关键词: 地震地貌学, 切片, 地震属性, 相干体, 三维可视化

Abstract: The development of fault and fracture is an important issue in 3D seismic interpretation during hydrocarbon exploration and development. The random noises contained in seismic data often affect the depiction accuracy of faults, so denoising technique is often used in the interpretation process to reduce noise and improve the accuracy of fault description. The conventional denoising technique can not only improve the S/N of seismic data, but also destroy the edge information of the reflection structure and the sharpness of fault, which affects the accuracy of structural interpretation and reservoir description. In order to solve the problems of structure protection denoising and fault detection, a method of fault recognition based on auto-adaptive dip-steering was proposed. A steering cube was generated by auto-adaptive dip-steering technology which contains dip and azimuth information of sampling point, and then the steering cube was used as a filter input for constraint denoising to improve S/N and retain important structural information, and finally, the similarity attributes were extracted from filtered seismic volume under the guidance of dip-steering so as to recognize fault and improve the accuracy of fault interpretation. The result shows that the similarity attributes extracted by auto-adaptive dip-steering technique effectively highlight the discontinuity of fault, and it has a good noise reduction result. Auto-adaptive dip-steering technique effectively enhances the fault recognition ability, so it is an effective method to describe the faults and the geological boundary, and has a good application prospect in complex reservoir exploration.

Key words: seismic geomorphology, slicing, seismic attribute, coherence, 3D visualization