field with high-precision by pre-stack depth migration tomographic velocity inversion based on ADCIGs. The pre-stack time migration velocity analysis based on CSP gathers can solve dispersion problem of reflection points for inclined layers, get focused velocity spectrum and obtain good initial velocity quickly. ADCIGs are obtained from prestack depth migration, not affected by multi-path problem and free of artifacts, and coupling relation of velocity and depth can be revealed clearly in AI）CIGs. The pre-stack depth tomographic velocity modeling based on ADCIGs can be easily implemented without considering complex reflection problems in ray tracing, and each step is updated on the basis of prestack depth migration in the prior step. Through the flatness degree of ADCIGs, it is easy to judge the velocity update and finally achieve a high precision velocity field. Examples of fault model and field dataset shows this method is feasible and valid, but its precision is restrained by the resolution of ADCIGs, and prestack data with low S/N may make impact on inversion results. So before the implement of pre-stack joint migration velocity modelling, data pre-process

关 键 词：叠前 多级优化 速度建模 共散射点道集 角度域共成像点道集 

分 类 号：P631[天文地球—地质矿产勘探]