A pre-stack hybrid inversion method for shale vti reservoirs based on the exact reflectance formula

Abstract

The invention discloses a pre-stack hybrid inversion method for shale VTI reservoirs based on the exact reflectance formula. The hybrid inversion strategy combines linear and nonlinear inversion, uses the linear inversion results as constraints, and generates smaller Search range and initial particles, perform nonlinear inversion to obtain five-parameter inversion results simultaneously. This method overcomes the disadvantage that conventional linear inversion tends to fall into local extremum, and improves the calculation efficiency and accuracy of nonlinear inversion. The exact expression of reflection coefficient can overcome the limitation of small angle, weak reflection and weak anisotropy of approximate expression, and improve the applicability of inversion. The effectiveness of the method was verified by the test of the logging model data, which laid the foundation for the follow-up research on the anisotropy of VTI media.

Description

technical field [0001] The invention relates to the technical field of seismic exploration of unconventional anisotropic reservoirs, in particular to a pre-stacking mixed inversion method for shale VTI reservoirs based on an accurate reflection coefficient. Background technique [0002] Prestack seismic inversion obtains elastic information of subsurface media by mining AVO / AVA (amplitude versus offset / incidence angle) responses in seismic data. At present, pre-stack inversion is an important technology for seismic exploration. In the past, for conventional sand and mudstone reservoirs, the inversion method based on the isotropic assumption has been used, and has achieved great success in the past few decades. However, with the increasing demand for oil and gas, the focus of exploration has gradually shifted from conventional reservoirs to unconventional oil and gas. Among them, shale oil and gas has become the focus and difficulty of current exploration. Shale rocks exhi...

AI Technical Summary

Problems solved by technology

However, since the inversion of anisotropic media needs to extract five target parameters at the same time, and the seismic sensitivity of anisotropic parameters is low, this leads to stronger ill-posedness of VTI inversion, and it is difficult to obtain better five-parameter results at the same time. In particular, the estimation of the anisotropy parameter has not been ideal

Therefore, Zhang et al. (2019) proposed a step-by-step inversion strategy based on approximate formulas, converting the five parameters into the other three target parameters through mathematical conversion, and then converting the anisotropic parameters through the indirect inversion method. However, this method introduces additional assumptions, and still use a less accurate approximation

Although the precise expression of the VTI medium reflection coefficient has been given, although its complex mathematical expression makes it difficult to obtain partial derivatives, it is currently rarely used in pre-stack inversion

[0004] Conventional linear inversion tends to fall into local extremums, while nonlinear inversion has problems such as large search range and low calculation efficiency, and approximate inversion has problems such as low accuracy at far angles and weak anisotropy limitations

Images