三维不规则地震数据重建方法

The reconstruction method for irregular seismic data

不规则采样地震数据会对多道处理技术产生不良影响,降低地震资料的处理质量。本文针对有限带宽三维不规则地震数据,将二维空间非均匀Fourier变换理论和贝叶斯参数反演方法相结合,进行反演重建。首先,采用分频重建策略,对每一个时间频率依据最小视速度确定出待重建数据的空间频率带宽,从不规则地震数据中估计出重建数据的空间Fourier谱。然后,将不规则地震数据重建视为谱重建的地球物理反演问题,运用贝叶斯参数反演理论估计出空间Fourier谱。在反演求解时,采用Delaunay三角网格剖分方法来确定不规则采样点的权值。此外,为避免复数矩阵求逆,使用预条件共轭梯度算法来保证求解的稳定性和收敛速度。理论模型和实际资料处理结果验证了本文方法的有效性和实用性。

Irregular sampling seismic data could cause problems for multi-channel seismic data processing and also could decrease data processing quality.For 3D irregular seismic data with the limited bandwidth inversion reconstruction was conducted by integration of 2D space non-uniform Fourier transform theory and Bayesian parameter inversion method.At first by using frequency division method and based on minimum apparent velocity for every time frequency the spatial frequency bandwidth was determined for the data which will be reconstructed and the spatial Fourier spectrum of the reconstructed data was estimated from the irregular seismic data,then by regarding the irregular seismic reconstruction as geophysical reverse problem,Bayesian parameter inversion theory was utilized to estimate the spatial Fourier spectrum.In the inversion solution,Delaunay triangle grid subdivision scheme was used to determine the weight value of the irregular sample points.In addition in order to avoid complex matrix inversion pre-condition conjugate gradient algorithm was utilized to insure the stability and convergence speed for the solution.The theory model and processing results for the field data proved the effectiveness and practicality of the method.