混合相位动态子波估计

常用的地震子波估计方法基于传统的褶积模型,没有考虑子波在传播过程中的动态衰减特性。我们采用改进的传统褶积模型,建立了能够描述子波衰减的动态褶积模型,以此为基础提出了动态子波估计方法:对衰减地震道做时频谱分析;假设反射系数的频谱为白谱,对衰减地震道频谱值做平滑,去除反射系数谱值的影响,获取动态子波振幅谱值;采用基于高阶统计量的双谱重构子波混合相位谱,实现混合相位动态子波估计。理论模型分析与实例计算结果表明,上述方法有效可行。

Estimation of mixed-phase nonstationary wavelet

The methods of seismic wavelet estimation at present are based on the traditional deconvolution model. It does not take the nonstationary attenuation character of wavelet into consideration. For solving the problem, we established the nonstationary convolution model which can describe the nonstationary wavelet by using the improved the traditional convolution model, and developed the nonstationary wavelet estimate method based on it. First, compute the spectrum of nonstationary attenuated seismic data. Second, suppose the spectrum of reflection coefficient is white, smoothing the spectrum of seismic data to delete the effect of the reflection coefficient value and get the spectrum of nonstationary wavelet. Third, use bispectrum based on higher order statistic to reconstruct the mixed-phase spectrum of wavelet, so the nonstationary wavelet which has the character of the mixed-phase can be acquired in the end. The theoretical model analysis and real data computation have proved that the method is right.