利用基于模型的层析速度反演进行低幅度构造成像

Low-relief structural imaging with model-based tomographic velocity inversion

基于层位的层析反演在低信噪比地区无法进行层位解释,在划分速度界面时也无法保证模型精度,局限性较大;基于网格的层析反演不容易收敛到实际速度模型,且模型没有地质意义的约束,有时会产生不符合地质规律的反演结果。为此,基于模型的层析反演方法,借鉴了层位约束和全局反演优点,以地质层位为基础建立模型,在横向上沿地质层位网格进行参数更新,在纵向上层位间网格更新尺度随着模型的地质层位而变化,既解决了薄层速度建模问题,又能得到精准的高、低频速度分量和各向异性参数。其次,基于模型的层析反演将模型中地质层位的深度作为一个变量纳入层析反演数据空间,可以同时反演速度结构与反射点位置(或反射层深度),具有更高的反演精度。将基于模型的层析反演方法应用于YM工区实际资料,并通过各向异性叠前深度偏移处理,提高了速度模型精度,解决了该区低幅度构造成像问题。

Horizon-based tomographic inversion performs velocity inversion by calculating ray path and residual delay time in multiple velocity interfaces of simple layered model,and it needs structural model for interpretation to be input.It is quite difficult to carry out horizon interpretation on low SNR seismic data based on this inversion,and the model accuracy cannot be guaranteed when classifying velocity interfaces.Grid-based tomographic inversion carries out velocity inversion by obtaining globally ray path and residual delay time,which can describe details of interval velocity in layered model.However sometimes velocity model cannot be appropriately converged,and the model does not have the constraints of geological significance,and the inversion results do not conform to geological rule.Therefore,model-based tomographic inversion method takes the advantage of horizon constraint and global inversion to build a model based on geological horizons,which updates horizontally parameters along geologic horizon grid,and updates interval grid scale with the geologic horizon of the model in the vertical direction.It solves the problem of velocity model building for thin layer and obtains precise high-frequency and low-frequency velocity components,and anisotropic parameters.Furthermore,model-based tomographic inversions takes the depth of geological horizon in the model as a variable into the tomographic inversion data space,and the velocity structure and reflection point position (or the reflector depth) can be simultaneously obtained by inversion.Thereby model-based tomographic inversion has higher accuracy of inversion.The proposed model-based tomographic inversion method is applied to the real data in the area YM,and the accuracy of velocity model is improved by anisotropic prestack depth migration,and low-relief structures in the area are correctly imaged. © 2017, Editorial Department OIL GEOPHYSICAL PROSPECTING. All right reserved.