抛物Radon变换重建地震数据方法研究

发布时间：2018-03-18 19:19

本文选题：地震数据重建　切入点：抛物Radon变换　出处：《东北石油大学》2015年硕士论文　论文类型：学位论文

【摘要】：地震勘探的目的是为了得到地下介质的精确成像。然而由于环境和人为因素的影响,地震数据在空间方向上不可避免的出现采样间距过大、数据缺失或不规则采样等现象,这会降低叠前偏移成像质量、减弱多次波压制效果、影响后续速度分析精度。为保证地震资料后续处理技术的准确使用,必须对缺失的地震数据进行重建。抛物Radon变换是一种稳健有效的地震数据重建方法,通过多次迭代经过部分动校正后的CMP道集,就可实现不规则地震数据的有效重建。该方法对输入数据要求较少,方法简单,易于实现。但面对叠前资料处理数据量巨大的问题,计算效率的高低成为好的算法是否实用的关键。传统抛物Radon变换方法在地震数据重建过程中,需要多次迭代Radon正反变换,计算量过大,直接矩阵求逆算法在求解抛物Radon正变换过程中计算效率较低,而且使用传统的曲率参数采样准则可能会产生振幅异常和假频问题。本文在传统Radon变换的基础上,完成以下方面内容:一是改进最小二乘抛物Radon变换地震数据重建方法,通过使用新的曲率参数采样准则抑制了振幅异常和假频的产生,应用Levinson递推法代替直接矩阵求逆的方法,提高了计算效率。二是提出应用高分辨率抛物Radon变换方法进行地震数据规则化重建。该方法在最小二乘抛物Radon变换的基础上,引入稀疏约束矩阵,提高地震数据在Radon域的分辨率,从而减少抛物Radon变换重建地震数据的迭代次数,减小计算量。三是?-f域加权抛物Radon变换地震数据重建方法,通过引入新变量?,消除了Radon变换算子对频率的依赖,使得Radon变换算子及算子的逆只需计算一次,显著提高计算效率。在迭代过程中,引入变化的权系数,更好地实现了?-f域的能量聚焦。
[Abstract]:The purpose of seismic exploration is to obtain accurate imaging of underground media. However, due to the influence of environment and human factors, the spatial space of seismic data is inevitably too large, the data is missing or irregular sampling, and so on. This will reduce the imaging quality of prestack migration, weaken the suppression effect of multiple waves, and affect the accuracy of subsequent velocity analysis. Parabolic Radon transform is a robust and effective method for seismic data reconstruction, and the CMP gather after partial NMO correction is iterated several times. This method requires less input data, and is simple and easy to implement. However, in the face of the problem of large amount of data in pre-stack data processing, this method can be used to reconstruct irregular seismic data effectively. The efficiency of calculation becomes the key to the practicability of good algorithms. In the process of seismic data reconstruction, the traditional parabolic Radon transformation method needs many iterations of Radon positive and negative transformations, and the computation is too large. The direct matrix inverse algorithm is inefficient in solving the parabolic Radon positive transformation, and the amplitude anomaly and false frequency may be caused by using the traditional curvature parameter sampling criterion. In this paper, on the basis of the traditional Radon transform, the amplitude anomaly and the false frequency problem may be caused by using the traditional curvature parameter sampling criterion. The main contents are as follows: firstly, the reconstruction method of seismic data based on least square parabolic Radon transform is improved. The amplitude anomaly and false frequency are restrained by using the new curvature parameter sampling criterion, and the inverse method of direct matrix is replaced by Levinson recursive method. The computation efficiency is improved. Secondly, the high resolution parabolic Radon transform method is proposed to regularize the reconstruction of seismic data. Based on the least square parabolic Radon transform, the sparse constraint matrix is introduced. Improve the resolution of seismic data in Radon domain, thus reduce the number of iterations of parabolic Radon transform to reconstruct seismic data, reduce the amount of calculation. The method of seismic data reconstruction based on weighted parabolic Radon transform in -f domain is introduced by introducing new variables. By eliminating the dependence of Radon transform operator on frequency, the Radon transform operator and its inverse only need to be calculated once, and the computational efficiency is improved significantly. In the iterative process, the variable weight coefficient is introduced to realize better? The energy focusing in the -f domain.
【学位授予单位】：东北石油大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P631.4

【参考文献】