【作者】 赵景霞；

【导师】 冯恩民； 张叔伦；

【作者基本信息】 大连理工大学 ， 运筹学与控制论， 2005， 博士

【摘要】 本论文围绕三维叠前深度偏移课题开展研究。论文内容包括共炮检距剖面相位移叠前时间偏移算法,三维合成震源记录叠前深度偏移方法及其并行算法,自适应空间分区裂步傅立叶延拓方法以及合成震源记录剩余偏移速度分析。 共炮检距剖面叠前时间偏移算法根据地震几何学的理论,从简单的点绕射代数公式出发,用数值拟合方法在常速条件下把零炮检距剖面的叠后相位移公式推广到共炮检距剖面。理论上共炮检距剖面时间偏移的相位移公式只在常速下是精确的,运用平均速度或叠加速度可使公式适用于纵向速度变化,合成记录和野外资料试算表明,既使存在较大速度变化,共炮检距剖面叠前时间偏移方法也能有效成像。 三维合成震源记录叠前深度偏移方法利用波动方程算子的三个性质合成炮震源及炮震源记录,从理论上统一了相位编码和面炮合成技术,推导出了包括面炮记录和相位编码的统一的合成震源和合成震源记录公式,并设计了几种不同功能的合成算子,使合成震源与震源记录设计更为灵活方便,理论更为完整;推导并实现了横测线方向和纵测线方向的联合编码,进一步提高效率数倍。设计、实现了基于MPI的三维叠前合成震源记录深度偏移的并行算法,算法并行效率和加速比理想。SEG/EAEG盐丘C3-NA数据模型上的试算结果和新疆三维起伏地表实际地震资料的处理结果进一步说明了三维叠前合成震源记录深度偏移方法是一个有效、实用、成像质量清晰适于复杂构造成像的偏移方法,同炮记录叠前深度偏移相比,合成震源记录叠前深度偏移在保持精度的同时具有更高的计算效率。 自适应空间分区裂步傅立叶延拓方法根据全局速度变化和算法误差来选择参考慢度,同一个参考慢度所对应的速度构成一个分区,每一个空间分区可以由几个空间上不连续的子分区组成,从而有效地减小了参考慢度的个数。每一延拓步的参考慢度、参考慢度的个数和如何构建分区都是根据速度函数自动生成。为了消除速度场强间断产生的人为噪音,设计了简单有效的f-k域的光滑滤波。 确定速度模型是叠前成像的关键步骤。本文在平面波剩余速度分析方法的基础上提出了基于波动理论的合成震源记录剩余偏移速度分析方法。剩余偏移速度分析在叠前深度偏移之后使用代数公式进行剩余深度校正来更新层速度,校正公式是解析的,减少了速度分析中的盲目性和人为性,计算速度快,可以交互实现。在复杂构造速度分析时,采用控制照明技术可校正平面波因复杂构造而引起的波前畸变,从而减小速度分析的误差。合成震源记录偏移方法用于速度分析,提高了速度分析中迭代偏移的计算效率和成像质量。模型和实际资料的试算结果验证了方法的有效性和正确性。更多还原

【Abstract】 The project related to 3D prestack depth migration of synthetic source records is studied in this dissertation. It includes common-offset prestack time migration, 3D parallel prestack depth migration of synthetic source records, adaptive spatial-division split-step Fourier migration method and residual migration velocity analysis by synthetic source records.In common-offset prestack time migration the formula of poststack phase shift migration is generalized to common-offset section for constant-velocity medium. The formula of phase shift migration of common offset sections is obtained by numerical fitting method. This migration method takes vertically varying velocity into account by defining average velocity. Theoretically, only when velocity is constant, is the formula accurate. However, synthetic and real examples indicate that data with reasonable rapid vertical velocity variation can also be imaged effectively.The shot sources and sources records can be synthesized by three properties of wave equation operator without any physical assumption, moreover, in terms of three synthetic formulae we unify area shot records synthetic method and phase encoding synthetic method under one theoretical framework. The method is flexible and selectable by designing different synthetic operator for different input data sets. Prestack depth migration method of synthetic source records retains accuracy of shot profile migration and overcomes high computational cost. The realization of parallel programming with MPI speeds up the computations. The test of 3D SEG/EAEG salt C3-NA data set and Sinkiang land rugged topograph seismic data set show that the method is efficient and practical.The reference slowness of adaptive spatial-division split-step Fourier migration method is determined by an error parameter of the perturbation term in split-step Fourier operator and whole velocity field variation. One reference slowness and its corresponding local slowness make up a division, which comprises several spatial discontinuous subdivisions commonly. The number of reference slowness is effectively reduced. The choice of reference slowness, including the number of reference slowness and construction of spatial divisions, is adaptive and more reasonable at each depth step. A simple and economical smoothing filter in the wave number-frequency domain is designed to avoid artifacts of the extrapolation of wave field in the presence of sharp discontinuities in the velocity field. The performance of the method is demonstrated on 2D models.Estimating the macro velocity model is considered to be the key step of prestack depth imaging procedures. We proposed residual migration velocity analysis by synthetic source records based on residual migration velocity analysis in the plane wave domain. The analytical equation is used to perform residual migration after prestack depth migration of synthetic source records. An interactive velocity analysis procedure is designed to build avelocity-depth model. Controlled illumination source is used for reducing the error resulting from aberration of plane wave transmitting complex geology structure. The correctness and effectiveness of the method is proved by the numerical experiments on the theoretical model and field datasets.更多还原