地震散射波场的高阶统计分析

以随机信号的二阶矩理论为基础的常规处理方法，不能适应地震信号是非平稳、时变和非最小相位等条件下的处理要求。本文研究了利用地震信号的高阶统计量来检测和识别不均匀储集体的方法。该方法的关键是依据高阶统计量原理，分析地震信号的高阶统计量特征。通过理论分析和对实际地震信号的检测表明，当具有广义高斯分布的地震信号中混有非高斯的，或非线性的低频信号时，能够利用地震序列广义高斯分布的对称性及时频双谱分析方法，检测出低频信号。该方法为识别碳酸盐岩不均匀储集体开辟了一条新的途径。

High-order statistic analysis of seismic dispersed wavefield.

The ordinary processing method based on second order cumulant theory of random signal can't meet the demands of processing when the seismic signal is non stationary,time variation and non minimum phase.The paper studied the method using high order statistic to detect and identify non uniform reservoir.The key of the method is to analyze the feature of high order statistic of seismic signal based on high order statistic.It is shown by theoretical analysis and detection of real seismic signal that the non symmetry of generalized Gauss distribution of seismic sequence and the time frequency double spectrum analysis method can be used for detection of low frequency signal when the seismic signal with generalized Gauss distribution is mixed with non Gaussian or non linear low frequency signal.The method opened a new way for identifying the non uniform carbonate reservoir.n 3 Dseismicinterpre