论文部分内容阅读
三维各向异性走时层析反演问题一直是地震数据处理中的难点,其核心是各向异性参数敏感核函数的建立和三维大规模数据的处理。基于球谐展开群速度表达式,计算走时关于各向异性参数的敏感核函数,引入整形正则化来处理大规模、不适定反问题,并利用共轭梯度法对垂直慢度、水平慢度和动校正慢度的参数化方法(第一种参数化方法)与Thomsen参数化方法(第二种参数化方法)进行反演,实现三维VTI介质初至波走时层析反演。理论模型和实际资料处理结果表明,与经典的Thomsen参数化方法相比,垂直慢度、水平慢度和动校正慢度的参数化方法,更有利于三维VTI介质走时层析速度反演和各向异性参数的建模,进一步提高了反演的精度,具有较好的应用前景。
The three-dimensional anisotropic traveltime tomography inversion has always been a difficult point in seismic data processing. The core of this method is the establishment of anisotropic parameter-sensitive kernel function and the processing of three-dimensional large-scale data. Based on the spherical harmonic expansion group velocity expression, the sensitive kernel function of anisotropy parameters is calculated and the regularization regularization is introduced to deal with large-scale and ill-posed inverse problems. The conjugate gradient method is used to measure the vertical slowness, horizontal slowness, The parametric method of moving calibration slowness (the first parameterization method) and the Thomsen parameterization method (the second parameterization method) are used to perform inversion to achieve the first-arrival time-of-flight tomographic inversion of 3D VTI media. The theoretical model and actual data processing results show that compared with the classical Thomsen parameterization method, the parametric method of vertical slowness, horizontal slowness and dynamic correction slowdown is more conducive to the 3D VTI media velocity inversion and The modeling of anisotropy parameters further improves the accuracy of inversion and has a good prospect of application.