奇异值谱分析在绕射波分离及成像中的应用

Diffraction wavefield separation and imaging using singular spectrum analysis

地层中的断层、裂隙(缝)和孔洞等小尺度地质目标体是常见的油气运移通道和储集体,对油气勘探具有重要意义。这些地质目标体的地震响应为弱能量的绕射波,在偏移成像前进行反射波和绕射波的分离,然后实现绕射波成像,有利于识别这些目标体。常规绕射波分离方法主要基于反射波与绕射波的运动学特征差异。频率-空间域的奇异值谱分析方法同时利用反射波与绕射波的运动学和动力学特征差异,压制共偏移距道集中具有强能量和线性特征的反射波场,突出并分离绕射波场,进而实现绕射体的目标成像。模型数据和实际地震资料的测试结果表明,奇异值谱分析绕射波分离方法可有效改善断层及其它小尺度地质体的成像质量。

Small-scale geological targets,such as faults,fractures and karst caves,which are usually oil-gas migration channels and reservoirs and are critical for hydrocarbon prospecting.However,the seismic responses of these geological targets are low-energy diffractions.Therefore,it is better to separate diffractions and reflections before migration imaging,which is helpful to realize diffraction imaging and identify the targets.Traditional separation methods are usually based on the kinematic difference between diffractions and reflections.The singular spectrum analysis （SSA） method in frequency-space domain is based on the difference in both kinematic and dynamic properties,suppresses the reflection wavefield with high-energy and linear features in common offset gathers,highlights and separates diffractions from the full wavefield,to realize the imaging of diffraction targets.Examples on synthetic data and real data demonstrate that the migration and imaging of these small-scale geological targets are improved by the diffraction wavefield separation using singular spectrum analysis method.