基于行波分解的绕射波成像方法研究

地震勘探中的绕射体如断层、裂缝、地层尖灭、孔洞、盐体等地质异常体是常见的油气运移通道和(或)储集体,对它们的刻画和描述在油气资源勘探中具有重要意义。在常规地震数据处理中,绕射体往往被能量更强的反射体所掩盖,为更好地识别这些小尺度地质异常体,可对绕射体进行单独成像。为此,发展了基于行波分解的绕射波逆时偏移成像方法,得到了高分辨率的绕射体成像结果。其关键在于利用绕射波和反射波传播方向的差异,单独提取绕射体成像结果。主要步骤如下:构建解析波场,对源检端地震波场分别进行分解得到下左和下右行波;修改逆时偏移成像条件,利用分解得到的下左行波和下右行波对所有入射角度的绕射体和满足Snell定律的正、负倾角的反射层分别进行成像;得到“正倾角反射层+绕射体”和“负倾角反射层+绕射体”两种成像结果,并将两种成像结果进行相关,从而达到了压制反射波,提取绕射波的目的。模拟数据和实际数据应用结果表明:该方法得到的绕射波成像结果可以有效压制反射层能量,精准定位地下介质中存在的绕射体。

Diffractors such as faults,fractures,pinch-outs,and karsts are typical migration channels or reservoirs for oil and gas.The quantitative characterization of subsurface diffractors is crucial for oil exploration.In traditional processing,diffractors are often obscured by reflectors with large impedance contrast or contaminated by imaging noises.To better locate and quantitatively characterize these small-scale anomalies,it is necessary to develop a target-oriented method to image the diffraction separately.We herein proposed a diffraction-wave reverse time migration (RTM) method based on directional wavefield decomposition.The key for separation of reflection and diffraction lies in the behavior of wavefield direction,as the reflections obey Snells law while scattering-waves have multiple emergence directions.In this method,the directional wavefield decomposition method was utilized to decompose the seismic wavefields into the downward-left and downward-right propagation waves.Moreover,the conventional RTM imaging condition was modified to produce two directional imaging results,i.e.the imaging for positive dip reflections with diffractions,and the imaging for negative dip reflections with diffractions.By correlating these two results,the reflection images were cancelled out,whereas the diffraction images were enhanced.Tests on both synthetic and real data demonstrated that the proposed method can efficiently suppress the imaging of reflectors,and accurately locate the subsurface diffractors.

国家重点研发计划变革性技术关键科学问题重点专项(2018YFA0702503)、国家自然科学基金(41774126)和国家科技重大专项(2016ZX05024-001,2016ZX05006-002)共同资助。