基于自适应子波分解的品质因子Q提取方法

Extraction method for quality factor Q based on adaptive wavelet decomposition

品质因子Q和速度一样是地层的固有参数,其大小决定地震波吸收衰减的强弱。由于Q值的大小与地震子波的传播时间和频率的乘积有关,因此,提出了一种基于自适应子波分解的品质因子提取方法。首先,利用自适应子波分解算法对地震信号进行子波分解,得到的匹配子波可认为是不同时刻的时变子波;然后,将不同匹配子波的中心时间和中心频率的乘积与相应匹配子波的对数振幅投影到平面坐标中,得到随该乘积值变化的对数振幅散点分布;最后通过拟合这些离散点斜率提取Q值。为压制离群值对拟合精度的影响,采用L1模线性回归方法对离散点进行线性拟合。理论模型测试和实际资料应用的结果表明,该方法能比较准确地提取品质因子,有效补偿地震波能量,提高了地震资料的分辨率。

Quality factor Q is the inherent parameter of strata, which determines the strength of seismic attenuation. Since the Q value is related with the product of travel time and frequency of seismic wavelet, we proposed a method to extract the value of Q based on the adaptive wavelet decomposition. Firstly,we decompose seismic signal using the adaptive wavelet decompo- sition algorithrrL The decomposed pursuit wavelets can be considered as time-varying seismic wavelets at different time. Then,the products of central time and central frequency versus the corresponding logarithmic amplitudes of the pursuit wavelets are projected into a coordinate plane, to obtain the scatter distribution of logarithmic amplitudes with these prod- ucts. Finally, the slope of these scatters is fitted to extract Q value. In order to decrease the influence of outliers,L1 norm lin- ear regression method is used in the linear fitting. Theoretical and real data tests show this method can accurately extract quality factor, compensate seismic wave energy and improve the resolution of seismic data.