A Pre-Stack Inversion Method Based on a Self-Facies-Control Low-Frequency Model and its Application in Offshore Reservoirs

Seismic inversion is one of the main methods for reservoir prediction by integrating logging and seismic information, but the rationality of the prediction results is heavily dependent on the initial model. For offshore oil fields, limited by the sparse and non-uniform distributed wells, the initial model obtained by mathematical interpolation is often deviated from the actual reservoir distribution, especially for the complex sedimentary environment case. Sedimentary facies control can solve this problem to a certain extent, but it is easily affected by the subjective understanding. In this paper, a seismic inversion technology based on self-facies-control low-frequency model is studied, which can make full use of the horizontal identification and spatial structure to constrain the ability of seismic data. Based on the self-facies-control initial model, the pre-stack geostatistical inversion is suitable for non-parallel structure reservoir, and the inversion results highlight the reservoir spatial structure. The method effectively improves the spatial structure resolution of the inversion result, and achieved the ideal application effect in the actual oilfield. The predicted coincidence of sand bodies with thickness greater than 12 m is more than 85%, and the average drilling rate of sandstone in the horizontal section of the horizontal well is 92.77%.