A pre-stack geostatistical inversion method under 3D dual control

Abstract

The invention relates to a prestack geostatistical inversion method under three-dimensional double control. The method comprises the steps of (1) target work area petrophysical analysis and well seismic calibration, (2) the construction of a low-frequency model under seismic velocity constraints, (3) prestack simultaneous inversion based on partial superimposed seismic data, (4) lithologic fluid probability analysis, and (5) geostatistical inversion under three-dimensional double control constraints. According to the method, in the calculation of a parameter of geostatistical inversion prior information lithology probability, firstly, the deterministic inversion is carried out, based on the Bayesian discriminant principle, elastic data obtained by inversion is converted into a three-dimensional lithological probability body, the three-dimensional lithological probability body is taken as prior information to replace traditional zero-dimensional lithologic probability constrained geostatistical inversion, the frequency merging and frequency-down processing are carried out on a final inversion result, therefore, while the longitudinal resolution of the geostatistical inversion is improved, original seismic structure characteristics still can be maintained, the inversion uncertainty is reduced, and the invention still has good predictive ability.

Description

technical field [0001] The invention relates to a seismic data inversion method in the process of predicting oil and gas reservoirs, in particular to a pre-stack geostatistical inversion technology suitable for compressional wave impedance superposition and strong heterogeneity reservoirs. Background technique [0002] Compared with post-stack inversion, only P-wave impedance information can be obtained, which is difficult to apply to the prediction of P-wave impedance superimposed reservoirs. Pre-stack inversion technology is based on the theory related to seismic reflection wave amplitude and reflection coefficient at different incident angles. Stack seismic data to invert various elastic parameters, such as compressional and shear wave impedance, density, Poisson's ratio, etc., and then predict reservoir lithology, physical properties and fluids. At present, pre-stack geostatistical inversion technology has developed into an effective method for pre-stack high-resolution ...

AI Technical Summary

Problems solved by technology

[0003] Although geostatistical inversion can obtain inversion results higher than seismic resolution, key parameters such as vertical and horizontal ranges and probability density functions in the inversion process often rely on experience and are selected by parameter testing. Low, lack of geological meaning, the inversion results are uncertain, low predictability, it is difficult to accurately predict the distribution of reservoirs with strong lateral heterogeneity, so it is necessary to explore methods that can further improve the predictability of geostatistical inversion

Images